INT J TUBERC LUNG DIS 6(10):903–908

© 2002 IUATLD
The value of urine testing for verifying adherence to
anti-tuberculosis chemotherapy in children and adults
in Uganda
P. E. Meissner,* P. Musoke,

A. Okwera,

J. E. G. Bunn,* J. B. S. Coulter*
* Liverpool School of Tropical Medicine, Pembroke Place, Liverpool, UK;

Department of Paediatrics & Child Health,
S U MMA R Y
Makerere University, Kampala,

National Tuberculosis Treatment Centre, Mulago Hospital, Kampala, Uganda
SETTI NG: Mulago Hospital, Kampala, Uganda.
OBJ ECTI VE: To evaluate the usefulness of urine dip-
sticks for monitoring adherence to anti-tuberculosis
chemotherapy.
DESI GN: In-house urine dipsticks for detection of iso-
niazid (INH) metabolites were compared to commercial
test strips. The value of n-butanol to detect rifampicin
was compared to coloration of the urine. Non-adherence
was assessed through a questionnaire and reviews of the
Mulago Hospital TB register.
RESULTS: Urine was obtained from 236 patients (127
adults and 109 children) on daily chemotherapy. Using
commercial test strips as standard, the sensitivity of in-
house urine dipsticks was 99.5% and specificity was
96.4%. The sensitivity and the specificity of n-butanol
and of coloration of urine to detect rifampicin were low
(64.0% and 54.9%, and 85.5% and 64.8%, respec-
tively). Fifty patients (21.2%) admitted non-adherence
to treatment during the previous month. An additional
15 (6.8%) were detected through urine testing. Of 911
patients in the TB register of Mulago Hospital who had
started treatment in the first 3 months of 2000, 39.7%
did not complete their treatment. Two-thirds of these
had discontinued treatment in the first 2 months.
CONCLUSI ON: In-house INH test strips are as effective
as commercially available strips for detecting isoniazid
in the urine. They are a simple tool for monitoring
adherence. Adherence to anti-tuberculosis chemother-
apy as determined by the use of isoniazid test strips and
review of the TB register showed poor compliance. Tests
for rifampicin are less sensitive and specific.
KEY WORDS: compliance; tuberculosis; urine testing;
isoniazid; paper test strips
THERE ARE 8 million new cases and 1.9 million
deaths due to tuberculosis (TB) annually world-wide.
1
The greatest burden of TB is concentrated in develop-
ing countries. Although DOTS is implemented in many
countries, TB case rates continue to increase, especially
in countries with a high prevalence of human immuno-
deficiency virus (HIV) infection.
2
Treatment adherence is a particular problem for
TB, because cure requires the use of medication for at
least 6 months. According to the US Centers for Dis-
ease Control and Prevention, approximately one-
third of all patients in the USA with active tuberculosis
fail to complete treatment or preventive therapy.
3
Non-
adherence may reach 90% among patients who are
homeless or alcoholics. Few studies have looked at
treatment adherence of patients in developing coun-
tries, and even less work has been done to assess com-
pliance among children.
Physicians’ predictions of non-adherence to treat-
ment are accurate in fewer than 50% of cases, and
unfortunately there is no gold standard by which to
measure adherence to medication.
4
Although simple
methods for detection of isoniazid (INH) in the urine
have been developed over the last decades,
5–7
com-
mercial test strips have been used in only a few devel-
oping countries.
8
In-house test strips have not been
described.
The aim of the study was to evaluate the usefulness
of urine tests in a developing country for monitoring
the drug treatment of paediatric and adult patients with
tuberculosis and assessing adherence to chemotherapy.
METHODS
Recruitment of patients for urine testing
Over a 7-week period between April and May 2001,
236 in- and out-patients (109 children and 127
adults) were recruited from the National Tuberculosis
Treatment Centre at Mulago Hospital, Kampala,
Uganda. All in-patients during this period were
Correspondence to: Dr Peter Meissner, Dep. Gen. Paediatrics, Im Neuenheimer Feld 150, 69120 Heidelberg, Germany. Fax:
(ϩ49) 89-2443-57981. e-mail: peter.meissner@urz.uni-heidelberg.de
Article submitted 4 January 2002. Final version accepted 4 June 2002.
904 The International Journal of Tuberculosis and Lung Disease
included, as were every out-patient child and every
