Nephrol Dial Transplant (2005) 20: 2797–2802
doi:10.1093/ndt/gfi060
Advance Access publication 4 October 2005
Original Article
Clinical outcomes of systemic lupus erythematosus patients
undergoing continuous ambulatory peritoneal dialysis
Yui Pong Siu1, Kay Tai Leung1, Matthew Ka Hang Tong1, Tze Hoi Kwan1 and Chi Chiu Mok2
Divisions of 1Nephrology and 2Rheumatology, Department of Medicine, Tuen Mun Hospital, Hong Kong, China
Abstract
Objectives. The purpose of this study was to evaluate
the outcome of systemic lupus erythematosus (SLE)
patients on continuous ambulatory peritoneal dialysis
(CAPD).
Methods. Eighteen SLE patients who had been under-
going CAPD for at least 3 months in our unit were
compared with 36 other age- and gender-matched non-
diabetic CAPD patients with an underlying primary
chronic glomerulonephritis (CGn). The clinical out-
come, infective complications, lupus activities, bio-
chemical parameters, haemoglobin level and the use of
erythropoietin were reviewed.
Results. The duration of dialysis of the two studied
groups was not different, with a mean of 35.4 months
for the SLE group and 36.7 months for the CGn group.
Before dialysis, SLE patients had a significantly lower
albumin level (30.4±6.6 vs 35.4±5.59 g/dl, P<0.01),
while the mean haemoglobin levels of the two groups
were similar (8.5±1.8 g/dl for SLE vs 9.0±1.9 g/dl
for the control group). However, the weekly dose of
erythropoietin (EPO) used was significantly higher
in the SLE group (6000 vs 3818 U/week, P<0.01) to
maintain a similar haemoglobin level during dialysis.
Regarding the infective complications, the SLE group
had a higher peritonitis rate (5.7 episodes/100 patient-
months vs 2.4 episodes/100 patient-months, P<0.05),
and an increase in the non catheter related infection rate
(6.67 episodes/100 patient-months vs 1.1 episodes/100
patient-months, P<0.001). However, no significant
difference could be demonstrated in the Tenckhoff
catheter exit site infection rate (2 episodes/100 vs 1.7
episode/100 patient-months). The number of patients
who received a kidney transplant or required a change
of mode to haemodialysis was similar among the
two groups. Seven patients died during the follow-up
period, and the overall mortality rate was much higher
Correspondence and offprint requests to: Dr Matthew Ka Hang Tong,
Department of Medicine, Tuen Mun Hospital, Tuen Mun,
Hong Kong, China. Email: khmtong@netvigator.com
ß The Author [2005]. Published by Oxford University Press on behalf of ERA-EDTA. All rights reserved.
For Permissions, please email: journals.permissions@oxfordjournals.org
Downloaded from https://academic.oup.com/ndt/article/20/12/2797/1924397 by Joongbu University user on 14 August 2020
in the SLE group than in the control group (0.83/100
vs 0.15/100 patient-months, P<0.05).
Conclusions. SLE patients on CAPD have a signifi-
cantly lower pre-dialysis serum albumin level and
use a higher dose of Epo to achieve a comparable
haemoglobin level than other non-diabetic CGn
CAPD patients. They also have a poorer prognosis
in terms of infective complications and mortality rate.
Keywords: continuous ambulatory peritoneal dialysis;
end-stage renal disease; mortality;
systemic lupus erythematosus
Introduction
Nephritis remains one of the most important complica-
tions in systemic lupus erythematosus (SLE) patients.
Up to 20% of lupus nephritis patients develop end-
stage renal disease (ESRD) within 10 years [1]. SLE
patients with ESRD may be at increased risk from
dialysis complications because these patients usually
have a history of pre-dialysis exposure to steroid or
cytotoxic drugs that would subject them to a number
of side effects and complications. Albeit reaching
ESRD, some patients may still have extra-renal lupus
flares which require additional immunosuppressants or
repeated courses of steriod for control, and this may
increase the susceptibility to infection. Furthermore,
SLE has been shown to be associated with premature
atherosclerosis, which may predispose SLE continuous
ambulatory dialysis (CAPD) patients to early coronary
artery diseases [2]. Consequently, patients on CAPD,
who are known to have increased risk of cardiovascular
complications, are endangered further by the concomi-
tant lupus disease.
