Adjunctive Therapies For AIDS Dementia Complex Cochrane

Adjunctive therapies for AIDS dementia complex (Review)
Uthman OA, Abdulmalik JO
This is a reprint of a Cochrane review, prepared and maintained by The Cochrane Collaboration and published in The Cochrane Library
2008, Issue 3
http://www.thecochranelibrary.com
Adjunctive therapies for AIDS dementia complex (Review)

Copyright © 2008 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
TABLE OF CONTENTS
HEADER . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
ABSTRACT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
PLAIN LANGUAGE SUMMARY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
BACKGROUND . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
Figure 1. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
Figure 2. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
Figure 3. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
OBJECTIVES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
METHODS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
Figure 4. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
RESULTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
Figure 5. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
DISCUSSION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
AUTHORS’ CONCLUSIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
ACKNOWLEDGEMENTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
REFERENCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
CHARACTERISTICS OF STUDIES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
DATA AND ANALYSES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
Analysis 1.1. Comparison 1 Lexipafant vs Placebo, Outcome 1 Neuropsychological test scores. . . . . . . . . 60
Analysis 1.2. Comparison 1 Lexipafant vs Placebo, Outcome 2 Tolerability. . . . . . . . . . . . . . . . 61
Analysis 1.3. Comparison 1 Lexipafant vs Placebo, Outcome 3 All cause mortality. . . . . . . . . . . . . 61
Analysis 1.4. Comparison 1 Lexipafant vs Placebo, Outcome 4 Adverse effects. . . . . . . . . . . . . . 62
Analysis 2.1. Comparison 2 Peptide T vs placebo, Outcome 1 Neurocognitive performance test. . . . . . . . 63
Analysis 2.2. Comparison 2 Peptide T vs placebo, Outcome 2 Tolerability. . . . . . . . . . . . . . . . 63
Analysis 2.3. Comparison 2 Peptide T vs placebo, Outcome 3 All cause mortality. . . . . . . . . . . . . 64
Analysis 2.4. Comparison 2 Peptide T vs placebo, Outcome 4 Adverse effects. . . . . . . . . . . . . . . 64
Analysis 3.1. Comparison 3 OPC-1147 vs placebo, Outcome 1 Neuropsychological test scores. . . . . . . . . 65
Analysis 3.2. Comparison 3 OPC-1147 vs placebo, Outcome 2 Tolerability. . . . . . . . . . . . . . . 66
Analysis 3.3. Comparison 3 OPC-1147 vs placebo, Outcome 3 All cause mortality. . . . . . . . . . . . . 66
Analysis 3.4. Comparison 3 OPC-1147 vs placebo, Outcome 4 Adverse effects. . . . . . . . . . . . . . 67
Analysis 4.1. Comparison 4 Valproic acid vs Placebo, Outcome 1 Neuropsychological test scores. . . . . . . . 68
Analysis 4.2. Comparison 4 Valproic acid vs Placebo, Outcome 2 Tolerability. . . . . . . . . . . . . . . 69
Analysis 4.3. Comparison 4 Valproic acid vs Placebo, Outcome 3 All cause mortality. . . . . . . . . . . . 69
Analysis 4.4. Comparison 4 Valproic acid vs Placebo, Outcome 4 Adverse effects. . . . . . . . . . . . . . 70
Analysis 5.1. Comparison 5 Memantine versus placebo, Outcome 1 Neuropsychological test scores. . . . . . . 70
Analysis 5.2. Comparison 5 Memantine versus placebo, Outcome 2 Tolerability. . . . . . . . . . . . . . 71
Analysis 5.3. Comparison 5 Memantine versus placebo, Outcome 3 All cause mortality. . . . . . . . . . . 71
Analysis 5.4. Comparison 5 Memantine versus placebo, Outcome 4 Adverse effects. . . . . . . . . . . . . 72
Analysis 6.1. Comparison 6 Thioctic acid vs Placebo, Outcome 1 Neuropsychological test scores. . . . . . . . 72
Analysis 6.2. Comparison 6 Thioctic acid vs Placebo, Outcome 2 Tolerability. . . . . . . . . . . . . . . 73
Analysis 6.3. Comparison 6 Thioctic acid vs Placebo, Outcome 3 All cause mortality. . . . . . . . . . . . 74
Analysis 6.4. Comparison 6 Thioctic acid vs Placebo, Outcome 4 Adverse effects. . . . . . . . . . . . . 74
Analysis 7.1. Comparison 7 Oral selegiline (deprenly) vs Placebo, Outcome 1 Neuropsychological test scores. . . . 75
Analysis 7.2. Comparison 7 Oral selegiline (deprenly) vs Placebo, Outcome 2 Tolerability. . . . . . . . . . 76
Analysis 7.3. Comparison 7 Oral selegiline (deprenly) vs Placebo, Outcome 3 All cause mortality. . . . . . . . 76
Analysis 7.4. Comparison 7 Oral selegiline (deprenly) vs Placebo, Outcome 4 Adverse effects. . . . . . . . . 77
Analysis 8.1. Comparison 8 Thiotic acid and oral selegiline (deprenyl) vs Placebo, Outcome 1 Neuropsychological test
scores. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78
Analysis 8.2. Comparison 8 Thiotic acid and oral selegiline (deprenyl) vs Placebo, Outcome 2 Tolerability. . . . . 79
Analysis 8.3. Comparison 8 Thiotic acid and oral selegiline (deprenyl) vs Placebo, Outcome 3 All cause mortality. . 79
Analysis 8.4. Comparison 8 Thiotic acid and oral selegiline (deprenyl) vs Placebo, Outcome 4 Adverse effects. . . . 80
Adjunctive therapies for AIDS dementia complex (Review) i
Analysis 9.1. Comparison 9 Low dose Transdermal selegiline (deprenyl) vs placebo, Outcome 1 Neuropsychological test
scores. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81
Analysis 9.2. Comparison 9 Low dose Transdermal selegiline (deprenyl) vs placebo, Outcome 2 Tolerability. . . . 82
Analysis 9.3. Comparison 9 Low dose Transdermal selegiline (deprenyl) vs placebo, Outcome 3 All cause mortality. . 82
Analysis 9.4. Comparison 9 Low dose Transdermal selegiline (deprenyl) vs placebo, Outcome 4 Adverse effects. . . 83
Analysis 10.1. Comparison 10 High dose Transdermal selegiline (deprenyl) vs placebo, Outcome 1 Neuropsychological test
scores. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84
Analysis 10.2. Comparison 10 High dose Transdermal selegiline (deprenyl) vs placebo, Outcome 2 Tolerability. . . 85
Analysis 10.3. Comparison 10 High dose Transdermal selegiline (deprenyl) vs placebo, Outcome 3 All cause mortality. 85
Analysis 10.4. Comparison 10 High dose Transdermal selegiline (deprenyl) vs placebo, Outcome 4 Adverse effects. . 86
Analysis 11.1. Comparison 11 Low dose (50mg) CPI-1189 vs Placebo, Outcome 1 Neuropsychological test scores. . 87
Analysis 11.2. Comparison 11 Low dose (50mg) CPI-1189 vs Placebo, Outcome 2 Tolerability. . . . . . . . . 88
Analysis 11.3. Comparison 11 Low dose (50mg) CPI-1189 vs Placebo, Outcome 3 All cause mortality. . . . . . 88
Analysis 11.4. Comparison 11 Low dose (50mg) CPI-1189 vs Placebo, Outcome 4 Adverse effects. . . . . . . 89
Analysis 12.1. Comparison 12 High dose (100mg) CPI-1189 vs Placebo, Outcome 1 Neuropsychological test scores. 90
Analysis 12.2. Comparison 12 High dose (100mg) CPI-1189 vs Placebo, Outcome 2 Tolerability. . . . . . . . 91
Analysis 12.3. Comparison 12 High dose (100mg) CPI-1189 vs Placebo, Outcome 3 All cause mortality. . . . . 91
Analysis 12.4. Comparison 12 High dose (100mg) CPI-1189 vs Placebo, Outcome 4 Adverse effects. . . . . . . 92
Analysis 13.2. Comparison 13 Low dose nimodipine vs Placebo, Outcome 2 Tolerability. . . . . . . . . . . 93
Analysis 13.3. Comparison 13 Low dose nimodipine vs Placebo, Outcome 3 All cause mortality. . . . . . . . 93
Analysis 13.4. Comparison 13 Low dose nimodipine vs Placebo, Outcome 4 Adverse effects. . . . . . . . . . 94
Analysis 14.2. Comparison 14 High dose nimodipine vs Placebo, Outcome 2 Tolerability. . . . . . . . . . . 94
Analysis 14.3. Comparison 14 High dose nimodipine vs Placebo, Outcome 3 All cause mortality. . . . . . . . 95
Analysis 14.9. Comparison 14 High dose nimodipine vs Placebo, Outcome 9 Adverse effects. . . . . . . . . 95
Analysis 15.1. Comparison 15 Any adjunctive therapy vs placebo, Outcome 1 Tolerability. . . . . . . . . . 96
Analysis 15.2. Comparison 15 Any adjunctive therapy vs placebo, Outcome 2 All cause mortality. . . . . . . . 97
Analysis 15.3. Comparison 15 Any adjunctive therapy vs placebo, Outcome 3 Number of patient experiencing any adverse
effects. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98
WHAT’S NEW . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98
HISTORY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98
CONTRIBUTIONS OF AUTHORS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99
DECLARATIONS OF INTEREST . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99
SOURCES OF SUPPORT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99
INDEX TERMS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99
Adjunctive therapies for AIDS dementia complex (Review) ii

[Intervention Review]
Adjunctive therapies for AIDS dementia complex
Olalekan A Uthman1 , Jubril O Abdulmalik2

1 Save the Youth Initiative, Ilorin, Nigeria. 2 Federal Neuropsychiatric Hospital, Maiduguri, Nigeria
Contact address: Olalekan A Uthman, Save the Youth Initiative, P. O. Box 5146, Ilorin, Kwara State, 240-001, Nigeria.
uthlekan@yahoo.com.
Editorial group: Cochrane HIV/AIDS Group.

Publication status and date: New, published in Issue 3, 2008.
Review content assessed as up-to-date: 11 February 2007.
Citation: Uthman OA, Abdulmalik JO. Adjunctive therapies for AIDS dementia complex. Cochrane Database of Systematic Reviews
2008, Issue 3. Art. No.: CD006496. DOI: 10.1002/14651858.CD006496.pub2.
ABSTRACT
Background
AIDS dementia complex is a common complication of human immunodeficiency virus type 1 (HIV-1) that continues to exist despite
the current use of potent antiretroviral therapy. It is a source of great morbidity and, when severe, is associated with limited survival.
Objectives
To determine efficacy and safety of adjunctive therapies for AIDS dementia complex
Search strategy
We searched the Cochrane HIV/AIDS group trials Specialized Register (December 2006), the Cochrane Central Register of Controlled
Trials (The Cochrane Library Issue 1, 2007), MEDLINE (January 1980 to February 2007), EMBASE (January 1980 to February
2007), AIDSearch (January 1980 to February 2007), PsycINFO (January 1980 to February 2007), PSYCHLIT (January 1980 to
February 2007), LILACS (January 1980 to February 2007), conference proceedings, trial registers, theses databases, and reference lists
of the articles. We also contacted manufacturers and researchers in the field.
Selection criteria
Randomized controlled trials, either published or published, that compared one type of adjunctive therapy to no therapy or placebo in
adults with AIDS dementia complex
Data collection and analysis
Two authors independently assessed trials quality, extracted data and entered data into RevMan 4.2 software. Where possible intention-
to-treat data were used and we contacted study authors for additional information. We collected neurocognitive performance, adverse
effects, tolerability and all-cause mortality information from the trials.
Main results
Ten trials involving 711 people were included. All the studies were phase 2 trials. Six studies used adequate methods for allocation of
sequence generation and unclear in the remaining four trials. Allocation concealment was adequate in five trials and unclear in the
remaining trials. The trials were heterogeneous in terms of types, dosages, routes and frequencies of administration of the adjunctive
therapies. There were no significant differences between the treated and placebo groups on neuropsychological test scores, number of
those that complete the assign dosage of experimental medication, adverse effects, and all-cause mortality.
Adjunctive therapies for AIDS dementia complex (Review) 1
Authors’ conclusions
This review confirms the absence of evidence that any of the adjunctive therapies improves cognitive performance or quality of life, or
both for patients with ADC, though they were well tolerated and safe.
PLAIN LANGUAGE SUMMARY

There is no evidence that adjunctive therapies for AIDS dementia are effective, though they are well-tolerated and safe.
The authors did a systematic review of ten studies, to see if adjunctive therapies (supplemental to the main treatment) for AIDS dementia
were effective and safe. They found no evidence that adjunctive therapies had any effect on the patients They did find adjunctive
therapies to be safe, and without any harmful effect on the patients.
BACKGROUND
ADC produces a highly variable clinical course ranging from sub-
AIDS Dementia Complex (ADC) is a common complication of tle cognitive (Stage 0) and motor impairments to profound de-
human immunodeficiency virus type 1 (HIV-1) that continues mentia (stage 4) (see Table 1) (McArthur 2005). Early symptoms
to exist despite the current use of potent antiretroviral therapy include word-finding difficulty, forgetfulness, psychomotor slow-
(Schifitto 2001). ADC (also known as AIDS-related dementia, ing, and diminished writing or visual/motor skills. Simple strate-
HIV encephalopathy, and HIV-associated dementia) is a condi- gies, such as written reminders, can be used to compensate for
tion in people with AIDS that results in the loss of mental aware- early deficits. Common psychiatric symptoms include depressed
ness, judgment, and inability to function in a social or occupa- mood and hypomania. Some patients experience a gradual mental
tional setting. It is a source of great morbidity and, when severe, is decline, whereas others deteriorate rapidly over a relatively short
associated with limited survival (Parnes 2006). ADC is common period of time (Table 2). Seizures, global cognitive deterioration,
in HIV-1 patients with high plasma HIV-1 RNA levels, low CD4 mutism, incontinence, and severe confusion are common clinical
cell count (Childs 1999), anemia, low body mass index, older age, features of late-stage ADC (AIDS institutes 2006).
injection drug use (Stern 2001) and female sex (Chiesi 1996).
The main direct targets of HIV infection are cells of the immune
There is variation in the reported incidence of ADC worldwide
system (Figure 1: Figure 2); the nervous system is often damaged
(see Figure 1). Gonzalez 2006 reported annual incidence of ADC
in the course of infection (Figure 2: Figure 3), not only by disease
among HIV-positive individuals was 7%, with an estimated 20%
processes that are secondary to immune dysfunction and its sys-
risk of ADC after developing full-blown AIDS in United State
temic manifestations but also by more fundamental effects of the
of America and 5% to 7% in Europe (Mollace 2001). Whether
retrovirus (Price 1996; Ghafouri 2006). Structural similarities be-
people living with AIDS in other regions suffer from similar rates
tween HIV and a protein (neuroleukin) that stimulates nerve cells
of ADC is unknown. Some have claimed that competing causes
may contribute to ADC, and thus may provide a way to reverse
of early death could translate into lower rates of ADC; however
that dementia. This similarity creates a competition between virus
studies of ADC in Africa have failed to resolve this question by
and protein for binding sites on neurons leading to inhibition of
reporting different rates from 3% to 68% (Birbeck 2005).
nerve function by the viruses, and in turn causing symptoms as-
sociated with dementia (Edward 1987; Gonzalez 2006).

Figure 1. prevalence.Prevalence of AIDS dementia complex according to country
Figure 2. HIV-1 Neuroinvasion.Ghafouri 2006

1) According to the “Trojan Horse hypothesis” entry of HIV-1 into the brain takes place by the migration of
infected monocytes which differentiate into perivascular macrophage.
2) the passage of infected CD4+ T cells can be another source of infection in the brain. Other probable causes
of CNS infection might be: 3) the direct entrance of the virus or 4) entrance of HIV-1 by transcytosis of brain
microvascular endothelial cells.

Figure 3. Neuropathogenesis.Ghafouri 2006
Two components of this mechanism are: A) the direct effect of the HIV-1 infection and B) the indirect effect -
secretion of cytokines and neurotoxins.
The infected macrophages and microglia participate actively in the neurodegeneration by:
1) shedding viral proteins and 2) releasing of cytokines and neurotoxins into the CNS.
3) Tat and TNF-alpha contribute to the disruption of the blood brain barrier, which in turn become more
permeable and the secreted pro-inflammatory cytokines activates 4) microglia and 5) astrocytes which in turn
secrete neurotoxin. 6) Multifactorial neuronal injury.
Table 1. Staging of AIDS dementia complex
Stage Grading Presentation
0 Normal
0.5 Subclinical or Equivocal - Minimal or equivocal symptoms

- Mild (soft) neurological signs
- No impairment of work or activities of daily living (ADL)
1 Mild - Unequivocal intellectual or motor impairment

- Able to do all but the most demanding work or

Table 1. Staging of AIDS dementia complex (Continued)
2 Moderate - Cannot work or perform demanding ADL

- Capable of self-care
- Ambulatory, but may need a single prop
- Major intellectual disability or
- Cannot walk unassisted
3 Severe - Major intellectual disability or

- Cannot walk unassisted
4 End-stage - Nearly vegetative

- Rudimentary cognition
- Para- or quadriplegic
Table 2. Clinical manifestations of ADC
Type of impairment Manifestations
Affective - Apathy (depression-like feature)

- Irritability
- Mania, new-onset psychosis
Behavioral - Psychomotor retardation (Slowed speech or response time)

- Personality change
- Social withdrawal
Cognitive - Lack of visuospatial memory (misplacing things)

- Lack of visuomotor coordination (eye movement abnormalities)
- Difficulty with complex sequencing
- Impaired concentration and attention
- Impaired verbal memory (word-finding ability)
- Mental slowing
Motor - Unsteady gait, loss of balance

- Dropping things
- Tremors, poor handwriting
- Decline in fine motor skills

Generally, three tests are used to diagnose ADC. Mental status
examination (neuropsychological testing) is the primary method
to diagnose and evaluate ADC and is designed to reveal problems
like memory loss, disorientation, concentration, abstract thinking
and mood swings. Standard scan like magnetic resonance imaging
(MRI) and computerized tomography (CT) scans are used to rule
out other causes of symptoms of ADC (Table 3). Examination of
cerebrospinal fluid can be use to assess the virus load and other
chemicals in the brain.
Just as there is no cure for AIDS, there is no cure for ADC. Treat-
ment of ADC is multidisciplinary and involves different therapeu-
tic modalities (antiretroviral therapy, treatment of symptoms and
adjunctive therapies). The efficacy of antiretroviral agents remains
uncertain because most do not achieve effective levels in the brain
and can cause neurologic complications as well (Mollace 2001;
Zink 2005). Hence, this concern provides an additional impetus
for the development of effective adjunctive therapy for ADC.
Table 3. Differential diagnosis of symptoms presenting as possible ADC
CNS disease Manifestations
Central nervous system disease Infectious

- CMV encephalitis
- Neurosyphilis
- Cryptococcal meningitis
- Tuberculous meningitis
- CNS toxoplasmosis
- Progressive multifocal leukoencephalopathy*
- HIV minor cognitive motor disorder
Tumors
- CNS lymphoma
- Metastatic disease
Vasculitis
Systemic/metabolic/endocrine disease - B12 deficiency

- Anemia
- Thyroid disease
- Addison’s disease
Primary psychiatric illness - Mood disorders (major depression, hypomania, dysthymia)

- Delirium
Substance withdrawal or intoxification - Alcohol

- Opioids
- Chronic cannabis

Table 3. Differential diagnosis of symptoms presenting as possible ADC (Continued)
Medications Medication reactions

- Psychotropics
- Antiretroviral medications
Attention has focused on adjunctive therapies targeted at reduc-

ing or counteracting the effects HIV-1 on the central nervous sys-
tem (Table 4). These have included the calcium blocker channel
antagonist nimodipine (McArthur 2005), peptide T (Heseltine
1998), and minocycline (Zink 2005).
Although there are general reviews on neurological complications

of HIV infection (McArthur 2005) and neuroprotective therapy
for ADC (Turchan 2003), to the best of our knowledge there is
no systematic review on adjunctive therapies for ADC. There is
a need for a systematic review of well designed treatment studies
to produce an up-to-date synthesis of research evidence on effec-
tiveness of adjunctive therapies for treatment of ADC, add to the
body of knowledge related to treatment of ADC, and help pol-
icy makers, medical practitioners, and HIV infected individuals
identify interventions for treating ADC which are supported by
reliable evidence of effectiveness.
Table 4. Adjunctive therapies for ADC
Agent Action
CPI-1189 TNF-alpha antagonist
Lexipafant Platelet activating factor antagonist
Memantine N-methyl-D-aspartate antagonist
Minocycline Anti-inflammatory and p38 MAP kinase inhibitor
Nimodipine Calcium Channel blocker
Nitroglycerin Vasodilator
OPC 14117 Antioxidant
Pentoxifylline Platelet activating factor antagonist
Peptide T Possibly Chemokine receptor blockade
Prednisone Macrophage suppression

Table 4. Adjunctive therapies for ADC (Continued)
Selegiline (deprenyl) Monoamine oxidase-B inhibitor
Thioctic acid Antioxidant
Valproic acid Unknown
Any intervention other than antiretroviral therapy used for the

treatment of ADC. Interventions may be combined with antiretro-
viral therapy as long as intervention and control group both re-
OBJECTIVES
ceived antiretrovirals and the only difference between groups is a
To determine the efficacy and safety of adjunctive therapies in the non-antiretroviral intervention. No limits were placed on the fre-
management AIDS dementia complex and effects on symptom quency of use, dose, and the duration of treatment.
severity and all-cause mortality.
Control
Placebo or no therapy.
METHODS
Criteria for considering studies for this review Types of outcome measures
The following outcomes were considered in this review:
Types of studies
Primary outcomes
Only randomized controlled trials, either published or unpub- 1. Change in cognitive function (as measured by neuropsycholog-
lished, that compared one type of ADC therapy to no therapy or ical tests)
placebo, in any language were eligible for inclusion in this review. 2. Change in mean Karnofsky performance score
3. Global impression of change
Secondary outcomes
Types of participants
1. All cause mortality
All HIV infected patients on antiretroviral therapy (either sex, aged 2. Activities of daily living
18 to 60 years) with dementia and randomly assigned to treatment 3. Center for Epidemiologic Studies - Depression Scale (CES-D)
were eligible for inclusion. Any clinically useful subgroup of ADC 4. Any adverse effects
was acceptable, for example minor cognitive-motor disorder, as
long as it is defined in the study’s inclusion criteria. There were no
exclusions based on country of study.
Search methods for identification of studies
Studies that include patients with history of a neurologic disease
unrelated to HIV infection, chronic seizures or head injuries, any See: HIV/AIDS Collaborative Review Group search strategy
psychiatric illness were excluded unless subgroup analysis is pre- We attempted to identify all relevant trials regardless of language
sented that allows exclusion of just these patients. or publication status (published, unpublished, in press, and in
progress).
Databases
Types of interventions We searched the following databases using the search terms and
Intervention strategy described in (Table 5):
Table 5. Search Strategy
Search Query
#1 HIV Infections[MeSH] OR HIV[MeSH] OR hiv[tw] OR hiv-1*[tw] OR hiv-2*[tw] OR hiv1[tw] OR hiv2[tw] OR

hiv infect*[tw] OR human immunodeficiency virus[tw] OR human immunedeficiency virus[tw] OR human immuno-
deficiency virus[tw] OR human immune-deficiency virus[tw] OR ((human immun*) AND (deficiency virus[tw])) OR