third adult out-patient waiting to see the doctor or to
collect new drugs. Only patients and children whose
guardians had given verbal consent were enrolled.
Compliance
Patients or guardians were asked a standard set of
questions. The questionnaire provided demographic
data and information pertaining to patients’ under-
standing of TB, and their perceptions of their own
ability to adhere to prescribed medication.
Patients were defined to be non-adherent if they
had taken their last dose Ͼ32 hours previously or if
they admitted that they had missed one or more doses
of anti-tuberculosis drugs in the last month. Those who
tested negative with the commercial urine test strips or
were lost to follow-up according to the TB register of
Mulago Hospital were also considered non-adherent.
Detection of isoniazid
To detect INH metabolites, the Arkansas Method was
used,
7
which includes chloramine-T, potassium cya-
nide and barbituric acid; the latter results in a change
to a clear blue colour in positive samples. Paper test
strips were developed similar to the description of
Kilburn.
6
The 6 cm ϫ 8 mm test strips were cut from
0.83 mm thick filter paper (Grade 222, Ahlstrom
Paper Group, Mt. Holly Springs, PA, USA). Each test
strip was spotted with 25 ␮l 5% barbituric acid in
water (pH adjusted to 5.2 with 50% sodium hydrox-
ide), 60% potassium thiocyanate in 8% citric acid,
and 50% aqueous chloramine T in three bands (Figure).
To permit solubilisation the chloramine T and barbi-
turic acid were heated while stirring and then pipetted
(using a multistep pipette) on to the paper strips while
still hot. The test strips were air-dried and stored.
To detect INH, the test strip was placed in 0.5 ml
of urine. For full colour development, the reaction
was performed in a screw-capped test tube (13 ϫ 100
mm). As the urine is absorbed up the paper strip, cyan-
ogen chloride is produced by mixing the potassium
thiocyanate with the chloramine T. The pyridine ring
of the INH metabolites isonicotinic acid and isonico-
tinoyl glycine is split by cyanogen chloride to form a
glutaconaldehyde derivative. The latter then con-
denses with barbituric acid to form a dark blue poly-
methine dye. A negative urine sample remains yellow.
For full colour development the test strips should
remain in the urine for at least 15 min.
The in-house test strips were compared to the
commercially available Taxo INH urine test strips
®
(Becton Dickinson, Cowley, UK), which follow the
same principle.
Detection of rifampicin
Coloration of the urine was recorded and compared
to a photographic scale. For the n-butanol method,
1 ml n-butanol was added to 5 ml of urine and the tube
was inverted.
9
In the presence of rifampicin an orange-
red colour appears in the upper butanol layer after 5–
10 minutes.
Review of the TB register at Mulago Hospital
Records of all patients who had commenced TB chemo-
therapy in the last first 3 months of 2000 were ana-
lysed for completion of therapy.
Ethical approval
The study received ethical approval from the Liverpool
School of Tropical Medicine, and Mulago Hospital.
RESULTS
In-house test strips compared to
Taxo INH urine test strips
®
In 234 of 236 cases the home-made test strips pro-
duced identical results compared to the Taxo INH
urine test strips
®
. In both tests 27 patients were nega-
tive and 207 samples were positive for isoniazid. In
one case the commercial test strips were negative and
the in-house test strips were slightly positive, and in
another case it was the reverse. The sensitivity of the
in-house test strips compared to commercial strips as
a gold standard was therefore 99.5% and the specific-
ity 96.4% (Table 1). Of 18 patients who admitted
that they had taken their last dose Ͼ32 hours ago
(i.e., they had missed at least one dose), six tested
weakly positive, with both the commercial and the in-
house INH-test strips. All six patients had taken their
last dose of INH between 33 and 60 hours prior to
testing (Table 1).
Rifampicin test and urine coloration
A total of 174 patients were presumed to have taken
rifampicin and INH on the day of testing, 59 patients
received INH and ethambutol and another three
patients only INH prophylaxis. Thus, the latter were
included in the rifampicin study as negative controls
(Table 2). To detect defaulters among the patients