All these adverse risk factors may impact signifi-
cantly on the outcomes of SLE patients on CAPD in
terms of complication rates and mortality. However,
data regarding this aspect are scarce. Huang et al. have
reported poorer outcomes of SLE patients compared
2798
with other patients on CAPD [3]. However, their
controls consisted of a group of CAPD patients with
heterogenous underlying renal diseases and variable
prognosis, including congenital urinary tract anomaly,
chronic glomerulonephritis (CGn), interstitial nephritis
and hypertension, thus making comparison difficult.
Among the CAPD population, primary CGn patients
in particular share similarity with lupus nephritis
patients in having a history of significant proteinuria
and exposure to steroids or other cytotoxic drugs.
Hence, we conducted a case–control study comparing
the complications and clinical outcomes of CAPD
patients with ESRD secondary to lupus nephritis with a
group of non-diabetic patients with primary CGn other
than SLE.
Methods
The study design was a case–control study matching age,
gender and race. Between January 1995 and December 2003,
20 patients with SLE, diagnosed with at least four of the
American Rheumatism Association criteria, complicated by
lupus nephritis resulting in ESRD and who had undergone
CAPD for at least 3 months, were identified. Two patients
were excluded from the study because of incomplete medical
records, and thus 18 patients were included for analysis. For
the control group, all non-diabetic patients on CAPD with
a diagnosis of biopsy-proven primary CGn within the same
recruitment period as the SLE group were reviewed. A total
of 96 patients were identified and, after matching gender and
age range, 72 patients were included. We then randomly
selected every other patient from the list, and thus 36 patients
were chosen as controls.
We made comparisons on five aspects between these
two groups, as well as making serial comparisons before
and after the initiation of dialysis within the same group:
(i) biochemical parameters; (ii) the haemoglobin level,
blood transfusion requirements and the use of recombinant
erythropoietin (Epo); (iii) lupus disease activities (SLE
group); (iv) infectious complications; and (v) the clinical
outcome. We classified the clinical outcomes into one of
the following four categories: (i) kidney transplantation;
(ii) switched over to haemodialysis; (iii) continued on
peritoneal dialysis; and (iv) all-cause mortality. The time to
achieve these clinical end-points was also analysed.
The biochemical parameters examined were the serum
creatinine, albumin, calcium and phosphate levels, fasting
total cholesterol, high-density and low-density lipoproteins
and the triglyceride levels. The lupus disease activities were
monitored serologically by means of the complement C3 and
C4 levels, and any clinical symptoms and signs of lupus
disease activity were also reviewed. The anti-double-stranded
DNA titre was not available in patients recruited before 2000;
this can cause comparison bias and therefore this titre level
was not included in the present study. We analysed the total
units of blood transfused from the start of dialysis till
reaching one of the end-points defined above; the use and
dosage of Epo received, and the serum ferritin levels were also
monitored regularly. All blood results were retrieved and
analysed at two time points: the first set of data were obtained
at the start of CAPD, and the second set of data were the
latest values before reaching clinical end-points. For those
Y. P. Siu et al.
who remained on peritoneal dialysis, the second set of data
was taken as the latest values available.
For the infectious complications, the incidences of
peritonitis, Tenckhoff catheter exit site infection (ESI) and
all other infections were recorded. The diagnosis of peritonitis
was made clinically in accordance with the International
Society of Peritoneal Dialysis guidelines, and the ESI was
defined as the presence of erythema, swelling and discharges
around the catheter exit area or a positive bacterial culture.