Table 5. Search Strategy (Continued)
acquired immunodeficiency syndrome[tw] OR acquired immunedeficiency syndrome[tw] OR acquired immuno-deficiency

syndrome[tw] OR acquired immune-deficiency syndrome[tw] OR ((acquired immun*) AND (deficiency syndrome[tw]))
OR “Sexually Transmitted Diseases, Viral”[MeSH:NoExp]
#2 “AIDS DEMENTIA COMPLEX”[MeSH ] OR AIDS DEMENTIA COMPLEX[tw]
#3 #1 OR #2
#4 randomized controlled trial [pt] OR controlled clinical trial [pt] OR randomized controlled trials [mh] OR random
allocation [mh] OR double-blind method [mh] OR single-blind method [mh] OR clinical trial [pt] OR clinical trials
[mh] OR (“clinical trial” [tw]) OR ((singl* [tw] OR doubl* [tw] OR trebl* [tw] OR tripl* [tw]) AND (mask* [tw] OR
blind* [tw])) OR ( placebos [mh] OR placebo* [tw] OR random* [tw] OR research design [mh:noexp] OR comparative
study [mh] OR evaluation studies [mh] OR follow-up studies [mh] OR prospective studies [mh] OR control* [tw] OR
prospectiv* [tw] OR volunteer* [tw]) NOT (animals [mh] NOT human [mh])
#5 ADJUNCTIVE THERAPY OR ADJUNCTIVE THERAPIES OR CPI-1189 OR LEXIPAFANT OR MEMANTINE

OR MINOCYCLINE OR NIMODIPINE OR NITROGLYCERIN OPC 14117 OR PENTOXIFYLLINE OR PEPTIDE
T OR PREDNISONE OR SELEGILINE OR THIOCTIC ACID OR VALPROIC ACID
#6 #3 AND #4 AND #5
#7 #6 Field: All Fields, Limits: from 1980 to 2006, HUMAN
• Cochrane HIV/AIDS Group Specialized Register • Conference on Retroviruses and Opportunistic Infections
(December 2006) (1996 to 2006)
• Cochrane Central Register of Controlled Trials • International AIDS Society (IAS) Pathogenesis and
(CENTRAL), published in The Cochrane Library (2007 Issue 1) Treatment (2001 to 2005)
• MEDLINE (1980 to February 2007) • International AIDS Conferences (1985 to 2004)
• EMBASE (1980 to February 2007) • International Congress on Drug Therapy in HIV Infection
(1992 to 2004)
• AIDSearch (1980 to February 2007)
• International HIV Drug Resistance Workshop (2002 to
• PsycINFO (1980 to February 2007)
2006)
• PSYCHLIT (1980 to February 2007)
• National HIV/AIDS Update Conference (2000 to 2005)
• LILACS (1980 to February 2007)
• National HIV Prevention Conference (1999 to 2005)
Conference proceedings • Eleventh International Conference on AIDS and STDs in

Africa (ICASA): Lusaka, Zambia (www.africa.upenn.edu/
We searched the following conference proceedings for relevant
Urgent˙Action/apic˙91599.html)
abstract:
• National Cancer Institute’s AIDS-related Malignancy
conferences; and
• Annual Conference of the British HIV Association (2001
to 2006) • International Congress on Drug Therapy in HIV Infection

Ongoing trials: • Dissertation Abstract Online (USA) (/wwwlib.umi.com/
dissertations/gateway)
• Australian Clinical Trials Registry (www.actr.org.au/)
Researchers, organization, and pharmaceutical companies
• ClinicalTrials.gov (http://clinicaltrials.gov/)
• CenterWatch Clinical Trials Listing Service For unpublished and ongoing trials, we contacted individual re-
(www.centerwatch.com/) searchers working in the field and the following organizations:
National Institutes of Health (NIH), World Health Organization
• Current Controlled Trials (www.controlled-trials.com/) (WHO) and the US Centers for Disease Control and Prevention
• PsiTri Database (mental health treatments) (http:// (CDC); and pharmaceutical companies such as Centaur pharma-
psitri.stakes.fi/EN/psitri.htm) ceuticals, Miles Pharmaceuticalsand Roche, Somerset Pharmaceu-
ticals, and Otsuka America Pharmaceuticals.
• GlaxoSmithKline Clinical Trial Register (http://
ctr.gsk.co.uk/) Reference lists
• NovartisClinicalTrials.com (www.novartisclinicaltrials.com) Reference lists of all studies and reviews identified by the above
methods were checked for further trials.
• Johns Hopkins HIV Neurology Group Clinical Trials and
Cohort Studies (/www.neuro.jhmi.edu/HIV/clinical˙trials.htm)
• National Institutes of Health (NIH) (www.nih.gov/) Data collection and analysis
Selection of Studies
Theses Both authors (AO and JO) independently applied the inclusion
criteria to all identified trials. We used the titles and abstracts of
the identified citations to exclude trials that clearly did not meet
• Networked Digital Library of Theses (www.ndltd.org/)
the inclusion criteria (see Figure 4). If either author judged that
• Australian Digital Theses Program (http://adt.caul.edu.au/) the trial might be eligible for inclusion, we obtained the full pa-
per. We independently screened the full articles of selected trials
• Canadian Theses and Dissertations
and resolved any disagreements by discussion. We gave reasons
(www.collectionscanada.ca/thesescanada/index-e.html)
for excluding potentially relevant trials in the “Characteristics of
• DATAD - Database of African Theses and Dissertations Excluded Studies.” We attempted to contact the authors for clar-
(www.aau.org/datad/backgrd.htm) ification when necessary.

Figure 4. Inclusion form.ELIGIBILITY FORM

Assessment of methodological quality and risk of bias
Two authors assessed the methodological quality of each trial in • Yes (the study appears to be free of other sources of bias)
terms of sequence generation, allocation concealment (Jüni 2001), • No (there is at least one important risk of bias)
blinding (of participants, personnels, and outcomes), complete- • Unclear (there may be a risk of bias, but there is either
ness of follow up, incomplete outcome data, selective outcome insufficient information or insufficient rationale or evidence that
reporting, and topic specific or design specific or potential threats an identified problem will introduce bias)
to validity.
Generation of allocation sequence Data extraction
Two authors independently extracted the data using a pre-designed
• Adequate (using a computer random number generator, data collection forms. For each of the studies, we extracted data the
coin tossing , shuffling cards or envelopes, throwing dice, following: citation, study design, methodological criteria, inclu-
drawing of lots, minimization) sion and exclusion criteria, comparison group intervention, par-
• Inadequate (non-random approach, allocation by alternate ticipant characteristics, trial setting, elements of intervention, all
allocation, and allocation by judgment of the clinician) relevant outcomes measures, and results. Where reports were un-
• Unclear (Insufficient information about the sequence certain or include summary measures, authors were contacted for
generation process) clarification. We also took note of any data that is consistently
Allocation concealment underreported, and highlighted this deficit along with future re-
search needs.
• Adequate (participants and the investigators enrolling We checked whether authors had conducted an intention-to-treat
participants cannot foresee assignment) analysis (all randomized participants should be analysed in the
• Inadequate (participants and investigators enrolling groups to which they were originally randomized). We calculated
participants can foresee upcoming assignment) the percentage lost to follow-up and reported this information. For
• Unclear (method not described) dichotomous outcome measures, we recorded the number of par-
Blinding ticipants experiencing the event and the number analysed in each
We recorded who was blinded in each trial, such as the participants, group. For continuous outcome measures, we extracted the mean
care provider, or outcome assessor. change from baseline, the standard deviation of the mean change,
Completeness of follow up and the number of patients for each treatment group at each assess-
A = >80%; B = <80%, C = unclear ment. Where standard errors were presented (eg Heseltine 1998),
(This is a change from the protocol; we proposed to discard studies standard transformation (Higgins 2006) was applied to estimate
if over 30% of participants were lost during follow up) the standard deviation by multiplying standard errors of means
Incomplete outcome data adequately addressed? by the square root of the sample size. Trialists were contacted to
supply missing information and to clarify issues. We resolved dis-
• Yes (no missing outcome data; reasons for missing outcome
crepancies through discussion.
data unlikely to be related to true outcome; missing outcome
Data analysis
data balanced in number across groups)
All eligible studies were summarized and analyzed in Review
• No (reason for missing outcome data likely to be related to
Manager 4.2. The two authors extracted the data, the first author
true outcome, with either imbalance in number across groups or
entered all data into RevMan and second author rechecked all en-
reasons for missing data having different profiles across groups)
tries. Disagreements were resolved by discussion.
• Unclear (insufficient reporting of attrition/exclusions)
The outcomes measured in clinical trials of dementia and cog-
Free of selective outcome reporting? nitive impairment often arise from ordinal rating scales, such as
neuropsychological test scores, clinical global impression scales,
• Yes (all pre-specified (primary and secondary) outcomes
and quality of life scales. Where the rating scales used in the trials
have been reported or it is clear that the published report
have a reasonably large number of categories (more than 10) the
includes all of the study’s pre-specified outcomes and all expected
data we treated as continuous variables arising from a normal dis-
outcomes)
tribution. We used weighted mean difference (WMD) when the
• No (not all pre-specified outcomes have been reported; one
pooled trials use the same rating scale or test, and the standardized
or more reported outcomes were not pre-specified; one or more
mean difference (SMD). Study results for dichotomous data were
outcomes are reported incompletely in a way related to the actual
expressed as relative risk (RR) and 95% confidence interval (CI).
results; or the study report fails to include results for an outcome
Stratification of results
that would be expected to have been reported for such a study)
Due to the variety of combinations of different treatments and dif-
• Unclear (insufficient information)
ferent comparators, it was not be possible to combine all included
Free of topic specific, design specific or other potential threats to trials in a single meta-analysis. We thus decided to use a narrative
validity? synthesis; to present details for each study and discuss them in turn.

For trials with more than two arms, we considered each pair wise Sensitivity test
comparison as if it were separate two-arm trials (Ramsay 2007). We planned sensitivity analyses to assess the robustness of the
For example, if a trial with two experimental groups, high dose meta-analysis by comparing the results using all trials and then
nimodipine and low dose nimodipine, and a placebo group; we in- excluding trials of a lower methodological quality.
cluded the high dose nimodipine versus placebo arm and low dose Sub-Group Analyses
nimodipine versus placebo as separate trials. We did not include We proposed to stratify our analysis by intervention setting
the two comparisons in one meta-analysis to avoid ’unit analysis (dosage, route of administration, antiretroviral therapy naive or
error’ (counting placebo patients twice). We proposed to prepare not) and CD4 count given sufficient data.
separate meta-analyses if there are more than one trial comparing Publication bias
similar interventions and comparison groups: (1) adjunctive ther- We planned to look for asymmetry in a funnel plot of the standard
apy versus another adjunctive therapy belonging to the same or error plotted against the risk ratio measured on a logarithmic scale
different drug class; and (2) adjunctive therapy grouped by drug as an indication of publication bias (Egger 1997).
class versus other adjunctive therapy belonging to a different drug
class (Table 4).
Not included in this review:
Due to insufficient data, we were unable to carry out the following RESULTS
analyses:
Heterogeneity
Description of studies
We planned to assess the presence of statistical heterogeneity in
the meta-analyses by looking at the forest plot and chi-squared test See: Characteristics of included studies; Characteristics of excluded
for heterogeneity using a P value of 0.10 to determine statistical studies; Characteristics of ongoing studies.
significance; and to quantify inconsistent across studies using a Trial selection
simplified categorization of I2 : low (I2 value of 25%), moderate (I We prepared a QUOROM (Moher 1999) statement flowchart to
2 value of 50%), and high (I2 value of 75%) (Higgins 2003). describe how we processed the references identified through the
search results (see Figure 5).

Figure 5. Trial flow.Quorom statement flow diagram of the process of identifying and including references
for the systematic review - Ajunctive therapies for AIDS dementia complex

The literature searches yielded 969 titles of potentially relevant The trials were homogenous in terms of the age groups stud-
articles. After scanning titles and abstracts, a total of 33 potentially ied [mean age of 38.50 years (Navia 1998) to 62.00 years (
relevant articles were identified and full-text copies were assessed Heseltine 1998)]. Most of the studies included more male par-
independently against the inclusion criteria by the two authors. ticipants [60.00% (Sacktor 2000) to 95.40% (Heseltine 1998)].
The reference lists of these studies were scanned, which resulted The proportion of white participants [27.70% (Schiffito 2006a) to
in another two papers for inclusion. Twenty papers were excluded 86.00%(Heseltine 1998)] and mean duration of known HIV in-
(see ’Characteristics of excluded studies’) and 12 papers (ten RCTs) fection [mean year = 4.60 (Dana 1998) to 10.40 (Schiffito 2006a)]
met the inclusion criteria (see ’Characteristics of included studies’). differed across studies.
All trials were identified from published literature. The following inclusion criteria were used in the studies reviewed:
Included studies • Patients who had HIV infection (confirmed by enzyme-

All trials had randomized double-blind controlled design (see linked immunosorbent assay (Dana 1997; Dana 1998; Heseltine
’Characteristics of Included Studies’ for detailed information 1998), Western blot (Clifford 2002; Dana 1997; Dana 1998;
about the individual trials). The trials were published between Heseltine 1998), or polymerase chain reaction(Heseltine 1998)),
1997 and 2007 in Neurology (Clifford 2002; Dana 1997; Dana • Evidence of cognitive impairment
1998; Navia 1998; Sacktor 2000; Schiffito 1999; Schiffito 2006a; • Stable antiretroviral regimen for 6 weeks before
Schiffito 2007b), Archives of neurology, Journal of neurovirology randomization
(Heseltine 1998), and AIDS (Schiffito 2007a).
Cognitive impairment was defined as performing at or below 1
Trial location
SD from the mean on at least two neuropsychological tests, or 2
All the studies reviewed were conducted in USA. Multicenter stud-
SDs below the mean on at least one test.
ies were organized by the Charles A. Dana Foundation Consor-
Most of the studies excluded patients with:
tium on the Therapy of HIV Dementia and Related Cognitive Dis-
orders (University of Rochester, Columbia University, Johns Hop- • Active opportunistic central nervous system infection
kins University); other sites included the University of Southern (within 12 weeks of initial treatment of toxoplasmosis or
California Los Angeles (USC), the University of Miami, Miami, cryptococcal meningitis),
Fla (UM), and the University of California, San diego (USCD). • Severe premorbid psychiatric illness likely to interfere with
protocol compliance,
Participants • History of a chronic neurologic disorder unrelated to HIV
This review was based on data from 711 participants. The break- infection,
down, with respect to individual studies, was as follows:
Heseltine 1998 excluded participants with previous use of Peptide
• Schiffito 1999: 30 participants T. In Schiffito 2006a no information was provided for exclusion
• Dana 1998: 36 participants of the participants.
• Schiffito 2006a: 16 participants Interventions
• Clifford 2002: 64 participants Most of the trials had one form experimental treatment of ad-
• Navia 1998: 38 participants junctive therapy (See Table 6) and control group (Dana 1997;
• Sacktor 2000: 14 participants Heseltine 1998; Sacktor 2000; Schiffito 1999; Schiffito 2006a).
• Heseltine 1998: 215 participants Three studies (Clifford 2002; Navia 1998;Schiffito 2007b) had
• Dana 1997: 30 participants three arms with two different doses and placebo. Dana 1998 had
• Schiffito 2007a: 140 participants four arms: thioctic acid alone, thioctic acid and deprenyl, deprenyl
• Schiffito 2007b: 128 participants alone, and control group.
Table 6. Drugs - Routes and regimen
Drug Route Dosage
Lexipafant Oral 250mg twice daily
Deprenyl (Selegiline) Oral 2.5mg three times a week
Thioctic acid Oral 600mg twice daily

Table 6. Drugs - Routes and regimen (Continued)
Valproic acid Oral 250mg twice daily
CPI-1189 Oral 50mg once daily
CPI-1189 Oral 100mg once daily
Nimodipine Oral 30mg three times daily
Nimodipine Oral 60mg five times daily
Deprenyl (Selegiline) Transdermal 1.0mg/cm X 15 cm2 delivering approximately 3.1 mg per 24 hours
Peptide T Intranasal Three times a day
OPC-14117 Oral 60mg twice daily for the first six weeks then maintenance dosage of 120mg twice daily
Memantine Oral 10mg daily
See Table 6 for details of drugs, route of administration, and

treatment regimen.
Outcome measures
Primary outcome of most of the trials was tolerability, defined by
whether the subject completed the study at the originally assigned
dosage of experimental medications, regardless of whether dosage
halving or suspension occurred during the study. A secondary mea-
sure of tolerability was whether the subject completed the study.
Measures of safety included frequencies of adverse experiences
and abnormal results on laboratory tests, and changes over time
in laboratory tests and vital signs. Measures of efficacy included
changes from baseline in neuropsychological test results (see Table
7), global impression, and measures of function and mood. In
Heseltine 1998 the primary efficacy end point was change in cog-
nitive function.
Table 7. Neuropsychological test scores
Domain Test
Rey Auditory Verbal Learning - Total Number Correct

- Number Correct - Trial 5
- Recall after Interference
- Delayed Recall
Mean Reaction Time - Choice

- Sequential

Table 7. Neuropsychological test scores (Continued)
Trail Making - Part A

- Part B
Digit Symbol
Grooved Pegboard - Dominant Hand

- Non-dominant Hand
Finger Tapping - Dominant Hand

- Non-dominant Hand
Timed Gait
Composite neurological Z score (NPZ)
All studies used validated neuropsychological test scores.

Sources of support
Of the eight trials, seven listed the study sponsors (Table 8) and
sources of support (Table 9). Heseltine 1998 provided no infor-
mation about the trial sponsor(s).
Table 8. Trials sponsors
Sponsor Study ID
The Charles A. Dana Foundation Schiffito 1999; Dana 1998; Dana 1997
National Institutes of Health (NIH) Schiffito 1999; Dana 1998; Schiffto 2006; Sacktor 2000; Dana 1997
Neurological AIDS Research Consortium Schiffito 1999; Dana 1998
National Institute of Mental Health Dana 1998
National Institute of Neurologic Diseases and Stroke grant Clifford 2002
General Clinical Research Centers (GCRC) at Columbia Clifford 2002
GCTC at Rochester Clifford 2002
Centaur pharmaceuticals Clifford 2002
National Institutes of Allergy and Infectious Diseases Navia 1998
Miles Pharmaceuticals Navia 1998
American Foundation for AIDS Research Navia 1998

Table 8. Trials sponsors (Continued)
Somerset Pharmaceuticals Sacktor 2000
Otsuka America Pharmaceuticals Dana 1997
Table 9. Sources of support
Company Support Study ID
British Biotech Donated Lexipafant and placebo Schiffito 1999
ASTA Medica Provided Thiotic acid and matching placebo Dana 1998

Excluded studies
Twenty papers identified by the search strategy were not relevant
and were excluded after reading the full-text. The reasons for ex-
clusion for each individual study are listed in the ’Characteristics
of excluded studies’ section, and can be summarized as follows: -
• Seven studies used antiretroviral interventions not

adjunctive therapies (Brew 2007; Day 1992; Galgani 1997;
Sacktor 2002; Schmitt 1988; Sidtis 1993; Tozzi 1993)
• Five were not a randomized study (Bridge 1989; Bridge
1991; Cysique 2006; Kosten 1997; Villemagne 1996)
• One was a comment / Letter (Anonymous 1996)
• One was drug interaction study (DiCenzo 2004)
• Three were validation studies (Chang 2004; Lee 2003;
Schifitto 2001), and
• Three were carried out on patient with no evidence of
cognitive impairment (Polianova 2003; Schifitto 2006b;
Simpson 1996).
Studies awaiting assessment

A randomized, placebo-controlled study (Vitiello 1998) of two
doses of nimodipine (90mg per day and 300mg per day) for 16
weeks in 49 patients with HIV-associated dementia has been com-
pleted, but was only presented during International Conference
on AIDS, 1998.
Ongoing studies
We identified two relevant trials (Nimodipine 1999; Minocycline
2006) from clinical trials registers (see ’Characteristics of Ongoing
Studies’ for detailed information about the individual trials).
Risk of bias in included studies

See Table 10 for details.
Table 10. Risk-of-bias assessment tool
Trial *Sequence *Conceal- *Blinding *Follow-up *Outcome *Outcome *Others

ment Data Reporting
Schiffito 1999 ADEQUATE ADEQUATE Participants = 90.00% YES YES YES

YES 27/30
Provider = GRADE A
YES
Assessor =
UNCLEAR

Table 10. Risk-of-bias assessment tool (Continued)
Dana 1998 ADEQUATE ADEQUATE Participants = 86.11% YES YES YES

YES 31/36
Provider = GRADE A
YES
Assessor =
UNCLEAR
Schiffito 2006 UNCLEAR UNCLEAR Participants = 93.75% YES YES YES

UNCLEAR 15/16
Provider = GRADE A
UNCLEAR
Assessor =
UNCLEAR
Clifford 2002 ADEQUATE ADEQUATE Participants = 85.94% YES YES YES

YES 55/64
Provider = GRADE A
YES
Assessor =
UNCLEAR
Navia 1998 UNCLEAR UNCLEAR Participants = 71.05% NO YES YES

UNCLEAR 27/38
Provider = GRADE B
UNCLEAR
Assessor =
UNCLEAR
Sacktor 2000 ADEQUATE ADEQUATE Participants = 85.71% YES YES YES

YES 12/14
Provider = GRADE A
YES
Assessor =
YES
Helsetine ADEQUATE UNCLEAR Participants = 66.51% NO YES YES

1998 UNCLEAR 143/215
Provider = GRADE B
UNCLEAR
Assessor =
UNCLEAR
Dana 1997 ADEQUATE ADEQUATE Participants = 60.00% YES YES YES

YES 18/30
Provider = GRADE B
YES
Assessor =
UNCLEAR

Table 10. Risk-of-bias assessment tool (Continued)
Schiffito UNCLEAR UNCLEAR Participants = 79% YES YES YES

2007a UNCLEAR 110/140
Provider = GRADE B
UNCLEAR
Assessor =
UNCLEAR
Schiffito UNCLEAR UNCLEAR Participants = 84% YES YES YES

2007b UNCLEAR 108/128
Provider = Grade A
UNCLEAR
Assessor =
UNCLEAR
*Genera- *Allocation *Complete- *Incomplete *Free of selec- *Free of topic

tion of alloca- concealment ness of follow outcome data tive outcome specific,
tion sequence up: A = >80%; adequately ad- reporting? design specific
B = <80%, C = dressed? or other po-
unclear tential threats
to validity?
Generation of allocation sequence Completeness of follow up

All trials were reported as randomized. Six trials used adequate All trials reported the number of participants lost to follow up
methods - computer-generated random sequences (Clifford 2002; and all randomized participants were included in analysis. The
Dana 1997; Dana 1998; Sacktor 2000; Schiffito 1999) and min- inclusion of randomized participants was adequate in six trials (
imization method as described by the author (Heseltine 1998). Clifford 2002; Dana 1998; Sacktor 2000; Schiffito 1999; Schiffito
The remaining trials reports did not describe their methods of al- 2006a; Schiffito 2007b) and inadequate in four trials (Heseltine
location sequence (Navia 1998; Schiffito 2006a; Schiffito 2007a; 1998; Navia 1998; Dana 1997; Schiffito 2007a). Participant loss
Schiffito 2007b) to follow-up ranged from 6 to 40% (mean 20% across all studies).
Allocation concealment See Table 10 for details of individual studies.
Five trials employed adequate methods for concealing allocation Incomplete outcome
using call-in computer enrolment modules (Clifford 2002; Dana
Incomplete outcome data were adequately addressed in eight trials
1997; Dana 1998; Sacktor 2000; Schiffito 1999). Allocation con-
(Clifford 2002; Dana 1997; Dana 1998; Sacktor 2000; Schiffito
cealed was unclear in the remaining trials (Heseltine 1998; Navia
1999; Schiffito 2006a; Schiffito 2007a; Schiffito 2007b). Reasons
1998; Schiffito 2006a; Schiffito 2007a; Schiffito 2007b).
for discontinuing treatment in the remaining studies were consid-
Blinding ered inadequate, because they were related to true outcomes:
All trials were reported as double-blind. Five trials reported that
both participants and providers were blinded to treatment as-
• Navia 1998: 11 patients out of 38 randomized were loss to
signment (Clifford 2002; Dana 1997; Dana 1998; Sacktor 2000;
follow up due to neurologic and toxicity endpoint
Schiffito 1999); and assessor (personal communication Sacktor
2000). Blinding methods were unclear in the other five trials ( • Heseltine 1998: number of deaths (eight in Peptide T
Heseltine 1998; Navia 1998; Schiffito 2006a; Schiffito 2007a; group vs three in placebo group) and entry into salvage protocol
Schiffito 2007b). (eight in Peptide T group vs six in placebo group).