who claimed that they had taken rifampicin, the Figure In-house urine dipsticks.
Control of TB treatment through urine testing 905
results were compared with the results for the com-
mercial INH test strips, because all patients were tak-
ing combination tablets containing rifampicin and
INH (Table 2).
The n-butanol test to detect rifampicin had a sen-
sitivity of only 64.0% and a specificity of 85.5%. The
sensitivity and specificity of interpreting the urine
colour by eye was even lower (54.9% and 62.3%,
respectively). Many patients had taken their drugs the
previous afternoon or evening and not in the morning
before their meal as recommended. Only 65 (29.0%)
of all tested patients reported having taken their last
drugs Ͻ12 hours prior to testing.
Adherence to treatment
Adherence was analysed for 127 adults (90 out- and
37 in-patients) and guardians of 109 children (80 out-
and 29 in-patients). Forty-nine (20.7%) of all inter-
viewed patients admitted that they had missed at least
one dose of anti-tuberculosis drugs during the last
month; of these, 27 (11.4%) missed more than one dose.
Fifteen (6.8%) patients were detected on urine testing to
be non-adherent on the day of testing. Six of them
admitted some non-adherence in the last month. They
all tested negative in both tests for INH and rifampicin.
Twenty-nine (31.0%) of the paediatric in-patients
who should have received TB treatment did not take
their drugs, compared to 37 (5.4%) on the adult
wards (P ϭ 0.007). There was a higher rate of adult
TB patients (28.3%) who were lost to follow-up dur-
ing the study period compared to paediatric patients
(22.3%) (P ϭ 0.02, OR 1.41, 95% confidence inter-
val [CI] 1.04–1.9).
There was no significant difference in age, sex,
level of education, ability to speak English or family
status of adult patients or the guardians of children
between the adherent and non-adherent groups.
Although 230 (97.5%) patients had a good knowl-
edge of the drug regimen, 35 patients (14.8%) said
that they had initially received no explanation from a
doctor.
According to the TB register about 52 patients of
the study population (18%) were lost to follow-up
during the study period.
Of 911 patients who started treatment for TB
between January and March 2000, 39.7% did not
complete their treatment (excluding patients who had
died or continued treatment elsewhere); of these, two-
thirds discontinued treatment in the first 2 months.
DISCUSSION
Value of urine testing
Improving adherence is essential for effective tuber-
culosis control, and a urine test for isoniazid is a use-
ful tool to detect non-adherence. The chemical
method for detection of INH metabolites in the urine
is well described in the literature and known to be
highly sensitive and specific.
10
However, the use of
non-commercial paper test strips, which were first
described by Kilburn et al. in 1972, has not been
reported in the literature since then.
6
Commercially
available test strips are unaffordable for most devel-
oping countries. The sensitivity and specificity of the
in-house paper test strips were identical to those of
the commercially available ones. However, for pa-
tients who last took their drugs the day before (Ͼ24 h
previously), the colour change in both methods
was usually initially weak and became positive after
Ͼ15 min.
Over 200 in-house test strips can be produced
within 3–4 hours at a cost of less than 1.5 US cents
Table 1 Sensitivity/specificity of the in-house test strips compared to the Taxo
INH urine test strips
®
Patients had
taken
isoniazid
Patients had
not taken
isoniazid
Results of in-house
test strips,
n ϭ 236
Positive 207 False positives
1 Sensitivity ϭ 99.5%
Negative False negatives 27
1 Specificity ϭ 96.4%
Table 2 Sensitivity/specificity of the n-butanol test for rifampicin*
Patients had
taken
rifampicin
Patients had
not taken
rifampicin
Results of
rifampicin test,
n ϭ 233
Positive 96 False positives
12 Sensitivity ϭ 64.0%
Negative False negatives 71
54 Specificity ϭ 85.5%
*Compared to INH test results (if patients were taking combination tablets) or admitted non-adherence.
906 The International Journal of Tuberculosis and Lung Disease
per test strip, and the method can be easily taught to
local laboratory technicians. Chemicals and chro-
matographic filter paper should be available in most