Statistical analysis
Downloaded from https://academic.oup.com/ndt/article/20/12/2797/1924397 by Joongbu University user on 14 August 2020
Statistical analysis was performed by Statistical Packages
for Social Sciences (SPSS version 11.0 for Windows). All
continuous variables were expressed as mean±SD unless
otherwise stated. Unpaired continuous variables were com-
pared by Student t-test or Mann–Whitney U-test, and paired
continuous variables were compared by Student t-test or
Wilcoxon test, where appropriate. Categorical data were
compared by 2 test. A P-value of <0.05 was considered
statistically significant. All tests were two-tailed.
Results
Demographic data
Eighteen Chinese SLE patients on CAPD using the
Baxter UltrabagÕ system were studied. There was
a female predominance (72%) with a mean age of
40.8±10.3 years. For the 36 race- and gender-matched
CGn controls, the mean age was 43.3±7.3 years
(P ¼ 0.57) (Table 1). In the SLE group, there were
11 (61%) WHO class IV lupus nephritis patients
and one class V nephritis patient, and the remaining
six patients did not have a renal biopsy. Two SLE
patients had been withdrawn from all immunosup-
pressive drugs before the initiation of CAPD. The
remaining 16 patients were maintained on prednisolone
with a dose ranging from 1 to 10 mg/day, two patients
were on concomitant cyclosporin and one patient was
on azathioprine. For the control group, all patients
had a biopsy-proven diagnosis of primary CGn,
including IgA nephropathy (47%), focal segmental
glomerulosclerosis (19%), mesangiocapillary (11%),
mesangioproliferative (9%) and membranous glomer-
ulonephritis (14%). Two patients were on maintenance
low dose prednisolone (5 mg/day) at the initiation of
dialysis.
Biochemical parameters
The SLE group had a significantly lower pre-dialysis
albumin level than the CGn group (30.4±6.6 vs
35.4±5.59 g/dl, P<0.01), while there was no difference
between the two groups in terms of the post-dialysis
albumin levels. For the other biochemical parameters
including the serum creatinine level, calcium and
phosphate values as well as the lipid profile (total
cholesterol, low-density lipoprotein, high-density
lipoprotein and triglyceride levels), there were no
statistical differences before and after the initiation of
Outcomes of SLE patients on CAPD 2799
Table 1. Baseline demographic data of the SLE group and CGn Table 3. Comparison of lupus activities before and after the
group initiation of dialysis, and the haemoglobin level, blood transfusion
and Epo use between the SLE group and CGn group
SLE group CGn group
SLE group CGn group P-value
No. of patients 18 36
Gender (F:M) 13:5 26:10 C3 level (g/dl)
Age 40.8±10.3 42.2±7.3 Pre-dialysis 0.68±0.4 – 0.1
System of dialysis Baxter Ultrabag Baxter Ultrabag Dialysisa 0.81±0.23
Duration of dialysis (months) 35.4±20.7 36.7±28.2 C4 level (g/dl)
No. of patients on 16 (88.9%) 2 (5.6%) Pre-dialysis 0.24±0.08 – 0.69
immunosuppressive drugs at Dialysis 0.28±0.16
the initiation of dialysis CRP (mg/l)

Downloaded from https://academic.oup.com/ndt/article/20/12/2797/1924397 by Joongbu University user on 14 August 2020

Pre-dialysis 36.85±91.2
Dialysis 24.74±74.5
Haemoglobin level 8.5±1.8 9.0±1.9 0.44
Table 2. Comparison of serum biochemical parameters between the (g/dl) dialysis
SLE and CGn group before and after the initiation of dialysis No. of units of blood 21 14 0.58
transfused (per 100
patient-months)
Biochemical parameters SLE group CGn group P-value No. of patients on Epo 7 (38.9%) 11 (30.6%) 0.5
Mean Epo dosage 6000±2309 3818±603 <0.01
Albumin (g/dl) per week (units)
Pre-dialysis 30.4±6.6 35.4±5.59 <0.01
Dialysisa 33.6±6.5 37.0±5.37 0.07 Epo ¼ erythropoietin.
a
Total cholesterol (mmol/l) Dialysis: the latest values before reaching clinical end-points.