Selective outcome reporting • Number of patient experiencing any adverse effects (RR
0.58; 95% CI 0.34 to 1.00)
All trials were free of suggestive outcome reporting.
• Diarrhoea (RR 0.29; 95% CI 0.01 to 6.69)
T opic specific, design specific or other potential threats to • Herpes zoster (RR 0.29; 95% CI 0.01 to 6.69)
validity • Blepharitis (RR 2.65; 95% CI 0.12 to 60.21)
• Headache (RR 0.58; 95% CI 0.11 to 3.00)
All trials were apparently free of design specific or other potential
• Gastroesophageal reflux (RR 0.58; 95% CI 0.11 to 3.00)
threat to validity.
• Vomiting (RR 0.18; 95% CI 0.01 to 3.39)
• Nephrolithiasis (RR 0.88; 95% CI 0.06 to 12.73)
Effects of interventions • Bacteria pneumonia (RR 0.88; 95% CI 0.06 to 12.73)
The studies could not be combined for meta-analysis due to het-
02. Peptide T vs placebo (Heseltine 1998)
erogeneity in interventions tested (Table 6). Results for individ-
02.01 Neurocognitive performance scores (see ’Analysis 2.1’)
ual comparisons are, therefore presented by outcome. For more No difference was found between the two treatment groups on
details on individual studies, please refer to the ’Characteristics of the score changes of the global and seven cognitive domains using
Included Studies’ and ’Analyses’
’available case analysis’ (all patients for whom complete outcome
01. Lexipafant vs placebo (Schiffito 1999)
data were available 66/106 in peptide T arm and 77/109 in placebo
01.01 Neuropsychological test scores (see ’Analysis 1.1’)
arm):
Several tests were used to assess neuropsychological function.
There was trend toward improvement in the Rey auditory verbal • Global (WMD 0.08; 95% CI -0.03 to 0.19)
learning (RAVL) - recall after interference (WMD 2.20; 95% CI • Verbal fluency (WMD -0.01; 95% CI -0.25 to 0.23)
0.68 to 3.72), favouring the lexipafant group. There were no sig- • Visuospatial ability (WMD -0.08; 95% CI -0.24 to 0.08)
nificant treatment effects for other measures of neuropsychologi- • Abstract thinking (WMD 0.11; 95% CI -0.05 to 0.27)
cal function: • Speed of information processing (WMD 0.09; 95% CI -
0.08 to 0.26)
• Rey auditory verbal learning - Total number correct (WMD
• Working memory (WMD 0.15; 95% CI -0.07 to 0.37)
4.10; 95% CI -0.53 to 8.73)
• Learning and retention (WMD 0.08; 95% CI -0.10 to
• Rey auditory verbal learning - Number correct Trial 5
0.26)
(WMD 0.80; 95% CI -0.39 to 1.99)
• Motor performance (WMD 0.01; 95% CI -0.16 to 0.18)
• Rey auditory verbal learning - Delayed Recall (WMD 1.40;
95% CI -0.40 to 3.20) 02.02 Tolerability (see ’Analysis 2.2’)
• Rey auditory verbal learning - Correct Recognition (WMD There was no significant difference between intervention and con-
-0.20; 95% CI -1.63 to 1.23) trol group in the number of those that completed the assign dosage
• Mean reaction time - Choice (WMD -5.10; 95% CI - of experimental medication (RR 0.88; 95% CI 0.73 to 1.07).
102.44 to 92.24) 02.03 All cause mortality (see ’Analysis 2.3’)
• Mean reaction time - Sequential (WMD 11.70; 95% CI - There were more death in peptide T group compared to placebo,
71.48 to 94.88) but the difference did not reach statistical significance (RR 2.88;
• Digit Symbol (WMD 1.30; 95% CI -2.87 to 5.47) 95% CI 0.74 to 11.18).
• Groove pegboard - Dominant Hand (WMD 3.50; 95% CI 02.04 Adverse effects (see ’Analysis 2.4’)
-8.39 to 15.39) Various adverse effects were recorded, peripheral neuropathy event
• Groove pegboard - Nondominant Hand (WMD 1.80; 95% show a significant difference in favour of peptide compared to
CI -9.32 to 12.92) placebo (RR 0.37; 95% CI 0.17 to 0.80). There were no significant
• Timed Gait (WMD 0.00; 95% CI -1.58 to 1.58) differences in all other adverse events:
01.02 Tolerability (see ’Analysis 1.2’) • Number of patient experiencing any adverse effects (RR
The analysis of tolerability (treatment completers) (RR 0.94; CI 1.03; 95% CI 0.88 to 1.21)
0.74 to 1.19) showed no significant difference between interven- • Worsening of mood disorder (RR 7.20; 95% CI 0.90 to
tion and control groups. 57.51)
01.03 All cause mortality • Rash (RR 0.83; 95% CI 0.58 to 1.20)
No deaths were reported in both arms. • Nasal congestion (RR 2.47; 95% CI 0.90 to 6.77)
01.04 Adverse effects (see ’Analysis 1.4’) • Proteinuria (RR 2.06; 95% CI 0.97 to 4.37)
Various adverse effects were recorded, none of the events showed
statistical difference between the intervention and control groups: 03. OPC-1147 vs placebo (Dana 1997)
03.01 Neuropsychological test scores (see ’Analysis 3.1’)

OPC-1147 compared placebo showed no statistical significant • Groove pegboard - Dominant Hand (WMD -10.33; 95%
benefit on neurocognitive functions as measured by the neuropsy- CI -27.31 to 6.65)
chological test scores: • Groove pegboard - Nondominant Hand (WMD -3.94;
95% CI -21.02 to 13.14)
• Rey auditory verbal learning - Total number correct (WMD • Timed Gait (WMD -0.39; 95% CI -1.34 to 0.56)
2.30; 95% CI -4.68 to 9.28) • Composite neuropsychological Z score (WMD 2.55; 95%
• Rey auditory verbal learning - Recall after Interference CI -2.57 to 7.67)
(WMD 1.70; 95% CI -0.34 to 3.74)
• Rey auditory verbal learning - Delayed Recall (WMD 1.50; 04.02 Tolerability (see ’Analysis 4.2’)
95% CI -0.50 to 3.50) There was no significant difference between intervention and con-
• Digit Symbol (WMD 0.40; 95% CI -3.45 to 4.25) trol group in the number of those that complete the assign dosage
• Groove pegboard - Dominant Hand (WMD -7.30; 95% of experimental medication (RR 1.07; 95% CI 0.70 to 1.63)
CI -24.27 to 9.67) 04.03 All cause mortality
• Groove pegboard - Nondominant Hand (WMD -4.20; No deaths were reported in both arms.
95% CI -14.81 to 6.41) 04.04 Adverse effects (see ’Analysis 4.4’)
• Timed Gait (WMD 1.10; 95% CI -0.59 to 2.79) Various adverse effects were recorded, none of the events showed
statistical difference between the intervention and control groups:
03.02 Tolerability (see ’Analysis 3.2’) • Number of patient experiencing any adverse effects (RR
There was no significant difference between intervention and con- 3.50; 95% CI 0.20 to 62.27)
trol group in the number of those that complete the assign dosage • Abnormal liver function (RR 2.10; 95% CI 0.10 to 44.40)
of experimental medication (RR 0.78; 95% CI 0.39 to 1.54) • Gastroesophageal reflux (RR 2.10; 95% CI 0.10 to 44.40)
03.03 All cause mortality
No deaths were reported in both arms. 05. Memantine vs placebo (Schiffito 2007a)
03.04 Adverse effects (see ’Analysis 3.4’) 05.01 Neuropsychological test scores (see ’Analysis 5.1’)
Fewer subjects experienced any adverse effects in OPC-1147 com- There was no statistically significance difference between meman-
pared to placebo group (RR 0.54; 95% CI; 0.30 to 0.96). Al- tine and placebo on composite NPZ test score (WMD 43; 95%
though, there were no significant differences in any of the indi- CI 10.16 to 96.16).
vidual adverse events: 05.02 Tolerability (see ’Analysis 5.2’)
There was no significant difference between intervention and con-
• Fatigue (RR 0.60; 95% CI; 0.17 to 2.07) trol group in the number of those that complete the assign dosage
• Diarrhoea (RR 1.50; 95% CI; 0.29 to 7.73) of experimental medication (RR 0.96; 95% CI 0.81 to 1.85)
• Nausea (RR 0.33; 95% CI; 0.04 to 2.85) 05.03 All cause mortality
• Cough (RR 0.14; 95% CI; 0.01 to 2.55) There were two deaths during the first 16 weeks, both in the
placebo arm (RR 0.20; 95% CI 0.01 to 4.09).
04. Valproic acid vs placebo (Schiffito 2006a) 05.04 Adverse effects (see ’Analysis 5.4’)
04.01 Neuropsychological test scores (see ’Analysis 4.1’) There were nine subjects in the memantine arm who experienced
For neuropsychological test scores there was no significant differ- Grade 1 toxicity and none in the placebo arm (RR 19; 95% CI 1.13
ence between valproic acid and placebo groups on: to 320.27). There was no statistically difference between number
• Rey auditory verbal learning - Total number correct (WMD of subjects who experienced grade 2 or more toxicities in both
2.22; 95% CI -5.16 to 9.60) arms (RR 1.20; 95% CI 0.73 to 1.96).
• Rey auditory verbal learning - Number correct Trial 5 06. Thioctic acid vs placebo (Dana (thioctic) 1998)
(WMD -0.39; 95% CI -2.37 to 1.59) 06.01 Neuropsychological test scores (see ’Analysis 6.1’)
• Rey auditory verbal learning - Recall after Interference Several tests were used to assess neuropsychological function. Sub-
(WMD 4.78; 95% CI -3.73 to 13.29) jects on thioctic acid performed worse on Rey auditory verbal
• Rey auditory verbal learning - Delayed Recall (WMD 1.78; learning (RAVL) - total number of correct (RR -11.00; 95% CI -
95% CI -0.35 to 3.91) 15.95 to -6.05) and better on mean reaction time sequential (RR
• Rey auditory verbal learning - Correct Recognition (WMD 70.70; 95% CI 12.15 to 129.25) than subjects did not receive
-0.72; 95% CI -5.60 to 4.16) thioctic acid. There were no significant treatment effects for other
• Mean reaction time - Choice (WMD -19.38; 95% CI - measures of neuropsychological function:
148.91 to 110.15) • Rey auditory verbal learning - Number correct Trial 5
• Mean reaction time - Sequential (WMD 56.04; 95% CI - (WMD -0.60; 95% CI -2.54 to 1.34)
54.18 to 166.26) • Rey auditory verbal learning - Recall after Interference
• Digit Symbol (WMD 4.56; 95% CI -2.43 to 11.55) (WMD -1.60; 95% CI -3.59 to 0.39)

• Rey auditory verbal learning - Delayed Recall (WMD - The total number of subjects that experienced any adverse effects
1.20; 95% CI -2.78 to 0.38) were more in experimental arm compared to placebo group (RR
• Rey auditory verbal learning - Correct Recognition (WMD 8.00; 95% CI 1.24 to 51.51). Although, there were no significant
-0.60; 95% CI -1.92 to 0.72) differences in any of the individual adverse events:
• Mean reaction time - Choice (WMD -0.50; 95% CI -53.58
to 52.58) • Headache (RR 2.00; 95% CI 0.22 to 18.33)
• Digit Symbol (WMD -0.30; 95% CI -7.64 to 7.04) • Nausea (RR 5.00; 95% CI 0.27 to 91.52)
• Groove pegboard - Dominant Hand (WMD 0.70; 95% CI • Acute lymphocytic leukemia (RR 5.00; 95% CI 0.27 to
-14.26 to 15.66) 91.52)
• Groove pegboard - Nondominant Hand (WMD 13.30; • Bacteria pneumonia (RR 3.00; 95% CI 0.14 to 65.16)
95% CI -1.43 to 28.03) • Flu (RR 3.00; 95% CI 0.14 to 65.16)
08. Thioctic acid and deprenyl (selegiline) vs placebo (Dana

06.02 Tolerability (see ’Analysis 6.2’)
(both) 1998)
There was no significant difference between intervention and con-
08.01. Neuropsychological test scores (see ’Analysis 8.1’)
trol group in the number of those that complete the assign dosage
No statistical significant differences between thioctic acid and de-
of experimental medication (RR 4.00; 95% CI 0.33 to 48.66)
prenyl (selegiline) and placebo group emerged in any of the neu-
06.03 All cause mortality
ropsychological test scores:
No deaths were reported in both arms.
06.04 Adverse effects (see ’Analysis 6.4’) • Rey auditory verbal learning - Total number correct (WMD
One case of headache was reported in each arm (RR 1.00; 95% -2.60; 95% CI -7.95 to 2.75)
CI 0.07 to 13.64) • Rey auditory verbal learning - Number correct Trial 5
07. Oral deprenyl (selegiline) vs placebo (Dana 1998) (WMD 1.00; 95% CI -0.56 to 2.56)
07.01 Neuropsychological test scores (see ’Analysis 7.1’) • Rey auditory verbal learning - Recall after Interference
Deprenyl arm performed significantly better on the Rey auditory (WMD -1.00; 95% CI -3.68 to 1.68)
verbal learning (RAVL) - number of correct trial 5 (RR 1.80; 95% • Rey auditory verbal learning - Delayed Recall (WMD 0.10;
CI 0.24 to 3.36) and RAVL-recall after interference (RR 1.90; 95% CI -1.37 to 1.57)
95% CI 0.07 to 3.73) compared to placebo arm. There were no • Rey auditory verbal learning - Correct Recognition (WMD
significant treatment effects for other measures of neuropsycho- -0.10; 95% CI -1.77 to 1.57)
logical function: • Mean reaction time - Choice (WMD -4.30; 95% CI -58.71
to 50.11)
• Rey auditory verbal learning - Total number correct (WMD
• Mean reaction time - Sequential (WMD -5.20; 95% CI -
2.80; 95% CI-2.10 to 7.70)
49.57 to 39.17)
• Rey auditory verbal learning - Delayed Recall (WMD 0.70;
• Groove pegboard - Dominant Hand (WMD -10.80; 95%
95% CI-1.11 to 2.51)
CI -26.85 to 5.25)
• Rey auditory verbal learning - Correct Recognition (WMD
• Groove pegboard - Nondominant Hand (WMD 4.70; 95%
1.30; 95% CI-0.41 to 3.01)
CI -12.53 to 21.93)
• Digit Symbol (WMD 5.90; 95% CI-0.05 to 11.85)
• Digit Symbol (WMD 0.60; 95% CI -4.78 to 5.98)
• Groove pegboard - Dominant Hand (WMD -10.00; 95%
CI-22.37 to 2.37) 08.02 Tolerability (see ’Analysis 8.2’)
• Groove pegboard - Nondominant Hand (WMD 3.10; 95% There was no significant difference between intervention and con-
CI-12.79 to 18.99) trol group in the number of those that complete the assign dosage
• Mean reaction time - Choice (WMD 24.40; 95% CI-28.86 of experimental medication (RR 1.46; 95% CI 0.91 to 2.35)
to 77.66) 08.03 All cause mortality (see ’Analysis 8.3’)
• Mean reaction time - Sequential (WMD 20.00; 95% CI- No deaths were reported in both arms.
33.12 to 73.12)
08.04 Adverse effects (see ’Analysis 8.4’)
07.02 Tolerability (see ’Analysis 7.2’) Various adverse effects were recorded, none of the events showed
There was no significant difference between intervention and con- statistical difference between the intervention and control groups:
trol group in the number of those that complete the assign dosage
of experimental medication (RR 1.33; 95% CI 0.80 to 2.23). • Number of patient experiencing any adverse effects (RR
07.03 All cause mortality 6.00; 95% CI 0.89 to 40.31)
No deaths were reported in both arms. • Headache (RR 2.00; 95% CI 0.22 to 18.33)
07.04 Adverse effects (see ’Analysis 7.4’) • Nausea (RR 5.00; 95% CI 0.27 to 91.52)

• Acute lymphocytic leukemia (RR 3.00; 95% CI 0.14 to 15.48)
65.16) 10. 04 Adverse effects (see ’Analysis 10.4’)
• Neutropenia (RR 3.00; 95% CI 0.14 to 65.16) Various adverse effects were recorded, none of the events showed
statistical difference between the intervention and control groups:
09. Low dose transdermal deprenyl (selegiline) vs placebo (
Sacktor 2000; Schiffito 2007b) • Rash (RR 3.00; 95% CI 0.13 to 71.65)
09.01 Neuropsychological test scores (see ’Analysis 9.1’) • Back pain (RR .033; 95% CI 0.01 to 7.96)
Two trials (Sacktor 2000; Schiffito 2007b) compared low dose • Parasthesia (RR 3.00; 95% CI 0.13 to 71.65)
transdermal selegiline with placebo. Changes from baseline on the • Headache (RR 0.20; 95% CI 0.01 to 4.05)
neuropsychological test scores results illustrate varieties of findings • Shortness of breath (RR .033; 95% CI 0.01 to 7.96)
in Sacktor 2000. The selegiline group performed better than the
placebo group on Rey Auditory Verbal Learning - Delayed Recall 11. Low dose (50mg) CPI-1189 vs placebo (Clifford 2002)
(WMD 2.90; 95% CI 0.13 to 5.67) and Groove pegboard - Non- 11.01 Neuropsychological test scores (see ’Analysis 11.1’)
dominant Hand (WMD 23.50; 95% CI 4.11 to 42.89). While No statistically significant evidence in favour of low dose (50mg)
placebo group showed improved on mean reaction time -choice CPI-1189 emerged in all the neuropsychological test scores:
test score (WMD -98.50; 95% CI -153.98 to -43.02). There • Rey Auditory Verbal Learning - Total Number Correct
were no differences between the groups on other tests. Individ- (WMD -2.60; 95% CI -7.99 to 2.79)
ual neuropsychological performance tests scores were not reported • Rey Auditory Verbal Learning - Number Correct - Trial 5
in Schiffito 2007b. However, no statistically significant evidence (WMD -1.40; 95% CI -2.79 to -0.01)
in favour of low dose transdermal selegiline emerged in all the • Rey Auditory Verbal Learning - Recall after Interference
reported three neuropsychological composite test scores (NPZ): (WMD -1.50; 95% CI -3.30 to 0.30)
NPZ- 6 (WMD -0.03; 95% CI -0.30 to 0.24), NPZ-8 (WMD - • Rey Auditory Verbal Learning - Delayed Recall (WMD -
0.01; 95% CI -0.22 to 0.20), and NPZ total (WD 0.06; 95% CI 0.80; 95% CI -2.92 to 1.32)
-0.17 to 0.29). • Mean Reaction Time - Choice (WMD 22.00; 95% CI -
09.02 Tolerability (see ’Analysis 9.2’) 24.22 to 68.22)
There was no significant difference between intervention and con- • Mean Reaction Time - Sequential (WMD -2.40; 95% CI -
trol group in the number of those that complete the assign dosage 52.47 to 47.67)
of experimental medication, with combined RR 0.99; 95% CI • Trail Making - Part A (WMD 0.00; 95% CI -0.10 to 0.10)
0.83 to 1.17, with no significant heterogeneity (I2 =0%; Chi2 • Trail Making - Part B (WMD 0.00; 95% CI -0.10 to 0.10)
=0.26; P=0.61) • Digit Symbol (WMD -2.60; 95% CI -8.47 to 3.27)
09.03 All cause mortality • Grooved Pegboard - Dominant Hand (WMD -2.40; 95%
No deaths were reported in both arms in Sacktor 2000 and CI -13.86 to 9.06)
Schiffito 2007b reported one death in each arm (RR 1.02; 95% • Grooved Pegboard - Nondominant Hand (WMD 5.40;
CI 0.07 to 15.84). 95% CI -4.57 to 15.37)
09. 04 Adverse effects (see ’Analysis 9.4’) • Finger Tapping - Dominant Hand (WMD -1.80; 95% CI -
Various adverse effects were recorded, none of the events showed 6.30 to 2.70)
statistical difference between the intervention and control groups. • Finger Tapping - Nondominant Hand (WMD -1.60; 95%
CI -5.46 to 2.26)
10. High dose transdermal deprenyl (selegiline) vs placebo ( • Timed Gait (WMD 0.40; 95% CI -0.54 to 1.34)
Schifitto (high)2007) • Composite neuropsychological Z score (WMD 0.30; 95%
10.01 Neuropsychological test scores (see ’Analysis 10.1’) CI -0.16 to 0.76)
Individual neuropsychological performance tests scores were not
reported. No statistically significant evidence in favour of high 11.02 Tolerability (see ’Analysis 11.2’)
dose transdermal selegiline emerged in all the reported three neu- More subjects were able to tolerate experimental medication com-
ropsychological composite test scores (NPZ): NPZ- 6 (WD 0.04; pared to placebo, but the difference did not reach statistical sig-
95% CI -0.26 to 0.34), NPZ-8 (WD -0.21; 95% CI -0.48 to 0.1), nificant (RR 1.19; 95% CI 0.90 to 1.57).
and NPZ total (WD 0.02; 95% CI -0.26 to 0.30).
10.02 Tolerability (see ’Analysis 10.2’) 11.03 All cause mortality (see ’Analysis 11.3’)
There was no significant difference between intervention and con- There was one death in placebo group (RR 0.33; 95% CI 0.01 to
trol group in the number of those that complete the assign dosage 7.74).
of experimental medication (RR 0.97; 95% CI 0.81 to 1.16) 11.04 Adverse effects (see ’Analysis 11.4)
10.03 All cause mortality The total number of subjects that experienced any adverse effects
One death was reported in each arm (RR 1.00; 95% CI 0.06 to were more in experimental arm compared to placebo group (RR