developing countries. Any paper with a high absorp-
tion and capillary rise Ͼ52 mm (e.g., blotting paper)
is suitable. To date only the study by Kilburn et al.
6
and the present study have assessed the sensitivity and
specificity of in-house paper test strips compared to
commercial ones. Further studies in conjunction with
DOTS, where the exact time of taking chemotherapy
is known, would be valuable in assessing the sensitiv-
ity of dipsticks and the measurement of urine concen-
tration of (acetyl) isoniazid.
Isoniazid, which is a component of every TB regi-
men, is, due to its pharmacokinetics, an ideal target
for the documentation of adherence to treatment.
INH metabolites are found in the urine of all patients
for at least 24 hours.
10
It may be argued that with the
few patients where the test was positive 48 hours or
more after taking INH, the urine positive rate would
overestimate the proportion of those patients con-
suming INH daily. However, in this study, all patients
who missed more than two doses of INH tested neg-
ative with the dipstick method.
Coloration of the urine or use of n-butanol for
detection of rifampicin can not be recommended for
control of compliance because of their low sensitivity
and specificity. The lack of sensitivity is mainly caused
by patients who took the drug Ͼ12 hours prior to the
test, due to the rapid excretion of rifampicin. Only 65
(29.0%) of all tested patients reported taking the last
drugs Ͻ12 hours previously.
By avoiding the use of chemicals such as potassium
cyanide outside the laboratory, dipsticks are very attrac-
tive for developing countries. In-house test strips to
detect INH provide health professionals involved in
DOTS with a simple, objective, laboratory-independent
and cheap method to assess adherence at community
health level after discharge from hospital. Although the
ideal use of this method is in conjunction with DOTS,
supervised therapy is costly and is still not generally
available. Floyd showed that only 2% of the cost for TB
treatment is for drugs, and Ͼ90% of the costs are due to
the diagnosis and the time patients spend at the hospital
being observed taking their treatment.
11
Where DOTS is
difficult to implement, occasional urine testing com-
bined with health education will support unsupervised
therapy and by itself have a positive impact on adher-
ence and thus on treatment outcomes. The INH test
strips are especially useful to test adherence where there
is doubt in the patient’s history or no response to treat-
ment. Urine testing together with analysis of adherence
predictors may also be useful in specific settings such as
homeless shelters or in prisons.
Adherence to treatment
This study describes adherence to treatment of paedi-
atric TB patients in a developing country for the first
time. The higher non-adherence rate of paediatric in-
patients (31.0%) compared to adult in-patients (5.4%)
at Mulago Hospital is mainly due to a rigid drug dis-
pensing system. Most paediatric patients were not
treated on the adult TB wards, where TB drugs are
available at all times. TB drugs were only given out by
the TB pharmacy on Wednesdays and Thursdays,
thus up to a week could elapse between the prescrip-
tion and obtaining the drugs. Parents had not only
less access to drugs, but they also had less informa-
tion about the drugs (Table 3). The study has helped
by rectifying this anomaly. Conversely, the non-
adherence rate of adult in-patients (5.4%) was low,
despite the fact that DOTS is not yet routine for in-
patients, because TB drugs were regularly available
on the adult TB wards (Table 3).
Important causes of poor adherence include lack
of access to health education, lack of financial
resources to support medical treatment or mistrust in
the health care system.
4
Socio-economic factors such
as crowded households seem to have more consistent
relationships with poor adherence.
12
Among different
studies which describe complex sets of predictor vari-
ables, the strongest predictor of non-adherence over
the full course of therapy is significantly associated
with non-adherence during the first month of treat-
ment.
4,12
A review of the TB register showed that of
patients who did not complete therapy, over two-
thirds discontinued treatment in the first 2 months.
Interruption of treatment is serious, because of the