Pre-dialysis 5.71±2.06 5.04±1.50 0.32
Dialysis 5.73±1.29 5.82±1.54 0.84
LDL-C (mmol/l)
thrombocytopenia, two patients (11%) developed
Pre-dialysis 3.38±1.85 3.10±1.25 0.64 vascular complications with retinal vasculitis, two
Dialysis 3.55±1.15 3.67±1.64 0.80 patients (11%) had evidence of active serositis and
HDL-C (mmol/l) one patient was complicated by transverse myelitis
Pre-dialysis 1.03±0.39 1.32±0.52 0.06 attributed to systemic lupus flare. Serologically, the
Dialysis 1.17±0.18 1.52±0.57 0.06 complement C3 and C4 levels did not show any
Triglyceride (mmol/l) significant change before and after the initiation of
Pre-dialysis 2.87±1.39 2.12±2.10 0.18 dialysis (P ¼ 0.1 and 0.69, respectively, Table 3).
Dialysis 2.52±2.28 2.46±2.76 0.94
The mean haemoglobin levels were also similar
Calcium (mmol/l)
Pre-dialysis 2.33±0.31 2.29±0.27 0.69
among the SLE and the control groups (8.52±1.78 vs
Dialysis 2.41±0.16 2.44±0.23 0.69 9.0±1.93 g/dl). The SLE CAPD patients had a trend
Phosphate (mmol/l) of requiring more frequent blood transfusions than
Pre-dialysis 2.08±0.68 2.00±0.52 0.68 the CGn CAPD patients (21 units vs 14 units per
Dialysis 1.99±0.65 1.86±0.57 0.50 100 patient-months), but this did not reach statistical
CRP (mg/l) significance. The number of patients that were put on
Pre-dialysis 36.85±91.2 17.1±39.8 0.46 Epo was similar in both groups, but the SLE patients
Dialysis 24.74±74.5 26.9±50.2 0.92 received a significantly higher mean dose of Epo
Residual RF (ml/min)b 1.47±2.15 1.42±1.49 0.08
Kt/V 1.91±0.62 1.83±0.41 0.66 (6000±2309 U/week in the SLE group vs 3818±
nPCR (g/kg/day) 1.05±0.16 1.10±0.20 0.36 603 U/week in the CGn group, P<0.01, Table 3).
LDL-C ¼ low-density lipoprotein cholesterol; HDL-C ¼ high-
density lipoprotein cholesterol; CRP ¼ C-reactive protein; Infectious complications
nPCR ¼ normalized protein catabolic rate.
a
Dialysis: the latest values before reaching clinical end-points. The SLE CAPD patients had a peritonitis rate of one
b
Residual renal function at the start of dialysis. episode per 17.5 patient-months, which was signifi-
cantly higher than the CGn group (one episode per 41.7
patient-months, P<0.05), but no difference could be
dialysis of both groups. The C-reactive protein level, demonstrated in the Tenckhoff catheter ESI rate
residual renal function, dialysis adequacy and the between the two groups (one episode in 50 patient-
nutritional status were also similar between the SLE months in the SLE group vs one episode in 59 patient-
and the CGn group (Table 2). months in the CGn group). Analysing the risks of
other non-catheter-related infections, the incidence of
infectious complications that resulted in hospitaliza-
Lupus activities, haemoglobin level,
tions was 6-fold higher in the SLE group than in the
blood transfusion and EPO use
controls (6.67 episodes vs 1.1 episodes per 100 patient-
Four patients (22%) had evidence of active lupus months, P<0.001). Most infections were respiratory;
activity manifested as haematological involve- other infections involved the musculoskeletal system
ment during CAPD with either pancytopenia or (cellulitis, subcutaneous abscesses and osteomyelitis),
2800
Table 4. Total number of episodes and type of infective complications
Types of infective complications Total no. of episodes (mean episodes per 100 patient-months)
SLE group
Peritonitis 40 (5.7±1.64)
Dialysis catheter exit site infection 16 (1.88±0.87)
Respiratory (pneumonia) 23 (3.25±1.08)
Cutaneous (cellulitis, subcutaneous abscess) 3 (1.08±0.68)
Gastrointestinal (gastroenteritis) 5 (1.59±0.89)
Cardiovascular (pericarditis, endocarditis) 4 (0.54±0.31)
Genitourinary (urinary tract infection, 1 (0.09±0.09)
epididymo-orchitis, vaginitis)
Skeletal (tuberculosis of spine) 0 (0)
the gastrointestinal system (gastroenteritis), the heart
(endocarditis and pericarditis) and the genitourinary
system (urinary tract infection, epididymo-orchitis and
vaginitis) (Table 4).