0.60; 95% CI 0.41 to 0.88). Although, there were no significant Various adverse effects were recorded, none of the events showed
differences in any of the individual adverse events: statistical difference between the intervention and control groups:
• Abdominal pain (RR 0.50; 95% CI 0.05 to 5.10) • Number of patient experiencing any adverse effects (RR
• Diarrhoea (RR 0.67; 95% CI 0.12 to 3.59) 0.95; 95% CI 0.81 to 1.12)
• Fatigue (RR 0.33; 95% CI 0.01 to 7.74) • Abdominal pain (RR 0.19; 95% CI 0.01 to 3.76)
• Fever (RR 1.00; 95% CI 0.07 to 14.95) • Diarrhoea (RR 0.95; 95% CI 0.22 to 4.21)
• Headache (RR 1.00; 95% CI 0.16 to 6.45) • Fatigue (RR 1.91; 95% CI 0.19 to 19.52)
• Rash (RR 0.60; 95% CI 0.16 to 2.20) • Fever (RR 0.95; 95% CI 0.06 to 14.30)
• Upper respiratory tract infection (RR 0.50; 95% CI 0.05 to • Headache (RR 1.91; 95% CI 0.39 to 9.35)
5.10) • Rash (RR 0.57; 95% CI 0.16 to 2.10)
• Weight decrease (RR 0.20; 95% CI 0.01 to 3.93) • Upper respiratory tract infections (RR 1.43; 95% CI 0.27
• Nausea (RR 1.00; 95% CI 0.16 to 6.45) to 7.73)
• Weight decrease (RR 0.48; 95% CI 0.05 to 4.88)
12. High dose (100mg) CPI-1189 vs placebo (Clifford (high) • Nausea (RR 1.43; 95% CI 0.27 to 7.73)
2002)
12.01 Neuropsychological test scores (see ’Analysis 12.1’) 13. Low dose nimodipine vs placebo (Navia 1998)
The composite neuropsychological Z score (WMD 0.50; 95% 13.01 Neuropsychological test scores
CI 0.08 to 0.92) and Grooved Pegboard - Nondominant Hand Individual neuropsychological performance tests scores were not
(WMD 11.50; 95% CI 0.98 to 22.02) were improved in the reported. A summary measure, neuropsychological percent change
100mg/day CPI-1189 arm compared with placebo. There were (NPC) was presented graphically. The authors reported an increase
no significant treatment effects on other test: in NPC in placebo group whereas the nimodipine group displayed
a marginal decrease.
• Rey Auditory Verbal Learning - Total Number Correct
(WMD 0.60; 95% CI -4.82 to 6.02) 13.02 Tolerability (see ’Analysis 13.2’)
• Rey Auditory Verbal Learning - Number Correct - Trial 5 There was no significant difference between intervention and con-
(WMD 0.00; 95% CI -1.08 to 1.08) trol group in the number of those that complete the assign dosage
• Rey Auditory Verbal Learning - Recall after Interference of experimental medication (RR 1.57; 95% CI 0.88 to 2.81)
(WMD 0.10; 95% CI -1.89 to 2.11) 13.03 All cause mortality (see ’Analysis 13.3’)
• Rey Auditory Verbal Learning - Delayed Recall (WMD One death was reported in each arm (RR 0.79; 95% CI 0.06 to
0.50; 95% CI -1.43 to 2.43) 11.20)
• Mean Reaction Time - Choice (WMD 24.20; 95% CI - 13.04 Adverse effects (see ’Analysis 13.4’)
21.38 to 69.78) Various adverse effects were recorded, none of the events showed
• Mean Reaction Time - Sequential (WMD -25.10; 95% CI statistical difference between the intervention and control groups:
-82.04 to 31.84)
• Trail Making - Part A (WMD 0.00; 95% CI -0.10 to 0.10) • Number of patient experiencing any adverse effects (RR
• Trail Making - Part B (WMD 0.00; 95% CI -0.11 to 0.11) 0.52; 95% CI 0.19 to 1.41)
• Digit Symbol (WMD -1.00; 95% CI -6.75 to 4.75) • Haematological toxicity (RR 0.52; 95% CI 0.11 to 2.61)
• Grooved Pegboard - Dominant Hand (WMD 2.60; 95% • Cardiac toxicity (RR 0.79; 95% CI 0.06 to 11.20)
CI -7.08 to 12.28) • Psychiatrics toxicity (RR 0.79; 95% CI 0.06 to 11.20)
• Finger Tapping - Dominant Hand (WMD -2.80; 95% CI - • Hypotension (RR 0.27; 95% CI 0.01 to 5.97)
7.59 to 1.99)
• Finger Tapping - Nondominant Hand (WMD -2.10; 95% 14. High dose nimodipine vs placebo (Navia (high) 1998)
CI -6.00 to 1.80) 14.01 Neuropsychological test scores
• Timed Gait (WMD 0.60; 95% CI -0.61 to 1.81 Individual neuropsychological performance tests scores were not
reported. A summary measure, neuropsychological percent change
12.02 Tolerability (see ’Analysis 12.2’) (NPC) was presented graphically. The authors reported an increase
More subjects were able to tolerate experimental medication com- in NPC in placebo group whereas the nimodipine group displayed
pared to placebo, but the difference did not reach statistical sig- a marginal decrease.
nificant (RR 1.19; 95% CI 0.91 to 1.57).
12.03 All cause mortality (see ’Analysis 12.3’) 14.02 Tolerability (see ’Analysis 14.2’)
There was one death in placebo group (RR 0.32; 95% CI 0.01 to There was no significant difference between intervention and con-
7.42]) trol group in the number of those that complete the assign dosage
12.04 Adverse effects (see ’Analysis 12.4’) of experimental medication (RR 1.88; 95% CI 0.35 to 9.98)

14.03 All cause mortality (see ’Analysis 14.3’) However, there was some evidence of a beneficial effect in terms
One death was reported in each arm (RR 0.85; 95% CI 0.06 to of improvement and better performance in some of the neuropsy-
12.01) chological test scores in some of the drugs (deprenyl, 100mg/day
14.04 Adverse effects (see ’Analysis 14.9’) CPI-1189, and valproic acid) compared to placebo. All the ad-
Various adverse effects were recorded, none of the events showed junctive therapies were well tolerated, safe, and no clinically sig-
statistical difference between the intervention and control groups: nificant toxic effects emerged in participants with ADC. Though
there were no significant differences among treatment groups re-
• Number of patient experiencing any adverse effects (RR garding the primary and secondary measures of tolerability; and
0.71; 95% CI 0.29 to 1.69) incidences of both common and severe adverse events. We found
• Haematological toxicity (RR 1.13; 95% CI 0.32 to 3.99) like Turchan 2003 that methods of analysis and reporting the bat-
• Cardiac toxicity (RR 0.29; 95% CI 0.01 to 6.38) teries of neuropsychological test scores vary between the studies
• Psychiatrics toxicity (RR 0.85; 95% CI 0.06 to 12.01) making them difficult to compare.
• Hypotension (RR 0.29; 95% CI 0.01 to 6.38)
Strengths and limitations of the review
15. Any adjunctive therapy vs placebo Our review had several strengths. We performed a comprehensive
For numerical details see ’comparisons and data tables’: search of several databases and sources to identify studies. We con-
tacted authors for original study articles. We evaluated the qual-
• Tolerability - Analysis 15.1
ity of included studies using established quality criteria. Screening
• All cause mortality- Analysis 15.2
for study eligibility and data abstraction were performed by two
• Number of patient experiencing any adverse effects -
authors. We tried to minimize publication bias by including both
Analysis 15.3.
published and unpublished data. Our review had noteworthy lim-
itation. This review was designed to access both the efficacy and
safety of adjunctive therapies for ADC, all the studies reviewed
were phase 2 trials. The primary end points were safety and toler-
ability, and were not designed primarily designed to detect clinical
DISCUSSION
improvement. To date not a single phase 3 trial of these agents has
This review of adjunctive therapies for AIDS dementia com- been conducted.
plex (ADC) has brought together evidence from ten randomized, Strengths and limitations of the available evidence
placebo controlled, double blind trials of peptide T, selegiline (de-
prenyl), CPI-1189, OPC-14117, valproic acid, nimodipine, me- Most of the studies reviewed are of good quality. Generation of
mantine, and lexipafant primarily from the last ten years incorpo- allocation sequence, allocation concealment, sample size calcula-
rating 711 subjects. tion, and blinding were adequate in most. All the trials reported
the number of participants lost to follow up and all randomized
Main findings participants were included in the analysis in most. All the trials
were free of suggestive outcome reporting and used validated bat-
A goal of systematic reviews and meta-analyses is to provide the
teries of neuropsychological test.
clinician or researcher with a balanced appraisal of the totality of
the evidence in an area, ideally by reading one review, as opposed We identified eight randomized trials, but they had important
to a multitude of individual studies. This allows for a more ob- limitation in term of sample size, most were small. The small size
jective appraisal of the evidence, which may lead to resolution of of these trials lack statistical power and make it difficult to draw
uncertainty and disagreement. This process should be free of bias, a reliable conclusion. The numbers of participants lost to follow
such as multiple publications of an individual study, and should up were very high in some (Heseltine 1998 [33.49%]; Dana 1997
synthesize the findings using clinically meaningful endpoints. In [40.00%]) and this may lead to imbalances in the comparison
reviewing the evidence presented here, we hoped to identify effec- groups. Almost all of the trials included are not independent of
tive intervention strategies for the treatment of ADC. Unfortu- the drug manufacturing or marketing companies. Care should be
nately, the results of the trials in this review did not allow for pool- taken in transposing these results to other settings because the data
ing of some key endpoints, so we have provided a predominantly are mainly from USA involving multicenter AIDS cohort study (
narrative review of the evidence. The trials were heterogeneous in Selnes 1991) subjects (which have inherent referral and selection
terms of types, dosages, routes and frequencies of administration biases(Sacktor 1997)) and subjects were recruited based on clinical
of the adjunctive therapies. Although we identified a number of confirmation of the American Academy of Neurology algorithm
studies for inclusion in the review, we found no clear evidence of for ADC (Dana 1996). In addition, data provided in one trial (
benefit (or harm) for using these adjunctive therapies to improve Navia 1998) were not complete to allow us to statistically re-eval-
clinical outcomes in patients with ADC. uate the evidence on neuropsychological test scores, and available

case analysis was used to analysis neurocognitive performance test • Review, and refine, the methodology for trials of
(Heseltine 1998). neuropsychological endpoints, in people with ADC;
• Recruit an adequate number of participants and provide
treatment for a sufficient duration and follow up;
AUTHORS’ CONCLUSIONS • Interventions need to be carefully specified and monitored;
include careful assessment and complete reporting of adverse
Implications for practice events; and
There is a lack of evidence that any of the adjunctive therapies • Define a priori an unambiguous, measurable, primary
improves cognitive performance or quality of life, or both for pa- endpoint; and report results for all outcomes using CONSORT
tients with ADC. The evidence of safety and tolerability of these statement.
drugs does not justify recommending its routine clinical use.
Implications for research

There is a need for well designed, adequately powered, phase III ACKNOWLEDGEMENTS
trials on adjunctive therapies for ADC, with main objective on Uthman AO was awarded a Reviews for Africa Programme
cognitive performance and/or quality of life. In addition there is Fellowship (www.mrc.ac.za/cochrane/rap.htm), funded from the
a need for more research in transitional and developing countries Nuffield Commonwealth Programme, through The Nuffield
with high burden of HIV and ADC. Foundation. We would like to acknowledge the support pro-
Future trials should address the following: vided by the staff at the South African Cochrane Centre and the
Cochrane HIV/AIDS Review group mentoring programme. Mar-
tie Muller (Uthman AO’s mentor).
REFERENCES
References to studies included in this review Therapy of HIV Dementia and Related Cognitive Disorders.
Neurology 1997;49(1):142–6.
Clifford (high) 2002 {published data only}
Clifford DB, McArthur JC, Schifitto G, Kieburtz K, McDermott Dana 1998 {published data only}
MP, Letendre S, et al.A randomized clinical trial of CPI-1189 for
∗
A randomized, double-blind, placebo-controlled trial of deprenyl
HIV-associated cognitive-motor impairment. Neurology 2002;59 and thioctic acid in human immunodeficiency virus-associated
(10):1568–73. cognitive impairment. Dana Consortium on the Therapy of HIV
Dementia and Related Cognitive Disorders. Neurology 1998;50(3):
Clifford 2002 {published data only}
645–51.
Clifford DB, McArthur JC, Schifitto G, Kieburtz K, McDermott
MP, Letendre S, et al.A randomized clinical trial of CPI-1189 for Heseltine 1998 {published data only}
HIV-associated cognitive-motor impairment. Neurology 2002;59 Goodkin K, Vitiello B, Lyman WD, Asthana D, Atkinson JH,
(10):1568–73. Heseltine PN, et al.Cerebrospinal and peripheral human
immunodeficiency virus type 1 load in a multisite, randomized,
Dana (both) 1998 {published data only}
double-blind, placebo-controlled trial of D-Ala1-peptide T-amide
for HIV-1-associated cognitive-motor impairment. Journal of
and thioctic acid in human immunodeficiency virus-associated
Neurovirology 2006;12(3):178–89.
cognitive impairment. Dana Consortium on the Therapy of HIV
Heseltine PN, Goodkin K, Atkinson JH, Vitiello B, Rochon J,
Dementia and Related Cognitive Disorders. Neurology 1998;50(3):
Heaton RK, et al.Randomized double-blind placebo-controlled
645–51.
trial of peptide T for HIV-associated cognitive impairment.
Dana (thioctic) 1998 {published data only} Archives of Neurology 1998;55(1):41–51.
and thioctic acid in human immunodeficiency virus-associated Navia (high) 1998 {published data only}
cognitive impairment. Dana Consortium on the Therapy of HIV Navia BA, Dafni U, Simpson D, Tucker T, Singer E, McArthur JC,
Dementia and Related Cognitive Disorders. Neurology 1998;50(3): et al.A phase I/II trial of nimodipine for HIV-related neurologic
645–51. complications. Neurology 1998;51(1):221–8.
Dana 1997 {published data only} Navia 1998 {published data only}
Safety and tolerability of the antioxidant OPC-14117 in HIV- Navia BA, Dafni U, Simpson D, Tucker T, Singer E, McArthur JC,
associated cognitive impairment. The Dana Consortium on the et al.A phase I/II trial of nimodipine for HIV-related neurologic
complications. Neurology 1998;51(1):221–8. Cysique 2006 {published data only}
∗
Cysique LA, Maruff P, Brew BJ. Valproic acid is associated with
Sacktor 2000 {published data only}
cognitive decline in HIV-infected individuals: a clinical
∗
Sacktor N, Schifitto G, McDermott MP, Marder K, McArthur
observational study. BMC neurology 2006;6:42.
JC, Kieburtz K. Transdermal selegiline in HIV-associated cognitive
impairment: pilot, placebo-controlled study. Neurology 2000;54 Day 1992 {published data only}
(1):233–5.
∗
Day JJ, Grant I, Atkinson JH, Brysk LT, McCutchan JA,
Hesselink JR, et al.Incidence of AIDS dementia in a two-year
Schiffito 1999 {published data only} follow-up of AIDS and ARC patients on an initial phase II AZT
Schifitto G, Sacktor N, Marder K, McDermott MP, McArthur JC, placebo-controlled study: San Diego cohort. The Journal of
Kieburtz K, et al.Randomized trial of the platelet-activating factor Neuropsychiatry and Clinical Neurosciences 1992;4(1):15–20.
antagonist lexipafant in HIV-associated cognitive impairment.
Neurological AIDS Research Consortium. Neurology 1999;53(2): DiCenzo 2004 {published data only}
391–6.
∗
DiCenzo R, Peterson D, Cruttenden K, Morse G, Riggs G,
Gelbard H, et al.Effects of valproic acid coadministration on
Schiffito 2006a {published data only} plasma efavirenz and lopinavir concentrations in human
Schifitto G, Peterson DR, Zhong J, Ni H, Cruttenden K, Gaugh immunodeficiency virus-infected adults. Antimicrobial Agents and
M, et al.Valproic acid adjunctive therapy for HIV-associated Chemotherapy 2004;48(11):4328–31.
cognitive impairment: a first report. Neurology 2006;66(6):919–21.
Galgani 1997 {published data only}
Schiffito 2007a {published data only} ∗
Galgani S, Balestra P, Narciso P, Tozzi V, Sette P, Pau F, et
Schifitto G, Navia BA, Yiannoutsos CT, Marra CM, Chang L, al.Nimodipine plus zidovudine versus zidovudine alone in the
Ernst T, et al.Memantine and HIV-associated cognitive treatment of HIV-1-associated cognitive deficits. AIDS 1997;11:
impairment: a neuropsychological and proton magnetic resonance 1520–1.
spectroscopy study. AIDS 2007;21(4):1877–86. Kosten 1997 {published data only}
Schiffito 2007b {published data only}
∗
Kosten TR, Rosen MI, McMahon TL, Bridge TP, O’Malley SS,
Schifitto G, Zhang J, Evans SR, Sacktor N, Simpson D, Millar LL, Pearsall R. Treatment of early AIDS dementia in intravenous drug
et al.A multicenter trial of selegiline transdermal system for HIV- users: high versus low dose peptide T. The American Journal of Drug
associated cognitive impairment. Neurology 2007;69(13):1314–21. and Alcohol Abuse 1997;23(4):543–53.
Lee 2003 {published data only}
Schifitto (high)2007 {published data only} ∗
Lee PL, Yiannoutsos CT, Ernst T, Chang L, Marra CM, Jarvik
Schifitto G, Zhang J, Evans SR, Sacktor N, Simpson D, Millar LL, JG, et al.A multi-center 1H MRS study of the AIDS dementia
et al.A multicenter trial of selegiline transdermal system for HIV- complex: validation and preliminary analysis. Journal of Magnetic
associated cognitive impairment. Neurology 2007;69(13):1314–21. Resonance Imaging 2003;17(6):625–33.
References to studies excluded from this review Polianova 2003 {published data only}
∗
Polianova MT, Ruscetti FW, Pert CB, Tractenberg RE, Leoung G,
et al.Antiviral and immunological benefits in HIV patients
Anonymous 1996 {published data only}
receiving intranasal peptide T (DAPTA). Peptides 2003;24:1093–8.
∗
Analyses of peptide T efficacy on neuropsychological
performance. AIDS Patient Care STDS 1996;10(1):52. Sacktor 2002 {published data only}
∗
Sacktor N, McDermott MP, Marder K, Schifitto G, Selnes OA,
Brew 2007 {published data only} McArthur JC, et al.HIV-associated cognitive impairment before
∗
Brew BJ, Halman M, Catalan J, Sacktor N, Price RW, Brown S, et and after the advent of combination therapy. Journal of
al.Factors in AIDS dementia complex trial design: results and Neurovirology 2002;8(2):136–42.
lessons from the abacavir trial. PLoS Clinical Trials 2007;2(3):e13.
Schifitto 2001 {published data only}
Bridge 1989 {published data only} Schifitto G, Kieburtz K, McDermott MP, McArthur J, Marder K,
∗
Bridge TP, Heseltine PN, Parker ES, Eaton E, Ingraham LJ, Gill Sacktor N, et al.Clinical trials in HIV-associated cognitive
M, et al.Improvement in AIDS patients on peptide T. Lancet 1989; impairment: cognitive and functional outcomes. Neurology 2001;
2:226–7. 56(3):415–8.
Bridge 1991 {published data only} Schifitto 2006b {published data only}
∗
Bridge TP, Heseltine PN, Parker ES, Eaton EM, Ingraham LJ, ∗
Schifitto G, Yiannoutsos CT, Simpson DM, Marra CM, Singer
McGrail ML, et al.Results of extended peptide T administration in EJ, Kolson DL, et al.A placebo-controlled study of memantine for
AIDS and ARC patients. Psychopharmacology bulletin 1991;27: the treatment of human immunodeficiency virus-associated sensory
237–45. neuropathy. Journal of Neurovirology 2006;12(4):328–31.
Chang 2004 {published data only} Schmitt 1988 {published data only}
∗
Chang L, Lee PL, Yiannoutsos CT, Ernst T, Marra CM, Richards ∗
Schmitt FA, Bigley JW, McKinnis R, Logue PE, Evans RW,
T, et al.A multicenter in vivo proton-MRS study of HIV-associated Drucker JL. Neuropsychological outcome of zidovudine (AZT)
dementia and its relationship to age. NeuroImage 2004;23(4): treatment of patients with AIDS and AIDS-related complex. The
1336–47. New England Journal of Medicine 1988;319(24):1573–8.
Sidtis 1993 {published data only} Childs 1999
∗
Sidtis JJ, Gatsonis C, Price RW, Singer EJ, Collier AC, Richman Childs EA, Lyles RH, Selnes OA, Chen B, Miller EN, Cohen BA, et
DD, et al.Zidovudine treatment of the AIDS dementia complex: al.Plasma viral load and CD4 lymphocytes predict HIV-associated
results of a placebo-controlled trial. AIDS Clinical Trials Group. dementia and sensory neuropathy. Neurology 1999;52(3):607–13.
Annals of Neurology 1993;33(4):343–9. Dana 1996
Simpson 1996 {published data only} Clinical confirmation of the American Academy of Neurology
∗
Simpson DM, Dorfman D, Olney RK, McKinley G, Dobkin J, algorithm for HIV-1-associated cognitive/motor disorder. The
So Y, et al.Peptide T in the treatment of painful distal neuropathy Dana Consortium on Therapy for HIV Dementia and Related
associated with AIDS: results of a placebo-controlled trial. The Cognitive Disorders. Neurology 1996;47(5):1247–53.
Peptide T Neuropathy Study Group. Neurology 1996;47(5): Edward 1987
1254–9. Edward DD. Competition’ cause of AIDS dementia? Science News
Tozzi 1993 {published data only} Sept 1987. www.findarticles.com/p/articles/mi˙m1200/is˙v132/
∗
Tozzi V, Narciso P, Galgani S, Sette P, Balestra P, Gerace C, et ai˙5197123 (accessed 23 July 2006).
al.Effects of zidovudine in 30 patients with mild to end-stage AIDS Egger 1997
dementia complex. AIDS 1993;7(5):683–92. Egger M, Davey Smith G, Schneider M, Minder C. Bias in meta-
Villemagne 1996 {published data only} analysis detected by a simple, graphical test. British Medical Journal
∗
Villemagne VL, Phillips RL, Liu X, Gilson SF, Dannals RF, Wong 1997;315(7109):629–34.
DF, et al.Peptide T and glucose metabolism in AIDS dementia Ghafouri 2006
complex. Journal of Nuclear Medicine 1996;37(7):1177–80. Ghafouri M, Amini S, Khalili K, Sawaya BE. HIV-1 associated
dementia: symptoms and causes. Retrovirology 2006;3:28.
References to studies awaiting assessment
Gonzalez 2006
Gonzalez-Scarano F, Martin-Garcia J. The neuropathogenesis of
Vitiello 1998 {unpublished data only}
AIDS. Nature Reviews Immunology 2005;5(1):69–81.
Vitiello B. Research on possible treatments of HIV-associated
cognitive impairment. The XII International AIDS Conference; Higgins 2003
1998 June 28 - July 3; Geneva, Switzerland. 1998:1148 (abstract Higgins JP, Thompson SG, Deeks JJ, Altman DG. Measuring
no. 60804). inconsistency in meta-analyses. British Medical Journal 2003;327
(7414):557–60.
References to ongoing studies Higgins 2006
Higgins JPT, Green S, editors. Analysis and presenting results.
Minocycline 2006 {unpublished data only} Cochrane Handbook for Systematic Reviews of Interventions 4.2.6
Minocycline for the Treatment of Decreased Mental Function in [updated September 2006]; Section 8. www.cochrane.org/
HIV Infected Adults. www.clinicaltrials.gov/ct/show/ resources/handbook/hbook.htm (accessed 6th October 2006).
NCT00361257?order=1 (accessed 28 May 2007).
Jüni 2001
Nimodipine 1999 {unpublished data only} Jüni P, Altman DG, Egger M. Systematic reviews in health care:
Randomized, Double-Blind, Placebo-Controlled Trial of Assessing the quality of controlled clinical trials. British Medical
Nimodipine for the Neurological Manifestations of HIV-1. Journal 2001;323(7303):42–6.
www.clinicaltrials.gov/ct/show/NCT00000738?order=1 (accessed McArthur 2005
28 May 2007). McArthur JC, Brew BJ, Nath A. Neurological complications of
HIV infection. Lancet Neurology 2005;4(9):543–55.
Additional references
Moher 1999
AIDS institutes 2006 Moher D, Cook DJ, Eastwood S, Olkin I, Rennie D, Stroup DF.
AIDS institutes. Cognitive disorders and HIV/AIDS: HIV - Improving the quality of reports of meta-analyses of randomised
associated dementia and delirium. www.hivguidelines.org/ controlled trials: the QUOROM statement. Quality of Reporting
GuideLine.aspx?pageID=261&guideLineID=44&vType=txt of Meta-analyses. Lancet 1999;354(9193):1896–900.
(accessed 28 August 2006). Mollace 2001
Birbeck 2005 Mollace V, Nottet HS, Clayette P, Turco MC, Muscoli C, Salvemini
Birbeck GL. Human immunodeficiency virus dementia patients in D, et al.Oxidative stress and neuroAIDS: triggers, modulators and
Africa: How many? Who cares? And where to from here?. Journal novel antioxidants. Trends in Neuroscience 2001;24(7):411–6.
of Neurovirolology 2005;11 Suppl 3:30–3. Parnes 2006
Chiesi 1996 Parnes RB. AIDS Dementia Complex. www.thirdage.com/
Chiesi A, Vella S, Dally LG, Pedersen C, Danner S, Johnson AM, et healthgate/files/43788.html (accessed 12 July 2006).
al.Epidemiology of AIDS dementia complex in Europe. AIDS in Price 1996
Europe Study Group. Journal of Acquired Immune Deficiency Price RW. Neurological complications of HIV infection. Lancet
Syndrome and Human Retrovirology 1996;11(1):39–44. 1996;348(9025):445–52.