risk of drug resistance and the related costs of treating
relapse. Control mechanisms to improve adherence of
TB patients to chemotherapy must focus especially on
the first months of treatment to prevent relapse and
development of resistance. Presently, primary drug
resistance rates in Uganda are as follows: INH 8%,
rifampicin 0.5%, streptomycin 4%; pyrazinamide
and ethambutol resistance are zero.
Beside DOTS, several other strategies to improve
adherence to treatment have been described. A study
of South African adult in-patients showed that adher-
ence may range from 66% to 84% in different hospi-
tals, depending on the degree of medical supervision
and staff motivation.
9
The importance of intense
supervision has also been emphasised.
13
Another
study from South Africa showed that the assumption
that DOTS is successfully practised in hospital is not
always correct.
14
Adherence to treatment of adult in-
patients was as low as 62%, but it could be increased
to over 90% with simple inexpensive measures such
as ‘dosage cards’ which have to be signed by both
patient and health worker. Especially in resource poor
countries, adaptation to logistical and economic diffi-
culties is necessary. Besides observation of treatment,
adherence requires appropriate and accessible health
care within the community.
11
Reminder letters sent to patients soon after they
had defaulted on clinic attendance increased the
Control of TB treatment through urine testing 907
patients’ rates of return by 47.7%, even among illit-
erate patients.
15
Health education given to mothers
and home visits also increased adherence among pae-
diatric patients.
16
Although not all of these strategies
to improve adherence to TB chemotherapy may be
practical in every setting, there is evidence that most
of the interventions are effective.
17
CONCLUSIONS
In-house INH test strips can be produced easily, have
the same high sensitivity (99.5%) and specificity
(96.4%) as commercially available ones and are
affordable in developing countries. Due to its phar-
macokinetics, INH is an ideal target for the control of
adherence to treatment. Both urine testing and self-
reporting are important, as are cheap tools to assess
non-adherence. In contrast, the n-butanol test and
coloration of the urine are unreliable tools to measure
adherence due to their low sensitivity and specificity,
and the fact that rifampicin can only be detected if
drugs have been taken Ͻ12 hours. The level of adher-
ence to chemotherapy can be predicted by the degree
of adherence in the first 2 months.
Acknowledgements
The study was undertaken as a project for the degree of Masters in
Tropical Paediatrics at the Liverpool School of Tropical Medicine
(PE Meissner).
The study was supported through a grant from the Dr Emil
Alexander Huebner und Gemahlin Stiftung (Germany).
References
1 Dye C, Scheele S, Dolin P, Pathania V, Raviglione M C. WHO
Global Surveillance and Monitoring Project: global burden of
tuberculosis. Estimated incidence, prevalence and mortality by
country. JAMA 1999; 282: 677–686.
2 Johnson J L, Ellner J J. Adult tuberculosis overview: African
versus Western perspectives. Curr Opin Pulm Med 2000; 6:
180–186.
3 CDC Update: Tuberculosis elimination—United States. MMWR
1990; 39: 153–156.
4 Sumartojo E. When tuberculosis treatment fails. A social be-
havioral account of patient adherence. Am Rev Respir 1993;
147: 1311–1320.
5 Eidus L, Hamilton E J. A new method for the determination of
N-acetyl isoniazid in urine of ambulatory patients. Am Rev
Respir Dis 1964; 89: 587–588.
6 Kilburn J O, Beam R E, David H L, Sanchez E, Corpe R F,
Dunn W. Reagent-impregnated paper strip for detection of
metabolic products of isoniazid in urine. Am Rev Respir Dis
1972; 106: 923–924.
7 Schraufnagel D E, Stoner R, Whiting E, Snukst-Torbeck G,
Werhane M J. Testing for isoniazid. An evaluation of the Ar-
kansas method. Chest 1990; 98: 314–316.
8 Pekovic V, Mayanja H, Vjecha M, et al. comparison of three
composite compliance indices in a trial of self-administered
preventive therapy for tuberculosis in HIV-infected Ugandan
adults. J Clin Epidemiol 1998; 51: 597–607.
9 Burkardt K R, Nel E E. Monitoring regularity of drug intake in
tuberculous patients by means of simple urine tests. S Afr Med
J 1980; 57: 981–985.
10 Elizaga J, Friedland J S. Monitoring compliance with anti- T
a
b
l
e