The risk factors for peritonitis and total infections
were analysed using logistic regression after we
adjusted for age, gender, primary renal disease as
SLE and non-SLE, the use of immunosuppressive
agents, haemoglobin level and the albumin level before
and after the start of dialysis. We found that the serum
albumin levels before and after the initiation of dialysis
were independent risk factors for CAPD peritonitis
(P ¼ 0.005 and P ¼ 0.011, respectively), whereas for
all-cause non-catheter-related infectious complications,
the pre-dialysis albumin level was the only independent
predictor (P ¼ 0.012).
Renal outcomes, transplantation and mortality
In the SLE group, five patients (28%) received a kidney
transplant during the study period, two patients were
switched over to haemodialysis (11%), while six patients
continued with CAPD (33%). For the CGn group,
15 patients were transplanted (42%), two patients
changed over to haemodialysis (6%) and 17 patients
remained on CAPD (46%). The mean duration of
dialysis till reaching any one of the above-defined end-
points in the SLE group, excluding those who remained
on CAPD, was 32.6±20.7 months, which was
comparable with the CGn group (25±16.1 months,
P ¼ 0.25).
A subgroup analysis of the duration till renal
transplantation showed that the SLE group had a
waiting time of 24.8 months, which was also similar
to the CGn group that had a mean waiting time
of 19.8 months (P ¼ 0.65). However, the SLE patients
had a much higher overall mortality rate compared
with the patients with CGn (0.83/100 vs 0.15/100
patient-months, P<0.05). Five patients in the
SLE group died during the study period. Besides one
fungal peritonitis, the most frequent cause of death was
cardiovascular events, including three acute myocardial
infarctions and one massive cerebral infarction. For the
CGn group, there were two mortalities: one patient
died of cerebrovascular accident and the other died
of miliary tuberculosis.
Y. P. Siu et al.
P-value
CGn group
37 (2.37±0.58) 0.02
14 (1.69±0.68) 0.87
4 (0.38±0.21) 0.001
3 (0.15±0.09) 0.06
4 (0.26±0.15) 0.047
0 (0) 0.014
4 (0.28±0.14) 0.37
Downloaded from https://academic.oup.com/ndt/article/20/12/2797/1924397 by Joongbu University user on 14 August 2020
1 (0.03±0.03) 0.32
Discussion
Despite the introduction of new immunosuppressive
agents and regimens, 11% of SLE patients suffering
from nephritis still require renal replacement therapy
within 8 years of diagnosis [4]. Alterations of SLE
disease activity in patients who have progressed to
ESRD have been studied extensively. Many investiga-
tors have observed that SLE activities do not remain
quiescent in ESRD or dialysis patients. Lupus flares
or active disease were reported in up to 54–67% of
patients after the initiation of dialysis in several series
[5–7], and some investigators have even described
an increased activity in dialysis patients [8,9]. In our
study, 50% (nine out of 18) of our cases continued
to show persistent disease activity clinically after the
initiation of peritoneal dialysis. In concordance with
previous reports, we found no significant change in
the serological activity (complements C3 and C4)
before and after starting dialysis. It is difficult to
predict the course of lupus disease in SLE dialysis
patients, as active disease pre-dialysis does not neces-
sarily predict an increased lupus activity after the
initiation of dialysis; however, it has been shown
that patients with a history of serositis, vasculitis or
seizure have a trend towards elevated lupus disease
activity [7].