Ramsay 2007
Ramsay C. How do you include trials with more than two groups
into a single meta-analysis?. www.epoc.uottawa.ca/
FAQmultiplegroups2003.pdf (accessed 06 June 2007).
Review Manager 4.2
The Nordic Cochrane Centre, The Cochrane Collaboration.
Review Manager (RevMan). 4.2 for Windows. Copenhagen: The
Nordic Cochrane Centre, The Cochrane Collaboration, 2003.
Sacktor 1997
Sacktor N, McArthur J. Prospects for therapy of HIV-associated
neurologic diseases. Journal of Neurovirology 1997;3(2):89–101.
Selnes 1991
Selnes OA, Jacobson L, Machado AM, Becker JT, Wesch J, Miller
EN. Normative data for a brief neuropsychological screening
battery. Multicenter AIDS Cohort Study. Perceptual and Motor
Skills 1991;73(2):539–50.
Stern 2001
Stern Y, McDermott MP, Albert S, Palumbo D, Selnes OA,
McArthur J, et al.Factors associated with incident human
immunodeficiency virus-dementia. Archives of Neurology 2001;58
(3):473–9.
Turchan 2003
Turchan J, Sacktor N, Wojna V, Conant K, Nath A.
Neuroprotective therapy for HIV dementia. Current HIV Research
2003;1:373–83.
Zink 2005
Zink MC, Uhrlaub J, DeWitt J, Voelker T, Bullock B, Mankowski
J, et al.Neuroprotective and anti-human immunodeficiency virus
activity of minocycline. Journal of the American Medical Association
2005;293(16):2003–11.
∗
Indicates the major publication for the study

CHARACTERISTICS OF STUDIES
Characteristics of included studies [ordered by study ID]
Clifford (high) 2002
Methods [Refer to Clifford 2002 for details; author described high dose of CPI-1189 versus placebo]
Participants
Interventions
Outcomes
Notes
Risk of bias
Item Authors’ judgement Description
Allocation concealment? Yes A - Adequate
Clifford 2002
Methods Design: RCT

Technique of randomization: Stratified and blocked
Allocation sequence: Computer generated (Adequate)
Allocation concealment: Adequate
Method of concealment: Call-in computer enrolment module
Binding: Double, participants and investigators
Inclusion of all randomized participants: Grade A
Loss to follow up: Overall (14.06%), placebo arm (23.81%), CPI-1189 (50mg) arm ( 9.52%), and CPI-
1189 (100mg) arm ( 9.09%)
Sample size calculation done a priori: Yes
Follow-up: 2, 6, 10 weeks
Participants Number: randomized (n=64), completed (n=55)

Inclusion criteria: HIV infection(confirmed by Western blot), evidence of cognitive impairment, and
stable antiretroviral regimen for six weeks
Exclusion criteria: Active opportunist CNS infection, severe pre-morbid psychiatric illness, history of a
chronic neurologic disorder unrelated to HIV infection
Age: Placebo (mean age = 42.2), CPI-1189 - 50mg (mean age = 44.7), CPI-1189 - 100mg (mean age =
43.3)
Male: Placebo (76.2%), CPI-1189 - 50mg (95.2%), CPI-1189 - 100mg (81.8%)
White: Placebo (42.9%), CPI-1189 - 50mg (66.7%), CPI-1189 - 100mg (45.5%)
Years of HIV: Placebo (mean = 7.7), CPI-1189 - 50mg (mean = 7.8), CPI-1189 - 100mg (mean = 8.3)

Clifford 2002 (Continued)
Interventions (1) CPI-1189 50mg orally once daily

(2) CPI-1189 100mg orally once daily
(3) Placebo
Outcomes (1) Tolerability of the drug - treatment completers

(2) Measures of safety - Adverse effects and abnormal laboratory results and eye examinations
(3) Measures of efficacy - change in neuropsychological test scores, neurologic exam results, measures of
function and mood
Outcome not included in this review
(4) HIV markers
Notes Study location: USA

Multi-center study: Yes
Study date: November 1998 - March 2000
Study sponsors: National Institute of Neurologic Diseases and Stroke grant, GCRC at Columbia, GCTC
at Rochester, and Centaur pharmaceuticals
Risk of bias
Dana (both) 1998
Methods [Refer to Dana 1998 for details; author described both deprenyl and thioctic acid versus placebo]
Participants
Interventions
Outcomes
Notes
Risk of bias
Dana (thioctic) 1998
Methods [Refer to Dana 1998 for details; author described both thioctic acid versus placebo]
Participants

Dana (thioctic) 1998 (Continued)
Interventions
Outcomes
Notes
Risk of bias
Dana 1997
Methods Design: RCT

Inclusion of all randomized participants: Grade B
Loss to follow up: Overall (40.00%), placebo arm (33.33%), and OPC-14117 arm (46.67%)
Follow-up: 2, 4, 8, 12 weeks

Inclusion criteria: HIV infection (confirmed by ELISA or Western blot), evidence of cognitive impairment,
and stable antiretroviral regimen for six weeks
Age: Placebo (mean age = 42.9), OPC-14117 (mean age = 40.5)
Male: Placebo (80%), OPC-14117 (87%)
White: Placebo (53%), OPC-14117 (60%)
Years of HIV: Placebo (4.9), OPC-14117 (5.6)
Interventions DRUGS
(1) OPC-14117
(4) Placebo
REGIMEN:
60mg tablet of OPC-14117 orally, twice daily for the first six weeks then maintenance dosage of 120mg
twice daily

(2) Measures of safety - Adverse effects and abnormal laboratory results
(3) Measures of efficacy - change in neuropsychological test scores

Dana 1997 (Continued)

Study sponsors: The Charles A. Dana Foundation, and National Institutes of Health (NIH), and Otsuka
America Pharmaceuticals, Inc., Rockville, MD
Risk of bias
Dana 1998
Methods Design: RCT

Loss to follow up: Overall (13.39%), placebo arm (33.33%), deprenyl arm (11.11%), thioctic acid arm
(11.11%), and thioctic acid+deprenyl (none)

Inclusion criteria: HIV infection (confirmed by ELISA or Western blot), evidence of cognitive impairment,
and stable antiretroviral regimen for six weeks
Age: Placebo (mean age = 38.5), Deprenyl (mean age = 40.6), Thiotic acid (mean age = 43.7), and Both
(mean age = 42.1)
Male: Placebo (77.8%), Deprenyl (66.7%), Thiotic acid (66.7%), and Both (77.8%)
White: Placebo (44%), Deprenyl (55%), Thiotic acid (44%), and Both (78%)
Years of HIV: Placebo (4.6), Deprenyl (5.3), Thiotic acid (6.6), and Both (6.3)
Interventions DRUGS
(1) Deprenyl alone 2.5mg orally
(2) Deprenyl (2.5mg) and Thiotic acid (600mg) orally
(3) Thiotic acid alone 600mg orally
(4) Placebo
REGIMEN
Deprenyl - three times a week
Thiotic acid twice daily


Dana 1998 (Continued)

Study sponsors: Neurological AIDS Research Consortium, the Charles A. Dana Foundation, and National
Institutes of Health (NIH), and National Institute of Mental Health
Sources of support: ASTA Medica provided Thiotic acid and matching placebo
Risk of bias
Heseltine 1998
Methods Design: RCT

Technique of randomization: Stratified
Allocation sequence: Minimization (Adequate)
Allocation concealment: NR
Method of concealment: NR
Binding: Double (not described)
Loss to follow up: Overall (33.49%), placebo arm (37.74%), and peptide T arm (29.36%)
Follow-up: Monthly for six months

INCLUSION CRITERIA:
(1) HIV infection (confirmed with ELISA, Western blot, or polymerase chain reaction),
(2) Cognitive dysfunction as measured by the neuropsychiatry screening battery and judged as likely due
to HIV
(3) Agreement to use barrier methods of contraception (for men) or a reliable method of contraception
(for women)
(4) Negative results on a pregnancy test within one month study entry
(5) CD4+ lymphocyte count of 0.500x109/L or less (for UM and UCSD sites only);
(6) expected survival of at least 6 months;
(7) screening laboratory results meeting the following values: total granulocyte count, 0.750x109/L or
more; hemoglobin, more than 80 g/L; platelet count, 75x109 or more; prothrombin time, more than 70%
of the control; creatinine, 130 µmol/L or less (1.5 mg/dL); aspartate aminotransferase, less than 5 times
normal; bilirubin, 51 µmol/L (3.0 mg/dL); and (8) a normal baseline electrocardiogram or urinalysis or
abnormalities for which the patient’s condition would still be judged medically stable.
EXCLUSION CRITERIA
(1) the presence of active AIDS-defining opportunistic infections or malignant neoplasm that required
treatment at study entry or Kaposi sarcoma or another malignant neoplasm likely to require chemotherapy
during the first 6 months of the study;
(2) a requirement of more than 2 transfusions per month to achieve a hemoglobin level of more than 80
g/L;
(3) alcohol or substance dependence or abuse during the last 3 months, judged by the investigators as

Heseltine 1998 (Continued)
likely to interfere with the study;

(4) underlying serious medical problems that could complicate the interpretation of the treatment results,
including unstable diabetes mellitus, clinical ischemic disease, uncontrolled hypertension, and hepatic or
renal failure;
(5) current or recent (within the past 6 months) DSM III-R Axis I psychiatric disorder (from the Diagnostic
and Statistical Manual of Mental Disorders, Third Edition, Revised)21 or history of psychotic disorder
or bipolar mania;
(6) a history of mental retardation or learning disability;
(7) treatment with psychoactive agents within 4 weeks, or within 8 weeks for long-acting agents (eg,
fluoxetine);
(8) previous use of peptide T; and
(9) Inability to participate in NP testing or to comply with the instructions for medication administration.
Age: Placebo (mean age = 60), Peptide T (mean age = 62)
Male: Placebo (95%), Peptide T (95.4%)
White: Placebo (78%), Peptide T (86%)
Interventions DRUGS
(1) Peptide T
(2) Placebo
REGIMEN
Intranasally three times a day for 6 months
Outcomes (1) Change in Cognitive function

(2) Adverse medical events

Study date: April 1993 to March 1994
Study sponsors: NR
Risk of bias
Allocation concealment? Unclear B - Unclear
Navia (high) 1998
Methods [Refer to Navia 1998 for details; author described high dose nimodipine versus placebo]
Participants
Interventions
Outcomes
Notes

Navia (high) 1998 (Continued)
Risk of bias
Navia 1998
Methods Design: RCT

Technique of randomization: NR
Allocation sequence: NR
Binding: Double blind
Loss to follow up: Overall (28.95%), placebo arm (30.77%), low dose nimodipine arm (14.29%), and
high dose nimodipine arm (45.45%)
Participants Number: randomized (n=38), completed (n= 27)

Inclusion criteria: HIV infection and evidence of cognitive impairment
Exclusion criteria: NR
Age: Nimodipine - high dose (mean age = 40.38), Nimodipine - low dose (mean age = 41.64), Placebo
(mean age = 38.64)
Interventions (1) Nimodipine 60mg orally, five times daily

(2) Nimodipine 30mg orally, three times daily (plus two doses of placebo)
(3) Placebo orally, five times daily
Outcomes (1) Neuropsychological test scores

Not included in this review
(3) Change in painful neuropathy
(4) HIV markers in cerebrospinal fluid and serum samples

Study date: May 1992 to September 1993
Study sponsors: National Institutes of Allergy and Infectious Diseases, Miles Pharmaceuticals, inc., and
American Foundation for AIDS Research.
Risk of bias

Sacktor 2000
Methods Design: RCT

Technique of randomization: NR
Binding: Double, participants, investigators, and assesors
Loss to follow up: Overall (14.29%), placebo arm (20.00%), and selegiline arm (11.11%)
Sample size calculation done a priori: NR
Follow-up: 4, 10 weeks

Age: Placebo (mean age = 43.3), Selegiline (mean age = 41.3)
Male: Placebo (60%), Selegiline (78%)
White: Placebo (80%), Selegiline (33%)
Years of HIV: Placebo (mean = 5.1), Selegiline (mean = 6.0)
Interventions DRUGS
(1) Selegiline
(2) Placebo
REGIMEN
Transdermal patch form of Selegiline (1.0mg/cm X 15 cm2) delivering approximately 3.1 mg per 24
hours.
One patch was applied every 24 hours (rotated at different sites)
Outcomes (1) Tolerability of the drug - Treatment completers

(3) Measures of efficacy - change in neuropsychological test scores, neurologic exam results, measures of
function and mood

Study sponsors: National Institutes of Health (NIH) and Somerset Pharmaceuticals
Risk of bias

Schiffito 1999
Methods Design: RCT

Loss to follow up: Overall (10.00%), placebo arm (7.14%), and lexipafant arm (12.50%)

Inclusion criteria: HIV infection, evidence of cognitive impairment, and stable antiretroviral regimen for
six weeks
Age: Lexipafant mean age = 43.2, placebo mean age = 42.1
Male: Lexipafant (64.3%), Placebo (81.2%)
White: Lexipafant (42.9%), Placebo (43.8%)
Years of HIV: Lexipafant (mean = 6.4), Placebo (mean = 6.5)
Interventions Tablet Lexipafant - 250mg twice daily

Matching placebo


Study sponsors: Neurological AIDS Research Consortium, the Charles A. Dana Foundation, and NIH
Sources of support: Lexipafant and placebo donated by British Biotech, Inc., Annapolis, MD
Risk of bias

Schiffito 2006a
Methods Design: RCT

Technique of randomization: Blocked and stratified
Binding: Double
Loss to follow up: Overall (6.25%). placebo arm (14.29%), and valproic acid arm (none)

Age: Placebo (mean age = 46.6), Valproic acid (mean age = 43.7)
Male: Placebo (72.7%), Valproic acid (81.8%)
White: Placebo (27.7%), Valproic acid (63.6%)
Years of HIV: Placebo (10.4), Valproic acid (8.3)
Interventions (1) Valproic acid 250mg twice daily

(2) Placebo

(3) Measures of efficacy - change in neuropsychological test scores, investigator Clinical Global Impression
Not included in this review
(4) MRI indices

Multi-center study: NR
Study sponsors: National Institutes of Health (NIH)
Risk of bias

Schiffito 2007a
Methods Design: RCT

Binding: Double
Loss to follow up: Overall (16.00%), placebo arm (22.8%), and memantine arm (20.00%)
Follow-up: 4, 8, 12, 16, 20 weeks

six weeks
Age: Memantine median age = 42, placebo median age = 44
Male: Memantine (93.0%), Placebo (87.0%)
White: Memantine (72.0%), Placebo (83.0%)
Interventions Tablet Memantine - 10mg daily during the first week and escalated by 10mg weekly
Matching placebo


Study date: February 1997 to December 1999
Risk of bias

Schiffito 2007b
Methods Design: RCT

Binding: Double
Loss to follow up: Overall (15.6%), placebo arm (14.0%), low dose transdermal selegilie arm (16.7%),
and high dose transdermal selegiline (16.3%)
Sample size calculation done a priori: NR

six weeks
Age: Low dose selegiline (median age = 45), high dose selegiline (median age=43), and placebo (median
age =47)
Male: Low dose selegiline (85.7%), high dose selegiline (88.4%), and placebo (88.4%)
White: Low dose selegiline (59.5%), high dose selegiline (48.8%), and placebo (44.1%)
Interventions Selegiline transdermal system (STS) 3mg/24 hour

Selegiline transdermal system (STS) 6mg/24 hour
Matching placebo


Study date: November 2001 to December 2004
Risk of bias
Schifitto (high)2007
Methods [Refer to Schiffito 2007b for details; author described high dose transdermal selegiline versus placebo]
Participants
Interventions

Schifitto (high)2007 (Continued)
Outcomes
Notes
Risk of bias
Allocation concealment? Unclear D - Not used
*Cognitive impairment was defined as performing at or below 1 SD from the mean on at least two neuropsychological tests, or 2 SDs
below the mean on at least one test.
*Neuropsychiatry test: Rey Auditory Verbal Learning Test, Digit Symbol Test, Grooved Pegboard (dominant and non-dominant hands),
timed gait, and the California Computerized Assessment package (Cal Cap) reaction time test
* RCT - Randomized controlled trial
* NR - Not reported
* Inclusion of all randomized participants: A = >80%; B = <80%, C = unclear

Characteristics of excluded studies [ordered by study ID]
Anonymous 1996 Letter / Commentary
Brew 2007 Efficacy study of antiretroviral intervention not adjunctive therapy
Bridge 1989 Not a randomized study
Bridge 1991 Not a randomized study
Chang 2004 Validation study
Cysique 2006 Not a randomized study
Day 1992 Efficacy study of antiretroviral intervention not adjunctive therapy
DiCenzo 2004 Drug interaction study
Galgani 1997 Efficacy study of antiretroviral intervention not adjunctive therapy
Kosten 1997 Not a randomized study
Lee 2003 Validation study
Polianova 2003 Aim of the study was not to treat participants with AIDS dementia complex
Sacktor 2002 Efficacy study of antiretroviral intervention not adjunctive therapy
Schifitto 2001 Validation study
Schifitto 2006b Aim of the study was to treat participants with HIV-associated sensory neuropathy
Schmitt 1988 Efficacy study of antiretroviral intervention not adjunctive therapy
Sidtis 1993 Efficacy study of antiretroviral intervention not adjunctive therapy
Simpson 1996 Aim of the study was to treat participants with HIV-associated distal neuropathy
Tozzi 1993 Efficacy study of antiretroviral intervention not adjunctive therapy
Villemagne 1996 Not a randomized study

Characteristics of ongoing studies [ordered by study ID]
Minocycline 2006
Trial name or title Minocycline for the Treatment of Decreased Mental Function in HIV Infected Adults
Methods
Participants Ages Eligible for Study: 18 Years - 65 Years, Genders Eligible for Study: Both
Inclusion Criteria: HIV infected, AIDS Dementia Scale (ADC) Stage greater than 0, Cognitive impairment,
and willing to use acceptable methods of contraception
Exclusion Criteria: current cancers, severe premorbid psychiatric illness, including schizophrenia and major
depression, which, in the opinion of the investigator, may interfere with the study, inability to undergo lumbar
punctures and breastfeeding.
Interventions Minocycline
Matching placebo
Outcomes Primary Outcome Measures:

“Change in cognitive performance from study entry to Week 24
Secondary Outcome Measures:
”24 week change in Global Deficit Score (summary measure of neuropsychological test defects)
“24 week change of clinical global impression by the investigator
”24 week change of cognitive domain scores
“24 week change of Karnofsky performance score
”toxicity and/or signs and symptoms of Grade 2 or higher
“time from treatment initiation to the development of a Grade 2 or higher toxicity and/or signs and symptoms
”24 week changes in Instrumental Activities of Daily Living Questionnaire
Starting date August 2006
Contact information Sofia Solis ssolis@mednet.ucla.edu

Study sponsors: National Institute of Allergy and Infectious Diseases (NIAID)
Neurologic AIDS Research Consortium (NARC)
Study ID Numbers: ACTG A5235
ClinicalTrials.gov identifier NCT00361257
Nimodipine 1999
Trial name or title Randomized, Double-Blind, Placebo-Controlled Trial of Nimodipine for the Neurological Manifestations of
HIV-1
Methods