3
C
h
a
r
a
c
t
e
r
i
s
t
i
c
s

o
f

s
t
u
d
y

p
a
t
i
e
n
t
s
N
o
n
-
a
d
h
e
r
e
n
c
e
(
1
)
N
e
w
l
y
d
i
a
g
n
o
s
e
d

T
B

p
a
t
i
e
n
t
s
R
e
t
r
e
a
t
m
e
n
t

a
f
t
e
r

d
e
f
a
u
l
t
(
2
)
R
e
t
r
e
a
t
m
e
n
t
a
f
t
e
r
r
e
l
a
p
s
e
P
a
t
i
e
n
t
s

s
a
i
d
t
h
a
t

t
h
e
y

r
e
c
e
i
v
e
d

e
x
p
l
a
n
a
t
i
o
n
o
f

t
r
e
a
t
m
e
n
t
R
e
c
e
i
v
e
d
n
o
e
x
p
l
a
n
a
t
i
o
n
(
3
)
M
a
r
i
t
a
l

s
t
a
t
u
s
*
T
o
t
a
l
P
u
l
m
o
n
a
r
y
T
B
E
x
t
r
a
-
p
u
l
m
o
n
a
r
y
T
B
T
B
m
e
n
i
n
g
i
t
i
s
M
a
r
r
i
e
d
S
i
n
g
l
e
S
e
p
a
r
a
t
e
d
C
h
i
l
d
r
e
n
O
u
t
-
p
a
t
i
e
n
t
s
8
0
2
1
7
0
5
5
7
5
4
1
8
4
1
1
6
5
1
2
3
I
n
-
p
a
t
i
e
n
t
s
2
9
9
2
6
2
1
2
4
2
3
1
4
1
5
1
0
4
A
d
u
l
t
s
O
u
t
-
p
a
t
i
e
n
t
s
9
0
2
6
7
5
9
6
8
1
9
0
1
1
7
4
4
3
3
4
1
3
I
n
-
p
a
t
i
e
n
t
s
3
7
2
1
2
1
0
1
5
3
5
1
1
6
1
9
1
3
5
T
h
e
r
e

w
a
s

n
o

s
i
g
n
i

c
a
n
t

d
i
f
f
e
r
e
n
c
e

i
n

s
e
x
,

l
e
v
e
l

o
f

e
d
u
c
a
t
i
o
n
,

t
h
e

a
b
i
l
i
t
y

t
o

s
p
e
a
k

E
n
g
l
i
s
h

o
r

f
a
m
i
l
y

s
t
a
t
u
s
.
1
)
F
i
g
u
r
e
s

f
o
r

n
o
n
-
a
d
h
e
r
e
n
c
e

d
o

n
o
t

i
n
c
l
u
d
e

p
a
t
i
e
n
t
s

w
h
o

d
i
d

n
o
t

r
e
t
u
r
n

f
o
r

t
r
e
a
t
m
e
n
t

d
u
r
i
n
g

t
h
e

s
t
u
d
y

p
e
r
i
o
d

(
a
n

a
d
d
i
t
i
o
n
a
l

5
2

p
a
t
i
e
n
t
s
)
.