Nimodipine 1999 (Continued)
Participants Ages Eligible for Study: 18 Years and above, Genders Eligible for Study: Both
Criteria
Inclusion Criteria: documented HIV infection, HIV-Associated Motor / Cognitive Complex, acceptable
neurological and neuropsychological impairment scores, and ability to provide written informed consent.
Exclusion Criteria: Active symptomatic AIDS-defining opportunistic infection, neoplasms other than basal
cell carcinoma, in situ carcinoma of the cervix, or Kaposi’s sarcoma without evidence of visceral involvement or
that do not require systemic chemotherapy, and confounding neurological disorders, including the following.
Interventions Nimodipine
Matching placebo
Outcomes
Starting date November 1999
Contact information

Study sponsors: National Institute of Allergy and Infectious Diseases (NIAID), Miles, and Glaxo Wellcome
Study ID Numbers: ACTG 162
Last Updated: June 23, 2005
Record first received: November 2, 1999
ClinicalTrials.gov Identifier: NCT00000738

DATA AND ANALYSES
Comparison 1. Lexipafant vs Placebo
No. of No. of
Outcome or subgroup title studies participants Statistical method Effect size
1 Neuropsychological test scores 1 Mean Difference (IV, Fixed, 95% CI) Totals not selected
1.1 Rey Auditory Verbal 1 Mean Difference (IV, Fixed, 95% CI) Not estimable
Learning - Total Number
Correct
Learning - Number Correct -
Trial 5
Learning - Recall after
Interference
Learning - Delayed Recall
Learning - Correct Recognition
1.6 Mean Reaction Time - 1 Mean Difference (IV, Fixed, 95% CI) Not estimable
Choice
Sequential
1.8 Digit Symbol 1 Mean Difference (IV, Fixed, 95% CI) Not estimable
1.9 Grooved Pegboard - 1 Mean Difference (IV, Fixed, 95% CI) Not estimable
Dominant Hand
Nondominant Hand
1.11 Timed Gait 1 Mean Difference (IV, Fixed, 95% CI) Not estimable
1.12 Composite 1 Mean Difference (IV, Fixed, 95% CI) Not estimable
neuropsychological Z score
2 Tolerability 1 Risk Ratio (M-H, Fixed, 95% CI) Totals not selected
3 All cause mortality 1 Risk Ratio (M-H, Fixed, 95% CI) Totals not selected
4 Adverse effects 1 Risk Ratio (M-H, Fixed, 95% CI) Totals not selected
4.1 Number of patient 1 Risk Ratio (M-H, Fixed, 95% CI) Not estimable
experiencing any adverse effects
4.2 Diarrhea 1 Risk Ratio (M-H, Fixed, 95% CI) Not estimable
4.3 Herpes zoster 1 Risk Ratio (M-H, Fixed, 95% CI) Not estimable
4.4 Blepharitis 1 Risk Ratio (M-H, Fixed, 95% CI) Not estimable
4.5 Headache 1 Risk Ratio (M-H, Fixed, 95% CI) Not estimable
4.6 Gastroesophageal reflux 1 Risk Ratio (M-H, Fixed, 95% CI) Not estimable
4.7 Vomiting 1 Risk Ratio (M-H, Fixed, 95% CI) Not estimable
4.8 Nephroliatisis 1 Risk Ratio (M-H, Fixed, 95% CI) Not estimable
4.9 Bacteria pneumonia 1 Risk Ratio (M-H, Fixed, 95% CI) Not estimable

Comparison 2. Peptide T vs placebo
No. of No. of
1 Neurocognitive performance test 1 Mean Difference (IV, Fixed, 95% CI) Totals not selected
1.1 Global 1 Mean Difference (IV, Fixed, 95% CI) Not estimable
1.2 Verbal fluency 1 Mean Difference (IV, Fixed, 95% CI) Not estimable
1.3 Visuospatial ability 1 Mean Difference (IV, Fixed, 95% CI) Not estimable
1.4 Abstract thinking 1 Mean Difference (IV, Fixed, 95% CI) Not estimable
1.5 Speed of information 1 Mean Difference (IV, Fixed, 95% CI) Not estimable
processing
1.6 working memory 1 Mean Difference (IV, Fixed, 95% CI) Not estimable
1.7 Learning and retention 1 Mean Difference (IV, Fixed, 95% CI) Not estimable
1.8 Motor performance 1 Mean Difference (IV, Fixed, 95% CI) Not estimable
3 All cause mortality 1 Odds Ratio (M-H, Fixed, 95% CI) Totals not selected
4.2 Worsening of mood 1 Risk Ratio (M-H, Fixed, 95% CI) Not estimable
disorder
4.3 Rash 1 Risk Ratio (M-H, Fixed, 95% CI) Not estimable
4.4 Nasal congestion 1 Risk Ratio (M-H, Fixed, 95% CI) Not estimable
4.5 Proteinuria 1 Risk Ratio (M-H, Fixed, 95% CI) Not estimable
4.6 Peripheral neuropathy 1 Risk Ratio (M-H, Fixed, 95% CI) Not estimable
Comparison 3. OPC-1147 vs placebo
No. of No. of
Correct
Interference
Dominant Hand
Nondominant Hand
4.2 Fatigue 1 Risk Ratio (M-H, Fixed, 95% CI) Not estimable
4.3 Diarrhea 1 Risk Ratio (M-H, Fixed, 95% CI) Not estimable
4.4 Nausea 1 Risk Ratio (M-H, Fixed, 95% CI) Not estimable
4.5 Cough 1 Risk Ratio (M-H, Fixed, 95% CI) Not estimable
Comparison 4. Valproic acid vs Placebo
No. of No. of
Correct
Trial 5
Interference
Choice
Sequential
Dominant Hand
Nondominant Hand
3 All cause mortality 1 15 Odds Ratio (M-H, Fixed, 95% CI) Not estimable
4.2 Abnormal liver function 1 Risk Ratio (M-H, Fixed, 95% CI) Not estimable
4.3 Gastroesophageal reflux 1 Risk Ratio (M-H, Fixed, 95% CI) Not estimable
Comparison 5. Memantine versus placebo
No. of No. of
1.1 NPZ-8 score 1 Mean Difference (IV, Fixed, 95% CI) Not estimable
4.1 Grade 1 toxicity 1 Risk Ratio (M-H, Fixed, 95% CI) Not estimable
4.2 Grade 2+ toxicity 1 Risk Ratio (M-H, Fixed, 95% CI) Not estimable
Comparison 6. Thioctic acid vs Placebo
No. of No. of
Correct
Trial 5
Interference
Choice
Sequential
Dominant Hand
Nondominant Hand
2 Tolerability 1 Odds Ratio (M-H, Fixed, 95% CI) Totals not selected
Comparison 7. Oral selegiline (deprenly) vs Placebo
No. of No. of
Correct
Trial 5
Interference
Dominant Hand
Nondominant Hand
Sequential
Choice
3 All cause mortality 1 18 Odds Ratio (M-H, Fixed, 95% CI) Not estimable
4.4 Acute lymphocytic 1 Risk Ratio (M-H, Fixed, 95% CI) Not estimable
leukemia
4.5 Bacteria pneumonia 1 Risk Ratio (M-H, Fixed, 95% CI) Not estimable
4.6 Flu 1 Risk Ratio (M-H, Fixed, 95% CI) Not estimable

Comparison 8. Thiotic acid and oral selegiline (deprenyl) vs Placebo
No. of No. of
Correct
Trial 5
Interference
Choice
Sequential
Dominant Hand
Nondominant Hand
4.4 Acute lymphocytic 1 Risk Ratio (M-H, Fixed, 95% CI) Not estimable
leukemia
4.5 Neutropenia 1 Risk Ratio (M-H, Fixed, 95% CI) Not estimable

Comparison 9. Low dose Transdermal selegiline (deprenyl) vs placebo
No. of No. of
Correct
Trial 5
Interference
Learning - correct recognition
Dominant Hand
Nondominant Hand
Choice
Sequential
1.12 NPZ-6 1 Mean Difference (IV, Fixed, 95% CI) Not estimable
1.13 NPZ-8 1 Mean Difference (IV, Fixed, 95% CI) Not estimable
1.14 NPZ-total score 1 Mean Difference (IV, Fixed, 95% CI) Not estimable
2 Tolerability 2 99 Risk Ratio (M-H, Fixed, 95% CI) 0.99 [0.83, 1.17]
3 All cause mortality 2 99 Risk Ratio (M-H, Fixed, 95% CI) 1.02 [0.07, 15.84]
4.2 Itching 1 Risk Ratio (M-H, Fixed, 95% CI) Not estimable
4.3 Erythema 1 Risk Ratio (M-H, Fixed, 95% CI) Not estimable
4.4 Herpes zoster 1 Risk Ratio (M-H, Fixed, 95% CI) Not estimable
4.5 Bell’s palsy 1 Risk Ratio (M-H, Fixed, 95% CI) Not estimable
4.8 Stomach ache 1 Risk Ratio (M-H, Fixed, 95% CI) Not estimable
4.9 Facial flush 1 Risk Ratio (M-H, Fixed, 95% CI) Not estimable
4.10 Vertigo 1 Risk Ratio (M-H, Fixed, 95% CI) Not estimable
4.11 Chest pain 1 Risk Ratio (M-H, Fixed, 95% CI) Not estimable
4.12 Calf infection 1 Risk Ratio (M-H, Fixed, 95% CI) Not estimable
4.14 Back pain 1 Risk Ratio (M-H, Fixed, 95% CI) Not estimable
4.15 Parasthesia 1 Risk Ratio (M-H, Fixed, 95% CI) Not estimable
4.16 Shortness of breath 1 Risk Ratio (M-H, Fixed, 95% CI) Not estimable
Comparison 10. High dose Transdermal selegiline (deprenyl) vs placebo
No. of No. of
1.1 Neurozological Z-score 6 1 Mean Difference (IV, Fixed, 95% CI) Not estimable
1.2 NPZ - 8 1 Mean Difference (IV, Fixed, 95% CI) Not estimable
1.3 NPZ total score 1 Mean Difference (IV, Fixed, 95% CI) Not estimable
4.2 Backpain 1 Risk Ratio (M-H, Fixed, 95% CI) Not estimable
4.3 Parasthesia 1 Risk Ratio (M-H, Fixed, 95% CI) Not estimable
4.5 Shortness of breath 1 Risk Ratio (M-H, Fixed, 95% CI) Not estimable
Comparison 11. Low dose (50mg) CPI-1189 vs Placebo
No. of No. of
Correct
Trial 5
Interference
Choice
Sequential
1.7 Trail Making - Part A 1 Mean Difference (IV, Fixed, 95% CI) Not estimable
1.8 Trail Making - Part B 1 Mean Difference (IV, Fixed, 95% CI) Not estimable
Dominant Hand
Nondominant Hand
1.12 Finger Tapping - 1 Mean Difference (IV, Fixed, 95% CI) Not estimable
Dominant Hand
Nondominant Hand
4.2 Abdominal pain 1 Risk Ratio (M-H, Fixed, 95% CI) Not estimable
4.3 Diarhea 1 Risk Ratio (M-H, Fixed, 95% CI) Not estimable
4.5 Fever 1 Risk Ratio (M-H, Fixed, 95% CI) Not estimable
4.8 Upper respiratory tract 1 Risk Ratio (M-H, Fixed, 95% CI) Not estimable
infection
4.9 Weight decrease 1 Risk Ratio (M-H, Fixed, 95% CI) Not estimable
Comparison 12. High dose (100mg) CPI-1189 vs Placebo
No. of No. of
Correct
Trial 5
Interference
Choice
Sequential
Dominant Hand
Nondominant Hand
Dominant Hand
Nondominant Hand
4.2 Abdominal pain 1 Risk Ratio (M-H, Fixed, 95% CI) Not estimable
4.3 Diarhea 1 Risk Ratio (M-H, Fixed, 95% CI) Not estimable
4.5 Fever 1 Risk Ratio (M-H, Fixed, 95% CI) Not estimable
4.8 Upper respiratory tract 1 Risk Ratio (M-H, Fixed, 95% CI) Not estimable
infection
4.9 Weight decrease 1 Risk Ratio (M-H, Fixed, 95% CI) Not estimable
Comparison 13. Low dose nimodipine vs Placebo
No. of No. of
1 Neuropsychological test scores 0 0 Mean Difference (IV, Fixed, 95% CI) Not estimable
1.1 Rey Auditory Verbal 0 0 Mean Difference (IV, Fixed, 95% CI) Not estimable
Correct
Trial 5
Interference
1.5 Mean Reaction Time - 0 0 Mean Difference (IV, Fixed, 95% CI) Not estimable
Choice
1.6 Mean Reaction Time - 0 0 Mean Difference (IV, Fixed, 95% CI) Not estimable
Sequential
1.7 Trail Making - Part A 0 0 Mean Difference (IV, Fixed, 95% CI) Not estimable
1.8 Trail Making - Part B 0 0 Mean Difference (IV, Fixed, 95% CI) Not estimable
1.9 Digit Symbol 0 0 Mean Difference (IV, Fixed, 95% CI) Not estimable
1.10 Grooved Pegboard - 0 0 Mean Difference (IV, Fixed, 95% CI) Not estimable
Dominant Hand
1.11 Grooved Pegboard - 0 0 Mean Difference (IV, Fixed, 95% CI) Not estimable
Nondominant Hand
1.12 Finger Tapping - 0 0 Mean Difference (IV, Fixed, 95% CI) Not estimable
Dominant Hand
1.13 Finger Tapping - 0 0 Mean Difference (IV, Fixed, 95% CI) Not estimable
Nondominant Hand
1.14 Timed Gait 0 0 Mean Difference (IV, Fixed, 95% CI) Not estimable
1.15 Composite 0 0 Mean Difference (IV, Fixed, 95% CI) Not estimable
4.2 Haematological toxicity 1 Risk Ratio (M-H, Fixed, 95% CI) Not estimable
4.3 Cardiac toxicity 1 Risk Ratio (M-H, Fixed, 95% CI) Not estimable
4.4 Psychiatrics toxicity 1 Risk Ratio (M-H, Fixed, 95% CI) Not estimable
4.5 Hypotension 1 Risk Ratio (M-H, Fixed, 95% CI) Not estimable
Comparison 14. High dose nimodipine vs Placebo
No. of No. of
Correct
Trial 5
Interference
Choice
Sequential
Dominant Hand
Nondominant Hand
Dominant Hand
Nondominant Hand
2 Tolerability 1 Odds Ratio (M-H, Fixed, 95% CI) Totals not selected
9.2 Haematological toxicity 1 Risk Ratio (M-H, Fixed, 95% CI) Not estimable
9.3 Cardiac toxicity 1 Risk Ratio (M-H, Fixed, 95% CI) Not estimable
9.4 Psychiatrics toxicity 1 Risk Ratio (M-H, Fixed, 95% CI) Not estimable
9.5 Hypotension 1 Risk Ratio (M-H, Fixed, 95% CI) Not estimable
Comparison 15. Any adjunctive therapy vs placebo
No. of No. of
3 Number of patient experiencing 14 Risk Ratio (M-H, Fixed, 95% CI) Totals not selected
any adverse effects

Analysis 1.1. Comparison 1 Lexipafant vs Placebo, Outcome 1 Neuropsychological test scores.
Review: Adjunctive therapies for AIDS dementia complex
Comparison: 1 Lexipafant vs Placebo
Outcome: 1 Neuropsychological test scores
Study or subgroup Lexipafant Placebo Mean Difference Mean Difference

N Mean(SD) N Mean(SD) IV,Fixed,95% CI IV,Fixed,95% CI
1 Rey Auditory Verbal Learning - Total Number Correct

Schiffito 1999 16 4.9 (5.9) 14 0.8 (6.9) 4.10 [ -0.53, 8.73 ]
2 Rey Auditory Verbal Learning - Number Correct Trial 5

Schiffito 1999 16 1.2 (1.6) 14 0.4 (1.7) 0.80 [ -0.39, 1.99 ]
3 Rey Auditory Verbal Learning - Recall after Interference

Schiffito 1999 16 1.9 (1.9) 14 -0.3 (2.3) 2.20 [ 0.68, 3.72 ]
4 Rey Auditory Verbal Learning - Delayed Recall

Schiffito 1999 16 1.6 (1.8) 14 0.2 (3) 1.40 [ -0.40, 3.20 ]
5 Rey Auditory Verbal Learning - Correct Recognition

Schiffito 1999 16 -0.1 (1.4) 14 0.1 (2.4) -0.20 [ -1.63, 1.23 ]
6 Mean Reaction Time - Choice

Schiffito 1999 16 -6.5 (117.4) 14 -1.4 (149.9) -5.10 [ -102.44, 92.24 ]
7 Mean Reaction Time - Sequential

Schiffito 1999 16 30.8 (85.1) 14 19.1 (137.4) 11.70 [ -71.48, 94.88 ]
8 Digit Symbol
Schiffito 1999 16 4.2 (5.2) 14 2.9 (6.3) 1.30 [ -2.87, 5.47 ]
9 Grooved Pegboard - Dominant Hand

Schiffito 1999 16 8.6 (16.2) 14 5.1 (16.9) 3.50 [ -8.39, 15.39 ]
10 Grooved Pegboard - Nondominant Hand

Schiffito 1999 16 3.5 (11.7) 14 1.7 (18.2) 1.80 [ -9.32, 12.92 ]
11 Timed Gait
Schiffito 1999 16 0.5 (2.2) 14 0.5 (2.2) 0.0 [ -1.58, 1.58 ]
12 Composite neuropsychological Z score

Schiffito 1999 16 0 (0) 14 0 (0) 0.0 [ 0.0, 0.0 ]
-10 -5 0 5 10
Favours Placebo Favours lexipafant

Analysis 1.2. Comparison 1 Lexipafant vs Placebo, Outcome 2 Tolerability.
Outcome: 2 Tolerability
Study or subgroup Lexipafant Placebo Risk Ratio Risk Ratio

n/N n/N M-H,Fixed,95% CI M-H,Fixed,95% CI
Schiffito 1999 14/16 13/14 0.94 [ 0.74, 1.19 ]
0.5 0.7 1 1.5 2

Favours lexipafant Favours placebo
Analysis 1.3. Comparison 1 Lexipafant vs Placebo, Outcome 3 All cause mortality.
Outcome: 3 All cause mortality

Schiffito 1999 0/16 0/14 0.0 [ 0.0, 0.0 ]
0.001 0.01 0.1 1 10 100 1000


Analysis 1.4. Comparison 1 Lexipafant vs Placebo, Outcome 4 Adverse effects.
Outcome: 4 Adverse effects

1 Number of patient experiencing any adverse effects

Schiffito 1999 8/16 12/14 0.58 [ 0.34, 1.00 ]
2 Diarrhea
Schiffito 1999 0/16 1/14 0.29 [ 0.01, 6.69 ]
3 Herpes zoster
Schiffito 1999 0/16 1/14 0.29 [ 0.01, 6.69 ]
4 Blepharitis
Schiffito 1999 1/16 0/14 2.65 [ 0.12, 60.21 ]
5 Headache
Schiffito 1999 2/16 3/14 0.58 [ 0.11, 3.00 ]
6 Gastroesophageal reflux
Schiffito 1999 2/16 3/14 0.58 [ 0.11, 3.00 ]
7 Vomiting
Schiffito 1999 0/16 2/14 0.18 [ 0.01, 3.39 ]
8 Nephroliatisis
Schiffito 1999 1/16 1/14 0.88 [ 0.06, 12.73 ]
9 Bacteria pneumonia
Schiffito 1999 1/16 1/14 0.88 [ 0.06, 12.73 ]
0.001 0.01 0.1 1 10 100 1000


Analysis 2.1. Comparison 2 Peptide T vs placebo, Outcome 1 Neurocognitive performance test.
Comparison: 2 Peptide T vs placebo
Outcome: 1 Neurocognitive performance test
Study or subgroup Peptide T Placebo Mean Difference Mean Difference

1 Global
Heseltine 1998 66 0.24 (0.41) 77 0.16 (0.26) 0.08 [ -0.03, 0.19 ]
2 Verbal fluency
Heseltine 1998 66 0.14 (0.73) 77 0.15 (0.7) -0.01 [ -0.25, 0.23 ]
3 Visuospatial ability
Heseltine 1998 66 0.17 (0.48) 77 0.25 (0.52) -0.08 [ -0.24, 0.08 ]
4 Abstract thinking
Heseltine 1998 66 0.34 (0.56) 77 0.23 (0.35) 0.11 [ -0.05, 0.27 ]
5 Speed of information processing

Heseltine 1998 66 0.23 (0.57) 77 0.14 (0.44) 0.09 [ -0.08, 0.26 ]
6 working memory
Heseltine 1998 66 0.23 (0.73) 77 0.08 (0.61) 0.15 [ -0.07, 0.37 ]
7 Learning and retention

Heseltine 1998 66 0.19 (0.57) 77 0.11 (0.53) 0.08 [ -0.10, 0.26 ]
8 Motor performance
Heseltine 1998 66 0.19 (0.49) 77 0.18 (0.53) 0.01 [ -0.16, 0.18 ]
-0.5 -0.25 0 0.25 0.5

Favours placebo Favours peptide T
Analysis 2.2. Comparison 2 Peptide T vs placebo, Outcome 2 Tolerability.
Study or subgroup Peptide T Placebo Risk Ratio Risk Ratio

Heseltine 1998 66/106 77/109 0.88 [ 0.73, 1.07 ]
0.5 0.7 1 1.5 2

Favours peptide T Favours placebo

Analysis 2.3. Comparison 2 Peptide T vs placebo, Outcome 3 All cause mortality.
Study or subgroup Peptide T Placebo Odds Ratio Odds Ratio

Heseltine 1998 8/106 3/109 2.88 [ 0.74, 11.18 ]
0.01 0.1 1 10 100

Analysis 2.4. Comparison 2 Peptide T vs placebo, Outcome 4 Adverse effects.
Study or subgroup Peptide T Placebo Risk Ratio Risk Ratio


Heseltine 1998 79/106 79/109 1.03 [ 0.88, 1.21 ]
2 Worsening of mood disorder

Heseltine 1998 7/106 1/109 7.20 [ 0.90, 57.51 ]
3 Rash
Heseltine 1998 34/106 42/109 0.83 [ 0.58, 1.20 ]
4 Nasal congestion
Heseltine 1998 12/106 5/109 2.47 [ 0.90, 6.77 ]
5 Proteinuria
Heseltine 1998 18/106 9/109 2.06 [ 0.97, 4.37 ]
6 Peripheral neuropathy
Heseltine 1998 8/106 22/109 0.37 [ 0.17, 0.80 ]
0.001 0.01 0.1 1 10 100 1000


Analysis 3.1. Comparison 3 OPC-1147 vs placebo, Outcome 1 Neuropsychological test scores.
Comparison: 3 OPC-1147 vs placebo
Study or subgroup OPC-1147 Placebo Mean Difference Mean Difference


Dana 1997 15 0.3 (9.6) 15 -2 (9.9) 2.30 [ -4.68, 9.28 ]

Dana 1997 15 0.8 (3.5) 15 -0.9 (2) 1.70 [ -0.34, 3.74 ]

Dana 1997 15 0.6 (3.4) 15 -0.9 (2) 1.50 [ -0.50, 3.50 ]
9 Digit Symbol
Dana 1997 15 0.1 (4.1) 15 -0.3 (6.4) 0.40 [ -3.45, 4.25 ]

Dana 1997 15 -3.3 (13.7) 15 4 (30.6) -7.30 [ -24.27, 9.67 ]

Dana 1997 15 -5 (20.8) 15 -0.8 (2.6) -4.20 [ -14.81, 6.41 ]
14 Timed Gait
Dana 1997 15 0.3 (2.1) 15 -0.8 (2.6) 1.10 [ -0.59, 2.79 ]

Dana 1997 15 0 (0) 15 0 (0) 0.0 [ 0.0, 0.0 ]
-10 -5 0 5 10
Favours Placebo Favours OPC-1147

Analysis 3.2. Comparison 3 OPC-1147 vs placebo, Outcome 2 Tolerability.
Study or subgroup OPC-1147 Placebo Risk Ratio Risk Ratio

Dana 1997 7/15 9/15 0.78 [ 0.39, 1.54 ]
0.1 0.2 0.5 1 2 5 10

Favours OPC-1147 Favours placebo
Analysis 3.3. Comparison 3 OPC-1147 vs placebo, Outcome 3 All cause mortality.

Dana 1997 0/15 0/15 0.0 [ 0.0, 0.0 ]
0.1 0.2 0.5 1 2 5 10


Analysis 3.4. Comparison 3 OPC-1147 vs placebo, Outcome 4 Adverse effects.


Dana 1997 7/15 13/15 0.54 [ 0.30, 0.96 ]
2 Fatigue
Dana 1997 3/15 5/15 0.60 [ 0.17, 2.07 ]
3 Diarrhea
Dana 1997 3/15 2/15 1.50 [ 0.29, 7.73 ]
4 Nausea
Dana 1997 1/15 3/15 0.33 [ 0.04, 2.85 ]
5 Cough
Dana 1997 0/15 3/15 0.14 [ 0.01, 2.55 ]
0.01 0.1 1 10 100


Analysis 4.1. Comparison 4 Valproic acid vs Placebo, Outcome 1 Neuropsychological test scores.
Comparison: 4 Valproic acid vs Placebo
Study or subgroup Valproic acid Placebo Mean Difference Mean Difference


Schiffito 2006a 9 2.89 (6.83) 6 0.67 (7.34) 2.22 [ -5.16, 9.60 ]

Schiffito 2006a 9 0.78 (2.05) 6 1.17 (1.83) -0.39 [ -2.37, 1.59 ]

Schiffito 2006a 9 -0.22 (2.91) 6 -5 (10.37) 4.78 [ -3.73, 13.29 ]

Schiffito 2006a 9 1.11 (0.78) 6 -0.67 (2.58) 1.78 [ -0.35, 3.91 ]

Schiffito 2006a 9 0.11 (4.14) 6 0.83 (5.08) -0.72 [ -5.60, 4.16 ]

Schiffito 2006a 9 -43.38 (107.57) 6 -24 (135.98) -19.38 [ -148.91, 110.15 ]

Schiffito 2006a 9 -8.63 (74.79) 6 -64.67 (123.47) 56.04 [ -54.18, 166.26 ]
8 Digit Symbol
Schiffito 2006a 9 6.56 (5.83) 6 2 (7.32) 4.56 [ -2.43, 11.55 ]

Schiffito 2006a 9 -9.33 (17.63) 6 1 (15.59) -10.33 [ -27.31, 6.65 ]

Schiffito 2006a 9 -6.44 (17.57) 6 -2.5 (15.81) -3.94 [ -21.02, 13.14 ]
11 Timed Gait
Schiffito 2006a 9 -0.65 (1.09) 6 -0.26 (0.79) -0.39 [ -1.34, 0.56 ]

Schiffito 2006a 9 4.56 (6.22) 6 2.01 (3.9) 2.55 [ -2.57, 7.67 ]
-10 -5 0 5 10
Placebo Valproic acid

Analysis 4.2. Comparison 4 Valproic acid vs Placebo, Outcome 2 Tolerability.
Study or subgroup Valproic acid Placebo Risk Ratio Risk Ratio

Schiffito 2006a 8/9 5/6 1.07 [ 0.70, 1.63 ]
0.01 0.1 1 10 100

Valproic acid Placebo
Analysis 4.3. Comparison 4 Valproic acid vs Placebo, Outcome 3 All cause mortality.
Study or subgroup Valproic acid Control Odds Ratio Odds Ratio

Schiffito 2006a 0/9 0/6 0.0 [ 0.0, 0.0 ]
Total (95% CI) 9 6 0.0 [ 0.0, 0.0 ]

Total events: 0 (Valproic acid), 0 (Control)
Heterogeneity: not applicable
Test for overall effect: Z = 0.0 (P < 0.00001)
0.1 0.2 0.5 1 2 5 10


Analysis 4.4. Comparison 4 Valproic acid vs Placebo, Outcome 4 Adverse effects.
Study or subgroup Valproic acid Placebo Risk Ratio Risk Ratio


Schiffito 2006a 2/9 0/6 3.50 [ 0.20, 62.27 ]
2 Abnormal liver function

Schiffito 2006a 1/9 0/6 2.10 [ 0.10, 44.40 ]
3 Gastroesophageal reflux
Schiffito 2006a 1/9 0/6 2.10 [ 0.10, 44.40 ]
0.001 0.01 0.1 1 10 100 1000

Analysis 5.1. Comparison 5 Memantine versus placebo, Outcome 1 Neuropsychological test scores.
Comparison: 5 Memantine versus placebo
Study or subgroup Memantine Placebo Mean Difference Mean Difference

1 NPZ-8 score
Schiffito 2007a 70 60 (210) 70 17 (86) 43.00 [ -10.16, 96.16 ]
-100 -50 0 50 100

Favours memantine Favours placebo

Analysis 5.2. Comparison 5 Memantine versus placebo, Outcome 2 Tolerability.
Study or subgroup Memantine Placebo Risk Ratio Risk Ratio

Schiffito 2007a 54/70 56/70 0.96 [ 0.81, 1.15 ]
0.5 0.7 1 1.5 2

Analysis 5.3. Comparison 5 Memantine versus placebo, Outcome 3 All cause mortality.

Schiffito 2007a 0/70 2/70 0.20 [ 0.01, 4.09 ]
0.01 0.1 1 10 100


Analysis 5.4. Comparison 5 Memantine versus placebo, Outcome 4 Adverse effects.

1 Grade 1 toxicity
Schiffito 2007a 9/70 0/70 19.00 [ 1.13, 320.27 ]
2 Grade 2+ toxicity
Schiffito 2007a 24/70 20/70 1.20 [ 0.73, 1.96 ]
0.01 0.1 1 10 100

Analysis 6.1. Comparison 6 Thioctic acid vs Placebo, Outcome 1 Neuropsychological test scores.
Comparison: 6 Thioctic acid vs Placebo
Study or subgroup Thioctic acid Placebo Mean Difference Mean Difference


Dana (thioctic) 1998 9 -6 (5.6) 9 5 (5.1) -11.00 [ -15.95, -6.05 ]

Dana (thioctic) 1998 9 -0.7 (2.2) 9 -0.1 (2) -0.60 [ -2.54, 1.34 ]

Dana (thioctic) 1998 9 -1.2 (2) 9 0.4 (2.3) -1.60 [ -3.59, 0.39 ]

Dana (thioctic) 1998 9 -1.2 (2.1) 9 0 (1.2) -1.20 [ -2.78, 0.38 ]

Dana (thioctic) 1998 9 -0.6 (0.7) 9 0 (1.9) -0.60 [ -1.92, 0.72 ]

Dana (thioctic) 1998 9 -9.1 (46.1) 9 -8.6 (66.9) -0.50 [ -53.58, 52.58 ]
-100 -50 0 50 100

Placebo Thioctic acid
(Continued . . . )

(. . . Continued)
Study or subgroup Thioctic acid Placebo Mean Difference Mean Difference
Dana (thioctic) 1998 9 63.6 (61.7) 9 -7.1 (65) 70.70 [ 12.15, 129.25 ]
8 Digit Symbol
Dana (thioctic) 1998 9 2 (9.9) 9 2.3 (5.3) -0.30 [ -7.64, 7.04 ]

Dana (thioctic) 1998 9 5.4 (17.8) 9 4.7 (14.4) 0.70 [ -14.26, 15.66 ]

Dana (thioctic) 1998 9 2.8 (12.9) 9 -10.5 (18.5) 13.30 [ -1.43, 28.03 ]

Dana (thioctic) 1998 9 0 (0) 9 0 (0) 0.0 [ 0.0, 0.0 ]
-100 -50 0 50 100

Placebo Thioctic acid
Analysis 6.2. Comparison 6 Thioctic acid vs Placebo, Outcome 2 Tolerability.
Study or subgroup Thioctic acid Placebo Odds Ratio Odds Ratio

Dana (thioctic) 1998 8/9 6/9 4.00 [ 0.33, 48.66 ]
0.01 0.1 1 10 100

Thioctic acid Placebo

Analysis 6.3. Comparison 6 Thioctic acid vs Placebo, Outcome 3 All cause mortality.
Study or subgroup Thioctic acid Placebo Odds Ratio Odds Ratio

Dana (thioctic) 1998 0/9 0/9 0.0 [ 0.0, 0.0 ]
0.1 0.2 0.5 1 2 5 10

Analysis 6.4. Comparison 6 Thioctic acid vs Placebo, Outcome 4 Adverse effects.
Study or subgroup Thioctic acid Placebo Risk Ratio Risk Ratio

1 Headache
Dana (thioctic) 1998 1/9 1/9 1.00 [ 0.07, 13.64 ]
0.001 0.01 0.1 1 10 100 1000


Analysis 7.1. Comparison 7 Oral selegiline (deprenly) vs Placebo, Outcome 1 Neuropsychological test
scores.
Comparison: 7 Oral selegiline (deprenly) vs Placebo
Study or subgroup Deprenyl Placebo Mean Difference Mean Difference


Dana 1998 9 7.8 (5.5) 9 5 (5.1) 2.80 [ -2.10, 7.70 ]

Dana 1998 9 1.7 (1.3) 9 -0.1 (2) 1.80 [ 0.24, 3.36 ]

Dana 1998 9 2.3 (1.6) 9 0.4 (2.3) 1.90 [ 0.07, 3.73 ]

Dana 1998 9 0.7 (2.5) 9 0 (1.2) 0.70 [ -1.11, 2.51 ]

Dana 1998 9 1.3 (1.8) 9 0 (1.9) 1.30 [ -0.41, 3.01 ]
6 Digit Symbol
Dana 1998 9 8.2 (7.4) 9 2.3 (5.3) 5.90 [ -0.05, 11.85 ]

Dana 1998 9 -5.3 (12.3) 9 4.7 (14.4) -10.00 [ -22.37, 2.37 ]

Dana 1998 9 -7.4 (15.8) 9 -10.5 (18.5) 3.10 [ -12.79, 18.99 ]

Dana 1998 9 17.3 (49.2) 9 -7.1 (65) 24.40 [ -28.86, 77.66 ]

Dana 1998 9 11.4 (46.2) 9 -8.6 (66.9) 20.00 [ -33.12, 73.12 ]

Dana 1998 9 0 (0) 9 0 (0) 0.0 [ 0.0, 0.0 ]
-100 -50 0 50 100

Favours Placebo Favours deprenyl

Analysis 7.2. Comparison 7 Oral selegiline (deprenly) vs Placebo, Outcome 2 Tolerability.
Study or subgroup Deprenyl Placebo Risk Ratio Risk Ratio

Dana 1998 8/9 6/9 1.33 [ 0.80, 2.23 ]
0.2 0.5 1 2 5
Favours deprenyl Favours placebo
Analysis 7.3. Comparison 7 Oral selegiline (deprenly) vs Placebo, Outcome 3 All cause mortality.
Study or subgroup Deprenyl Control Odds Ratio Odds Ratio

Dana 1998 0/9 0/9 0.0 [ 0.0, 0.0 ]
Total (95% CI) 9 9 0.0 [ 0.0, 0.0 ]

Total events: 0 (Deprenyl), 0 (Control)
Heterogeneity: not applicable
Test for overall effect: Z = 0.0 (P < 0.00001)
0.1 0.2 0.5 1 2 5 10


Analysis 7.4. Comparison 7 Oral selegiline (deprenly) vs Placebo, Outcome 4 Adverse effects.
Study or subgroup Deprenyl Placebo Risk Ratio Risk Ratio


Dana 1998 8/9 1/9 8.00 [ 1.24, 51.51 ]
2 Headache
Dana 1998 2/9 1/9 2.00 [ 0.22, 18.33 ]
3 Nausea
Dana 1998 2/9 0/9 5.00 [ 0.27, 91.52 ]
4 Acute lymphocytic leukemia

Dana 1998 2/9 0/9 5.00 [ 0.27, 91.52 ]
5 Bacteria pneumonia
Dana 1998 1/9 0/9 3.00 [ 0.14, 65.16 ]
6 Flu
Dana 1998 1/9 0/9 3.00 [ 0.14, 65.16 ]
0.001 0.01 0.1 1 10 100 1000


Analysis 8.1. Comparison 8 Thiotic acid and oral selegiline (deprenyl) vs Placebo, Outcome 1
Neuropsychological test scores.
Comparison: 8 Thiotic acid and oral selegiline (deprenyl) vs Placebo
Study or subgroup Thioctic+deprenyl Placebo Mean Difference Mean Difference


Dana (both) 1998 9 2.4 (6.4) 9 5 (5.1) -2.60 [ -7.95, 2.75 ]

Dana (both) 1998 9 0.9 (1.3) 9 -0.1 (2) 1.00 [ -0.56, 2.56 ]

Dana (both) 1998 9 -0.6 (3.4) 9 0.4 (2.3) -1.00 [ -3.68, 1.68 ]

Dana (both) 1998 9 0.1 (1.9) 9 0 (1.2) 0.10 [ -1.37, 1.57 ]

Dana (both) 1998 9 -0.1 (1.7) 9 0 (1.9) -0.10 [ -1.77, 1.57 ]

Dana (both) 1998 9 -12.9 (49.6) 9 -8.6 (66.9) -4.30 [ -58.71, 50.11 ]

Dana (both) 1998 9 -12.3 (19.7) 9 -7.1 (65) -5.20 [ -49.57, 39.17 ]

Dana (both) 1998 9 -6.1 (19.9) 9 4.7 (14.4) -10.80 [ -26.85, 5.25 ]

Dana (both) 1998 9 -5.8 (18.8) 9 -10.5 (18.5) 4.70 [ -12.53, 21.93 ]
10 Digit Symbol
Dana (both) 1998 9 2.9 (6.3) 9 2.3 (5.3) 0.60 [ -4.78, 5.98 ]

Dana (both) 1998 9 0 (0) 9 0 (0) 0.0 [ 0.0, 0.0 ]
-100 -50 0 50 100

Placebo Thioctic+deprenyl

Analysis 8.2. Comparison 8 Thiotic acid and oral selegiline (deprenyl) vs Placebo, Outcome 2 Tolerability.
Study or subgroup Thioctic+deprenyl Placebo Risk Ratio Risk Ratio

Dana (both) 1998 9/9 6/9 1.46 [ 0.91, 2.35 ]
0.2 0.5 1 2 5
Thioctic+deprenyl Placebo
Analysis 8.3. Comparison 8 Thiotic acid and oral selegiline (deprenyl) vs Placebo, Outcome 3 All cause
mortality.
Study or subgroup Thioctic+deprenyl Placebo Odds Ratio Odds Ratio

Dana (both) 1998 0/9 0/9 0.0 [ 0.0, 0.0 ]
0.1 0.2 0.5 1 2 5 10

Favours treatment Favours control

Analysis 8.4. Comparison 8 Thiotic acid and oral selegiline (deprenyl) vs Placebo, Outcome 4 Adverse
effects.
Study or subgroup Thioctic+deprenyl Placebo Risk Ratio Risk Ratio


Dana (both) 1998 6/9 1/9 6.00 [ 0.89, 40.31 ]
2 Headache
Dana (both) 1998 2/9 1/9 2.00 [ 0.22, 18.33 ]
3 Nausea
Dana (both) 1998 2/9 0/9 5.00 [ 0.27, 91.52 ]
4 Acute lymphocytic leukemia

Dana (both) 1998 1/9 0/9 3.00 [ 0.14, 65.16 ]
5 Neutropenia
Dana (both) 1998 1/9 0/9 3.00 [ 0.14, 65.16 ]
0.01 0.1 1 10 100

Thioctic+deprenyl Placebo

Analysis 9.1. Comparison 9 Low dose Transdermal selegiline (deprenyl) vs placebo, Outcome 1
Comparison: 9 Low dose Transdermal selegiline (deprenyl) vs placebo
Study or subgroup Transdermal deprenyl Placebo Mean Difference Mean Difference


Sacktor 2000 9 9 (4.6) 5 6.3 (5.7) 2.70 [ -3.13, 8.53 ]

Sacktor 2000 9 1.2 (1.3) 5 2.5 (1.9) -1.30 [ -3.17, 0.57 ]

Sacktor 2000 9 1.9 (1.7) 5 1.3 (2.8) 0.60 [ -2.09, 3.29 ]

Sacktor 2000 9 2.6 (1.7) 5 -0.3 (2.9) 2.90 [ 0.13, 5.67 ]
5 Rey Auditory Verbal Learning - correct recognition

Sacktor 2000 9 1.8 (2.4) 5 1.3 (2.1) 0.50 [ -1.92, 2.92 ]
6 Digit Symbol
Sacktor 2000 9 3.6 (7.8) 5 1 (8.1) 2.60 [ -6.14, 11.34 ]

Sacktor 2000 9 9.4 (20.5) 5 -9.3 (34.1) 18.70 [ -14.05, 51.45 ]

Sacktor 2000 9 13.2 (26.4) 5 -10.3 (10.1) 23.50 [ 4.11, 42.89 ]

Sacktor 2000 9 -11.7 (50.1) 5 86.8 (51.1) -98.50 [ -153.98, -43.02 ]

Sacktor 2000 9 -25.6 (133.9) 5 6.3 (76.1) -31.90 [ -141.91, 78.11 ]
11 Timed Gait
Sacktor 2000 9 0.8 (1) 5 0.2 (1.3) 0.60 [ -0.71, 1.91 ]
12 NPZ-6
Schiffito 2007b 43 0.16 (0.71) 42 0.19 (0.57) -0.03 [ -0.30, 0.24 ]
13 NPZ-8
Schiffito 2007b 43 0.34 (0.6) 42 0.35 (0.34) -0.01 [ -0.22, 0.20 ]
14 NPZ-total score
Schiffito 2007b 43 0.24 (0.52) 42 0.18 (0.55) 0.06 [ -0.17, 0.29 ]
-100 -50 0 50 100

Placebo Transdermal deprenyl

Analysis 9.2. Comparison 9 Low dose Transdermal selegiline (deprenyl) vs placebo, Outcome 2 Tolerability.
Study or subgroup Transdermal deprenyl Placebo Risk Ratio Weight Risk Ratio
Sacktor 2000 8/9 4/5 12.3 % 1.11 [ 0.68, 1.82 ]
Schiffito 2007b 35/42 37/43 87.7 % 0.97 [ 0.81, 1.16 ]
Total (95% CI) 51 48 100.0 % 0.99 [ 0.83, 1.17 ]

Total events: 43 (Transdermal deprenyl), 41 (Placebo)
Heterogeneity: Chi2 = 0.26, df = 1 (P = 0.61); I2 =0.0%
Test for overall effect: Z = 0.16 (P = 0.87)
0.2 0.5 1 2 5
Transdermal deprenyl Placebo
Analysis 9.3. Comparison 9 Low dose Transdermal selegiline (deprenyl) vs placebo, Outcome 3 All cause
mortality.
Study or subgroup Transdermal deprenyl Placebo Risk Ratio Risk Ratio

Sacktor 2000 0/9 0/5 0.0 [ 0.0, 0.0 ]
Schiffito 2007b 1/42 1/43 1.02 [ 0.07, 15.84 ]
Total (95% CI) 51 48 1.02 [ 0.07, 15.84 ]

Total events: 1 (Transdermal deprenyl), 1 (Placebo)
Heterogeneity: Chi2 = 0.0, df = 0 (P = 1.00); I2 =0.0%
Test for overall effect: Z = 0.02 (P = 0.99)
0.001 0.01 0.1 1 10 100 1000

Transdermal deprenyl Placebo

Analysis 9.4. Comparison 9 Low dose Transdermal selegiline (deprenyl) vs placebo, Outcome 4 Adverse
effects.


2 Itching
Sacktor 2000 0/9 1/5 0.20 [ 0.01, 4.17 ]
3 Erythema
Sacktor 2000 1/9 2/5 0.28 [ 0.03, 2.35 ]
4 Herpes zoster
Sacktor 2000 1/9 1/5 0.56 [ 0.04, 7.09 ]
5 Bell’s palsy
Sacktor 2000 0/9 1/5 0.20 [ 0.01, 4.17 ]
6 Headache
Sacktor 2000 0/9 1/5 0.20 [ 0.01, 4.17 ]
Schiffito 2007b 0/42 2/43 0.20 [ 0.01, 4.14 ]
7 Nausea
Sacktor 2000 2/9 0/5 3.00 [ 0.17, 52.53 ]
8 Stomach ache
Sacktor 2000 2/9 0/5 3.00 [ 0.17, 52.53 ]
9 Facial flush
Sacktor 2000 1/9 0/5 1.80 [ 0.09, 37.49 ]
10 Vertigo
Sacktor 2000 1/9 0/5 1.80 [ 0.09, 37.49 ]
11 Chest pain
Sacktor 2000 1/9 0/5 1.80 [ 0.09, 37.49 ]
12 Calf infection
Schiffito 2007b 1/42 0/43 3.07 [ 0.13, 73.30 ]
13 Fatigue
Schiffito 2007b 1/42 0/43 3.07 [ 0.13, 73.30 ]
14 Back pain
Schiffito 2007b 0/42 1/43 0.34 [ 0.01, 8.14 ]
15 Parasthesia
0.001 0.01 0.1 1 10 100 1000

Favours treatment Favours placebo
(Continued . . . )

(. . . Continued)
Schiffito 2007b 0/42 1/43 0.34 [ 0.01, 8.14 ]
16 Shortness of breath
Schiffito 2007b 0/42 1/43 0.34 [ 0.01, 8.14 ]
0.001 0.01 0.1 1 10 100 1000

Analysis 10.1. Comparison 10 High dose Transdermal selegiline (deprenyl) vs placebo, Outcome 1
Comparison: 10 High dose Transdermal selegiline (deprenyl) vs placebo
Study or subgroup Selegiline Placebo Mean Difference Mean Difference

1 Neurozological Z-score 6
Schifitto (high)2007 28 0.23 (0.61) 34 0.19 (0.57) 0.04 [ -0.26, 0.34 ]
2 NPZ - 8
Schifitto (high)2007 24 0.14 (0.46) 28 0.35 (0.34) -0.21 [ -0.43, 0.01 ]
3 NPZ total score

Schifitto (high)2007 33 0.2 (0.66) 41 0.18 (0.55) 0.02 [ -0.26, 0.30 ]
-1 -0.5 0 0.5 1

Analysis 10.2. Comparison 10 High dose Transdermal selegiline (deprenyl) vs placebo, Outcome 2
Tolerability.
Study or subgroup Selegiline Placebo Risk Ratio Risk Ratio

Schifitto (high)2007 36/43 37/43 0.97 [ 0.81, 1.16 ]
0.5 0.7 1 1.5 2

Favours selegiline Favours placebo
Analysis 10.3. Comparison 10 High dose Transdermal selegiline (deprenyl) vs placebo, Outcome 3 All cause
mortality.