T
h
e
r
e

i
s

a

s
t
a
t
i
s
t
i
c
a
l
l
y

s
i
g
n
i

c
a
n
t

d
i
f
f
e
r
e
n
c
e

b
e
t
w
e
e
n

t
h
e

a
d
u
l
t

a
n
d

p
a
e
d
i
a
t
r
i
c

i
n
-
p
a
t
i
e
n
t
s
(
F
i
s
h
e
r

s

e
x
a
c
t
:

2
-
t
a
i
l
e
d

P

ϭ

0
.
0
0
7
)
.
2
)
O
n
e

t
h
i
r
d

o
f

a
l
l

t
e
s
t
e
d

a
d
u
l
t
s

a
r
e

r
e
l
a
p
s
e

o
r

r
e
t
r
e
a
t
e
d

T
B
-
p
a
t
i
e
n
t
s
,

c
o
m
p
a
r
e
d

t
o

o
n
l
y

1
1
.
9
%

o
f

a
l
l

c
h
i
l
d
r
e
n
.

T
h
i
s

d
i
f
f
e
r
e
n
c
e

i
s

s
i
g
n
i

c
a
n
t

f
o
r

i
n
-
p
a
t
i
e
n
t
s

(
P

Ͻ

0
.
0
0
1
;

O
R

5
.
4
3
,

9
5
%
C
I

2
.
5

1
1
.
9
)
.
3
)
T
h
e
r
e

i
s

a

s
i
g
n
i

c
a
n
t

d
i
f
f
e
r
e
n
c
e

b
e
t
w
e
e
n

p
a
e
d
i
a
t
r
i
c

a
n
d

a
d
u
l
t

p
a
t
i
e
n
t
s

w
h
o

c
l
a
i
m
e
d

t
h
a
t

t
h
e
y

i
n
i
t
i
a
l
l
y

r
e
c
e
i
v
e
d

n
o

e
x
p
l
a
n
a
t
i
o
n

a
b
o
u
t

t
r
e
a
t
m
e
n
t

(
P

ϭ

0
.
0
0
1
,

O
R

3
.
4
8
,

9
5
%
C
I

1
.
5

8
.
2
4
)
.
*
F
o
r

c
h
i
l
d
r
e
n
,

s
t
a
t
u
s

o
f

t
h
e

g
u
a
r
d
i
a
n
.
908 The International Journal of Tuberculosis and Lung Disease
tuberculosis treatment by detection of isoniazid in urine. Lan-
cet 1997; 350: 1225–1226.
11 Floyd K, Wilkinson D, Gilks C. Comparison of cost effective-
ness of directly observed treatment and conventionally deliv-
ered treatment for tuberculosis: experience from rural South
Africa. BMJ 1997; 315: 1407–1411.
12 Water T E, Naude J M, Donald P R, et al. Twice weekly vs.
daily chemotherapy for childhood tuberculosis. Ped Infect Dis
J 2000; 19: 405–410.
13 Jin B W, Kim S C, Shimao T. The impact of intensified supervi-
sory activities on tuberculosis treatment. Tubercle Lung Dis
1993; 74: 267–272.
14 Sonnenberg P, Ross M H, Shearer S C P M, Murray J. The ef-
fect of dosage cards on compliance with directly observed
tuberculosis therapy in hospital. Int J Tuberc Lung Dis 1998; 2:
168–171.
15 Paramasivan R, Partharsarathy RT, Rajasekaran S. Short
course chemotherapy: a controlled study of indirect defaulter
retrieval method. Indian J Tuberc 1993; 40: 185–190.
16 Sanmarti L, Megias J A, Gomez M N, et al. Evaluation of the
efficacy of health education on the compliance with anti-tuber-
culosis chemoprophylaxis in school children. A randomized
clinical trial. Tubercle Lung Dis 1993; 74: 28–31.
17 Volmink J, Garner P. Systematic review of randomised con-
trolled trials of strategies to promote adherence to tuberculosis
treatment. BMJ 1997; 315: 1403–1406.
R É S U MÉ
CADRE : Hôpital Mulago, Kampala, Ouganda.
OBJ ECTI F : Evaluer l’utilité des tigettes urinaires pour le
suivi de l’adhésion à la chimiothérapie antituberculeuse.
SCHÉMA : Des tigettes « maison » pour l’urine en vue de
la détection des métabolites de l’isoniazide ont été com-
parées aux tigettes des tests commerciaux. De plus, la
valeur du n-butanol pour la détection de la rifampicine a
été comparée à celle de la coloration de l’urine. L’absence
d’adhésion a été évaluée par questionnaire et révision du
registre de TB de l’Hôpital Mulago.
RÉSULTATS : On a prélevé des urines chez 236 patients
(127 adultes et 109 enfants) sous chimiothérapie quoti-
dienne. Si l’on utilise les tigettes commerciales comme stan-