Schifitto (high)2007 1/43 1/43 1.00 [ 0.06, 15.48 ]
0.5 0.7 1 1.5 2

Favours selegiline Favours placebo

Analysis 10.4. Comparison 10 High dose Transdermal selegiline (deprenyl) vs placebo, Outcome 4 Adverse
effects.

1 Rash
Schifitto (high)2007 1/43 0/43 3.00 [ 0.13, 71.65 ]
2 Backpain
Schifitto (high)2007 0/43 1/43 0.33 [ 0.01, 7.96 ]
3 Parasthesia
Schifitto (high)2007 1/43 0/43 3.00 [ 0.13, 71.65 ]
4 Headache
Schifitto (high)2007 0/43 2/43 0.20 [ 0.01, 4.05 ]
5 Shortness of breath
Schifitto (high)2007 0/43 1/43 0.33 [ 0.01, 7.96 ]
0.001 0.01 0.1 1 10 100 1000


Analysis 11.1. Comparison 11 Low dose (50mg) CPI-1189 vs Placebo, Outcome 1 Neuropsychological test
scores.
Comparison: 11 Low dose (50mg) CPI-1189 vs Placebo
Study or subgroup CPI-1189 Placebo Mean Difference Mean Difference


Clifford 2002 21 -1.8 (8.4) 21 0.8 (9.4) -2.60 [ -7.99, 2.79 ]

Clifford 2002 21 -0.8 (2.2) 21 0.6 (2.4) -1.40 [ -2.79, -0.01 ]

Clifford 2002 21 -0.8 (2.2) 21 0.7 (3.6) -1.50 [ -3.30, 0.30 ]

Clifford 2002 21 -0.6 (3.4) 21 0.2 (3.6) -0.80 [ -2.92, 1.32 ]

Clifford 2002 21 11.6 (83.9) 21 -10.4 (68.1) 22.00 [ -24.22, 68.22 ]

Clifford 2002 21 48.9 (88.1) 21 51.3 (77.1) -2.40 [ -52.47, 47.67 ]
7 Trail Making - Part A

Clifford 2002 21 0.1 (0.1) 21 0.1 (0.2) 0.0 [ -0.10, 0.10 ]
8 Trail Making - Part B

Clifford 2002 21 0.1 (0.2) 21 0.1 (0.1) 0.0 [ -0.10, 0.10 ]
9 Digit Symbol
Clifford 2002 21 5 (6.5) 21 7.6 (12.1) -2.60 [ -8.47, 3.27 ]

Clifford 2002 21 3.3 (21.7) 21 5.7 (15.7) -2.40 [ -13.86, 9.06 ]

Clifford 2002 21 6.1 (11.8) 21 0.7 (20.1) 5.40 [ -4.57, 15.37 ]
12 Finger Tapping - Dominant Hand

Clifford 2002 21 0.5 (5.9) 21 2.3 (8.7) -1.80 [ -6.30, 2.70 ]
13 Finger Tapping - Nondominant Hand

Clifford 2002 21 1.5 (5) 21 3.1 (7.5) -1.60 [ -5.46, 2.26 ]
14 Timed Gait
Clifford 2002 21 0.6 (1.4) 21 0.2 (1.7) 0.40 [ -0.54, 1.34 ]

Clifford 2002 21 0.6 (0.9) 21 0.3 (0.6) 0.30 [ -0.16, 0.76 ]
-10 -5 0 5 10
Favours Placebo Favours CPI-1189

Analysis 11.2. Comparison 11 Low dose (50mg) CPI-1189 vs Placebo, Outcome 2 Tolerability.
Study or subgroup CPI-1189 Placebo Risk Ratio Risk Ratio

Clifford 2002 19/21 16/21 1.19 [ 0.90, 1.57 ]
0.2 0.5 1 2 5
Favours CPI-1189 Favours control
Analysis 11.3. Comparison 11 Low dose (50mg) CPI-1189 vs Placebo, Outcome 3 All cause mortality.

Clifford 2002 0/21 1/21 0.33 [ 0.01, 7.74 ]
0.01 0.1 1 10 100


Analysis 11.4. Comparison 11 Low dose (50mg) CPI-1189 vs Placebo, Outcome 4 Adverse effects.


Clifford 2002 12/21 20/21 0.60 [ 0.41, 0.88 ]
2 Abdominal pain
Clifford 2002 1/21 2/21 0.50 [ 0.05, 5.10 ]
3 Diarhea
Clifford 2002 2/21 3/21 0.67 [ 0.12, 3.59 ]
4 Fatigue
Clifford 2002 0/21 1/21 0.33 [ 0.01, 7.74 ]
5 Fever
Clifford 2002 1/21 1/21 1.00 [ 0.07, 14.95 ]
6 Headache
Clifford 2002 2/21 2/21 1.00 [ 0.16, 6.45 ]
7 Rash
Clifford 2002 3/21 5/21 0.60 [ 0.16, 2.20 ]
8 Upper respiratory tract infection

Clifford 2002 1/21 2/21 0.50 [ 0.05, 5.10 ]
9 Weight decrease
Clifford 2002 0/21 2/21 0.20 [ 0.01, 3.93 ]
10 Nausea
Clifford 2002 2/21 2/21 1.00 [ 0.16, 6.45 ]
0.01 0.1 1 10 100

Favours CPI-1189 Favours placebo

Analysis 12.1. Comparison 12 High dose (100mg) CPI-1189 vs Placebo, Outcome 1 Neuropsychological test
scores.
Comparison: 12 High dose (100mg) CPI-1189 vs Placebo
Study or subgroup CPI-1189 Placebo Mean Difference Mean Difference


Clifford (high) 2002 22 1.4 (8.7) 21 0.8 (9.4) 0.60 [ -4.82, 6.02 ]

Clifford (high) 2002 22 0.6 (0.8) 21 0.6 (2.4) 0.0 [ -1.08, 1.08 ]

Clifford (high) 2002 22 0.8 (3) 21 0.7 (3.6) 0.10 [ -1.89, 2.09 ]

Clifford (high) 2002 22 0.7 (2.8) 21 0.2 (3.6) 0.50 [ -1.43, 2.43 ]

Clifford (high) 2002 22 13.8 (83.9) 21 -10.4 (68.1) 24.20 [ -21.38, 69.78 ]

Clifford (high) 2002 22 26.2 (111.1) 21 51.3 (77.1) -25.10 [ -82.04, 31.84 ]
7 Trail Making - Part A

Clifford (high) 2002 22 0.1 (0.1) 21 0.1 (0.2) 0.0 [ -0.10, 0.10 ]
8 Trail Making - Part B

Clifford (high) 2002 22 0.1 (0.2) 21 0.1 (0.1) 0.0 [ -0.09, 0.09 ]
9 Digit Symbol
Clifford (high) 2002 22 6.6 (6) 21 7.6 (12.1) -1.00 [ -6.75, 4.75 ]

Clifford (high) 2002 22 8.3 (16.7) 21 5.7 (15.7) 2.60 [ -7.08, 12.28 ]

Clifford (high) 2002 22 12.2 (14.5) 21 0.7 (20.1) 11.50 [ 0.98, 22.02 ]
12 Finger Tapping - Dominant Hand

Clifford (high) 2002 22 -0.5 (7.2) 21 2.3 (8.7) -2.80 [ -7.59, 1.99 ]
13 Finger Tapping - Nondominant Hand

Clifford (high) 2002 22 1 (5.3) 21 3.1 (7.5) -2.10 [ -6.00, 1.80 ]
14 Timed Gait
Clifford (high) 2002 22 0.8 (2.3) 21 0.2 (1.7) 0.60 [ -0.61, 1.81 ]

Clifford (high) 2002 22 0.8 (0.8) 21 0.3 (0.6) 0.50 [ 0.08, 0.92 ]
-100 -50 0 50 100

Favours Placebo Favours CPI-1189

Analysis 12.2. Comparison 12 High dose (100mg) CPI-1189 vs Placebo, Outcome 2 Tolerability.

Clifford (high) 2002 20/22 16/21 1.19 [ 0.91, 1.57 ]
0.2 0.5 1 2 5
Analysis 12.3. Comparison 12 High dose (100mg) CPI-1189 vs Placebo, Outcome 3 All cause mortality.

Clifford (high) 2002 0/22 1/21 0.32 [ 0.01, 7.42 ]
0.01 0.1 1 10 100


Analysis 12.4. Comparison 12 High dose (100mg) CPI-1189 vs Placebo, Outcome 4 Adverse effects.


Clifford (high) 2002 20/22 20/21 0.95 [ 0.81, 1.12 ]
2 Abdominal pain
Clifford (high) 2002 0/22 2/21 0.19 [ 0.01, 3.76 ]
3 Diarhea
Clifford (high) 2002 3/22 3/21 0.95 [ 0.22, 4.21 ]
4 Fatigue
Clifford (high) 2002 2/22 1/21 1.91 [ 0.19, 19.52 ]
5 Fever
Clifford (high) 2002 1/22 1/21 0.95 [ 0.06, 14.30 ]
6 Headache
Clifford (high) 2002 4/22 2/21 1.91 [ 0.39, 9.35 ]
7 Rash
Clifford (high) 2002 3/22 5/21 0.57 [ 0.16, 2.10 ]
8 Upper respiratory tract infection

Clifford (high) 2002 3/22 2/21 1.43 [ 0.27, 7.73 ]
9 Weight decrease
Clifford (high) 2002 1/22 2/21 0.48 [ 0.05, 4.88 ]
10 Nausea
Clifford (high) 2002 3/22 2/21 1.43 [ 0.27, 7.73 ]
0.01 0.1 1 10 100


Analysis 13.2. Comparison 13 Low dose nimodipine vs Placebo, Outcome 2 Tolerability.
Comparison: 13 Low dose nimodipine vs Placebo
Study or subgroup Nimodipine Placebo Risk Ratio Risk Ratio

Navia 1998 12/14 6/11 1.57 [ 0.88, 2.81 ]
0.2 0.5 1 2 5
Favours nimodipine Favours placebo
Analysis 13.3. Comparison 13 Low dose nimodipine vs Placebo, Outcome 3 All cause mortality.

Navia 1998 1/14 1/11 0.79 [ 0.06, 11.20 ]
0.01 0.1 1 10 100

Favours nimodipine Favours control

Analysis 13.4. Comparison 13 Low dose nimodipine vs Placebo, Outcome 4 Adverse effects.


Navia 1998 4/14 6/11 0.52 [ 0.19, 1.41 ]
2 Haematological toxicity
Navia 1998 2/14 3/11 0.52 [ 0.11, 2.61 ]
3 Cardiac toxicity
Navia 1998 1/14 1/11 0.79 [ 0.06, 11.20 ]
4 Psychiatrics toxicity
Navia 1998 1/14 1/11 0.79 [ 0.06, 11.20 ]
5 Hypotension
Navia 1998 0/14 1/11 0.27 [ 0.01, 5.97 ]
0.01 0.1 1 10 100

Analysis 14.2. Comparison 14 High dose nimodipine vs Placebo, Outcome 2 Tolerability.
Comparison: 14 High dose nimodipine vs Placebo
Study or subgroup Nimopidine Placebo Odds Ratio Odds Ratio

Navia (high) 1998 9/13 6/11 1.88 [ 0.35, 9.98 ]
0.1 0.2 0.5 1 2 5 10

Favours nimopidine Favours control

Analysis 14.3. Comparison 14 High dose nimodipine vs Placebo, Outcome 3 All cause mortality.

Navia (high) 1998 1/13 1/11 0.85 [ 0.06, 12.01 ]
0.01 0.1 1 10 100

Favours nimopidine Favours control
Analysis 14.9. Comparison 14 High dose nimodipine vs Placebo, Outcome 9 Adverse effects.
Study or subgroup Nimodipine (high) Placebo Risk Ratio Risk Ratio


Navia (high) 1998 5/13 6/11 0.71 [ 0.29, 1.69 ]
2 Haematological toxicity
Navia (high) 1998 4/13 3/11 1.13 [ 0.32, 3.99 ]
3 Cardiac toxicity
Navia (high) 1998 0/13 1/11 0.29 [ 0.01, 6.38 ]
4 Psychiatrics toxicity
Navia (high) 1998 1/13 1/11 0.85 [ 0.06, 12.01 ]
5 Hypotension
Navia (high) 1998 0/13 1/11 0.29 [ 0.01, 6.38 ]
0.01 0.1 1 10 100


Analysis 15.1. Comparison 15 Any adjunctive therapy vs placebo, Outcome 1 Tolerability.
Comparison: 15 Any adjunctive therapy vs placebo
Study or subgroup Adjunctive therapy Placebo Risk Ratio Risk Ratio

Clifford (high) 2002 20/22 16/21 1.19 [ 0.91, 1.57 ]
Clifford 2002 19/21 16/21 1.19 [ 0.90, 1.57 ]
Dana (both) 1998 9/9 6/9 1.46 [ 0.91, 2.35 ]
Dana (thioctic) 1998 8/9 6/9 1.33 [ 0.80, 2.23 ]
Dana 1997 7/15 9/15 0.78 [ 0.39, 1.54 ]
Dana 1998 8/9 6/9 1.33 [ 0.80, 2.23 ]
Heseltine 1998 66/106 77/109 0.88 [ 0.73, 1.07 ]
Navia (high) 1998 9/13 6/11 1.27 [ 0.66, 2.43 ]
Navia 1998 12/14 6/11 1.57 [ 0.88, 2.81 ]
Sacktor 2000 8/9 4/5 1.11 [ 0.68, 1.82 ]
Schiffito 1999 14/16 13/14 0.94 [ 0.74, 1.19 ]
Schiffito 2006a 8/9 5/6 1.07 [ 0.70, 1.63 ]
Schiffito 2007a 54/70 56/70 0.96 [ 0.81, 1.15 ]
Schiffito 2007b 35/42 37/43 0.97 [ 0.81, 1.16 ]
Schifitto (high)2007 36/43 37/43 0.97 [ 0.81, 1.16 ]
0.2 0.5 1 2 5
Adjunctive therapy Placebo

Analysis 15.2. Comparison 15 Any adjunctive therapy vs placebo, Outcome 2 All cause mortality.

Clifford (high) 2002 0/22 1/21 0.32 [ 0.01, 7.42 ]
Clifford 2002 0/21 1/21 0.33 [ 0.01, 7.74 ]
Dana (both) 1998 0/9 0/9 0.0 [ 0.0, 0.0 ]
Dana (thioctic) 1998 0/9 0/9 0.0 [ 0.0, 0.0 ]
Dana 1997 0/15 0/15 0.0 [ 0.0, 0.0 ]
Dana 1998 0/9 0/9 0.0 [ 0.0, 0.0 ]
Heseltine 1998 8/106 3/109 2.74 [ 0.75, 10.06 ]
Navia (high) 1998 1/13 1/11 0.85 [ 0.06, 12.01 ]
Navia 1998 1/14 1/11 0.79 [ 0.06, 11.20 ]
Sacktor 2000 0/9 0/5 0.0 [ 0.0, 0.0 ]
Schiffito 1999 0/16 0/14 0.0 [ 0.0, 0.0 ]
Schiffito 2006a 0/9 0/6 0.0 [ 0.0, 0.0 ]
Schiffito 2007a 0/70 2/70 0.20 [ 0.01, 4.09 ]
Schiffito 2007b 1/42 1/42 1.00 [ 0.06, 15.47 ]
Schifitto (high)2007 0/43 1/43 0.33 [ 0.01, 7.96 ]
0.001 0.01 0.1 1 10 100 1000


Analysis 15.3. Comparison 15 Any adjunctive therapy vs placebo, Outcome 3 Number of patient
experiencing any adverse effects.
Outcome: 3 Number of patient experiencing any adverse effects

Clifford (high) 2002 20/22 20/21 0.95 [ 0.81, 1.12 ]
Clifford 2002 12/21 20/21 0.60 [ 0.41, 0.88 ]
Dana (both) 1998 6/9 1/9 6.00 [ 0.89, 40.31 ]
Dana (thioctic) 1998 1/9 1/9 1.00 [ 0.07, 13.64 ]
Dana 1997 7/15 13/15 0.54 [ 0.30, 0.96 ]
Dana 1998 8/9 1/9 8.00 [ 1.24, 51.51 ]
Heseltine 1998 79/106 79/109 1.03 [ 0.88, 1.21 ]
Navia (high) 1998 5/13 6/11 0.71 [ 0.29, 1.69 ]
Navia 1998 4/14 6/11 0.52 [ 0.19, 1.41 ]
Schiffito 1999 8/16 12/14 0.58 [ 0.34, 1.00 ]
Schiffito 2006a 2/9 0/6 3.50 [ 0.20, 62.27 ]
Schiffito 2007a 33/70 20/70 1.65 [ 1.06, 2.58 ]
Schiffito 2007b 2/42 5/43 0.41 [ 0.08, 2.00 ]
Schifitto (high)2007 1/43 5/43 0.20 [ 0.02, 1.64 ]
0.01 0.1 1 10 100

WHAT’S NEW
Last assessed as up-to-date: 11 February 2007.
14 May 2008 Amended Finished; converted to new review format.

HISTORY
Protocol first published: Issue 2, 2007
Review first published: Issue 3, 2008
12 September 2006 Feedback has been incorporated Substantive amendment
CONTRIBUTIONS OF AUTHORS
Olalekan Uthman contributed to the protocol design, conducted the literature search, selected studies for inclusion, located copies of
study reports, extracted data, and did the data entry and analysis; and has overall responsibility for maintaining the review.
Abdulmalik Jibril selected studies for inclusion, extracted data and assisted with data entry and analysis.
All authors contributed to the writing of the report.
DECLARATIONS OF INTEREST
We declare that we have no affiliation with or involvement in any organization or entity with a financial interest in the subject matter
of the review; for example, employment, consultancy, stock ownership, honoria or expert testimony.
SOURCES OF SUPPORT
Internal sources
• Save the Youth Initiative, Nigeria.
External sources
• Reviews for Africa Programme Fellowship, South Africa.

• South African Cochrane Centre (SACC) HIV/AIDS Mentoring Program, South Africa.
INDEX TERMS

Medical Subject Headings (MeSH)
AIDS Dementia Complex [∗ therapy]; HIV-1; Randomized Controlled Trials as Topic
MeSH check words

Humans


Adjunctive Therapies For AIDS Dementia Complex Cochrane

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Adjunctive Therapies For AIDS Dementia Complex Cochrane

Uploaded by

Copyright:

Available Formats

Adjunctive therapies for AIDS dementia complex (Review)

Uthman OA, Abdulmalik JO

Adjunctive therapies for AIDS dementia complex (Review)

Adjunctive therapies for AIDS dementia complex (Review) ii

Adjunctive therapies for AIDS dementia complex

Olalekan A Uthman1 , Jubril O Abdulmalik2

Editorial group: Cochrane HIV/AIDS Group.

PLAIN LANGUAGE SUMMARY

Adjunctive therapies for AIDS dementia complex (Review) 2

Figure 2. HIV-1 Neuroinvasion.Ghafouri 2006

Adjunctive therapies for AIDS dementia complex (Review) 3

Table 1. Staging of AIDS dementia complex

Stage Grading Presentation

0.5 Subclinical or Equivocal - Minimal or equivocal symptoms

1 Mild - Unequivocal intellectual or motor impairment

Adjunctive therapies for AIDS dementia complex (Review) 4

2 Moderate - Cannot work or perform demanding ADL

3 Severe - Major intellectual disability or

4 End-stage - Nearly vegetative

Table 2. Clinical manifestations of ADC

Type of impairment Manifestations

Affective - Apathy (depression-like feature)

Behavioral - Psychomotor retardation (Slowed speech or response time)

Cognitive - Lack of visuospatial memory (misplacing things)

Motor - Unsteady gait, loss of balance

Adjunctive therapies for AIDS dementia complex (Review) 5

Table 3. Differential diagnosis of symptoms presenting as possible ADC

CNS disease Manifestations

Central nervous system disease Infectious

Systemic/metabolic/endocrine disease - B12 deficiency

Primary psychiatric illness - Mood disorders (major depression, hypomania, dysthymia)

Substance withdrawal or intoxification - Alcohol

Adjunctive therapies for AIDS dementia complex (Review) 6

Medications Medication reactions

Attention has focused on adjunctive therapies targeted at reduc-

Although there are general reviews on neurological complications

Table 4. Adjunctive therapies for ADC

CPI-1189 TNF-alpha antagonist

Lexipafant Platelet activating factor antagonist

Memantine N-methyl-D-aspartate antagonist

Minocycline Anti-inflammatory and p38 MAP kinase inhibitor

Nimodipine Calcium Channel blocker

OPC 14117 Antioxidant

Pentoxifylline Platelet activating factor antagonist

Peptide T Possibly Chemokine receptor blockade

Prednisone Macrophage suppression

Adjunctive therapies for AIDS dementia complex (Review) 7

Selegiline (deprenyl) Monoamine oxidase-B inhibitor

Thioctic acid Antioxidant

Valproic acid Unknown

Any intervention other than antiretroviral therapy used for the

Table 5. Search Strategy

#1 HIV Infections[MeSH] OR HIV[MeSH] OR hiv[tw] OR hiv-1*[tw] OR hiv-2*[tw] OR hiv1[tw] OR hiv2[tw] OR

Adjunctive therapies for AIDS dementia complex (Review) 8

acquired immunodeficiency syndrome[tw] OR acquired immunedeficiency syndrome[tw] OR acquired immuno-deficiency

#2 “AIDS DEMENTIA COMPLEX”[MeSH ] OR AIDS DEMENTIA COMPLEX[tw]

#5 ADJUNCTIVE THERAPY OR ADJUNCTIVE THERAPIES OR CPI-1189 OR LEXIPAFANT OR MEMANTINE

#7 #6 Field: All Fields, Limits: from 1980 to 2006, HUMAN

• MEDLINE (1980 to February 2007) • International AIDS Conferences (1985 to 2004)

Conference proceedings • Eleventh International Conference on AIDS and STDs in

Adjunctive therapies for AIDS dementia complex (Review) 9

Adjunctive therapies for AIDS dementia complex (Review) 10

Adjunctive therapies for AIDS dementia complex (Review) 11

#1 HIV Infections[MeSH] OR HIV[MeSH] OR hiv[tw] OR hiv-1[tw] OR hiv-2[tw] OR hiv1[tw] OR hiv2[tw] OR

Trial Sequence Conceal- Blinding Follow-up Outcome Outcome *Others

Genera- Allocation Complete- Incomplete Free of selec- Free of topic