dard, la sensibilité des tigettes « maison » pour l’urine a
été de 99,5% et la spécificité de 96,4%, respectivement.
La sensibilité et la spécificité du n-butanol et celles de la
coloration de l’urine pour la détection de la rifampicine
ont été basses, respectivement 64,0%/54,9% et 85,5%/
64,8%. Cinquante patients (21,2%) ont admis leur non-
adhésion à la thérapeutique au cours du mois précédent.
Chez 15 patients de plus (6,8%), cette non-adhésion a
été détectée par le test urinaire. Sur 911 patients du re-
gistre TB de l’Hôpital Mulago qui avaient commencé
leur traitement au cours des trois premiers mois de 2000,
39,7% ne l’ont pas achevé. Deux tiers de ceux-ci qui
n’avaient pas achevé le traitement ont interrompu ce
dernier au cours des 2 premiers mois.
CONCLUSI ON : Les tigettes « maison » pour le test INH
sont aussi efficaces que leur équivalent commercial pour
la détection d’isoniazide dans l’urine. Ils représentent un
outil simple pour le suivi de l’adhésion. L’adhésion à la
chimiothérapie antituberculeuse, telle que déterminée
par tigettes du test à l’isoniazide, et la révision du regi-
stre TB ont démontré une piètre adhésion. Les tests pour
la rifampicine sont moins sensibles et moins spécifiques.
R E S U ME N
MARCO DE REFERENCI A : Hospital Mulago, Kampala,
Uganda.
OBJ ETI VO : Evaluar la utilidad de las bandeletas urina-
rias para el seguimiento de la adhesión a la quimiotera-
pia antituberculosa.
MÉTODO : Se compararon bandeletas urinarias ‘caseras’
para la detección de metabolitos de isoniacida con aquel-
las de los tests comerciales. Además, se comparó el valor
del n-butanol con la coloración de la orina, para la
detección de la rifampicina. La ausencia de adhesión fue
evaluada por un cuestionario y por revisión de los regis-
tros del Hospital Mulago.
RESULTADOS : Se obtuvo orina de 236 pacientes (127
adultos y 109 niños) que recibían una quimioterapia dia-
ria. Utilizando las bandeletas de los tests comerciales
como estándar, las bandeletas caseras tuvieron una sen-
sibilidad de 99,5% y una especificidad de 96,4%. La
sensibilidad y la especificidad del n-butanol y de la colo-
ración de la orina eran bajas : 64,0%/54,9% y 85,5%/
64,8%, respectivamente. Cincuenta pacientes (21,2%)
admitieron una ausencia de adhesión al tratamiento en el
mes precedente. En otros 15 pacientes (6,8%) la ausencia
de adhesión fue detectada por el test urinario. De un total
de 911 pacientes del registro de TB del Hospital de
Mulago que habían comenzado el tratamiento en los tres
primeros meses del año 2000, el 39,7% de ellos no com-
pletaron el tratamiento. Dos tercios de los que no
habían completado el tratamiento lo habían interrump-
ido durante los 2 primeros meses.
CONCLUSI ÓN : Las bandeletas caseras para detectar la
isoniacida en la orina son tan eficaces como las comer-
ciales y constituyen una herramienta simple para el
seguimiento de la adhesión al tratamiento. La adhesión a
la quimioterapia antituberculosa determinada por las
bandeletas de detección de la isoniacida y por la revisión
de los registros de TB mostraron un cumplimiento defi-
ciente del tratamiento. Los tests para la rifampicina son
menos sensibles y menos específicos.