The haemoglobin levels of our patients in the present
study were lower than recommended by international
guidelines because of the financial constraints on the
use of Epo in our locality, but there was no difference in
the haemoglobin level between the SLE and the control
groups. Concerning the use of Epo, the weekly dose of
Epo used in the SLE group was significantly higher
than in the CGn group. This higher Epo dose used may
be related to the persistence of lupus activity in the SLE
CAPD patients, resulting in increased peripheral red
cell destruction or marrow suppression with red cell
asplasia [13]. Moreover, SLE patients are apt to
produce autoantibodies, and SLE CAPD patients
may have circulating anti-Epo antibodies, thus
inducing a state of partial Epo resistance, and this
may be another reason for a higher Epo requirement in
these patients [6].
Patients on dialysis are known to have an
increased susceptibility to infections, probably related
Outcomes of SLE patients on CAPD
to the impairment of polymorphonuclear leucocyte
phagocytosis, and this will be aggravated further in
immunocompromised subjects or patients receiving
immunosuppressive drugs [9]. There is no consensus
on the optimal immunosuppressive protocol in SLE
dialysis patients. In common practice, patients are
usually maintained on low dose steroid with or without
concomitant cytotoxic drugs such as azathioprine, or
cyclosporin. It has been shown that steroid-treated
dialysis patients are associated with a less favourable
nutritional condition, lower serum albumin level and
body mass index, and a higher C-reactive protein [10],
and immunosuppressed CAPD patients have also been
demonstrated to experience more peritonitis episodes,
more frequent hospital admissions and more days off
CAPD, and required more laparotomies to remove
infected Tenckhoff catheters [11].
In our SLE CAPD patients, 89% (16 out of 18) were
maintained on some form of immunosuppressive
therapy. They had a lower pre-dialysis albumin level
together with a significantly higher peritonitis rate in
comparison with the CGn group (2.5 times more
episodes per 100 patient-months). When we compared
the overall non-catheter-related infectious complica-
tions, the SLE group had an incidence rate nearly
6-fold higher than the CGn patients. Most of the
serious non-catheter-related infections involved the
respiratory tract, and infections of the skin and soft
tissue as well as the gastrointestinal tract were also
common. Although we could not demonstrate SLE
to be an independent risk factor for infection in
CAPD patients, SLE patients did have a significantly
lower pre-dialysis albumin level (P ¼ 0.01), which we
identified to be the only independent predictor.
Therefore, SLE may cause increased infection by
inducing a poorer nutritional state as a result of chronic
inflammation.
We found an increased mortality in SLE patients
undergoing CAPD, around five times higher than in
patients with an underlying renal pathology of primary
CGn. This is in concordance with previous literature
which reported a 4.3 times increased risk of death and
a 5-year survival rate of 58–73% for SLE dialysis
patients in contrast to a 95% 5-year survival in non-
SLE patients on dialysis [12,13]. Previous studies
have attributed this increased mortality to infection.
However, in our present study, only one of the five
deaths was due to sepsis, while the remaining four were
due to cardiovascular events. SLE has been shown to
induce premature atherosclerosis because of its chronic
inflammatory, immune complex- and autoantibody-
inducing properties [2]. Unfortunately, the antiphos-
pholipid antibody status was not available in our
SLE patients to determine its role in the develop-
ment of cardiovascular complications. The prevalence
of antiphospholipid antibodies in SLE patients
undergoing CAPD is not known. In a study performed
in non-SLE haemodialysis patients, it has been
shown that there is an
30% prevalence of IgG-
anticardiolipin antibodies and 22–30% of lupus anti-
coagulant [14], and we may expect even higher values
2801
in SLE dialysis patients. Hence, besides the atherogenic
inflammatory effects of SLE, the presence of antiphos-
pholipid antibodies may also play a significant role in
the increased incidence of cardiovascular complications
in SLE dialysis patients; however, further studies are
required for clarification.
It has been reported previously that SLE patients on
CAPD have a higher incidence of technique failure and
a reduced chance of receiving a kidney transplant [3].
However, in our study, the number of SLE CAPD
patients with technique failure requiring a change of
Downloaded from https://academic.oup.com/ndt/article/20/12/2797/1924397 by Joongbu University user on 14 August 2020
mode to haemodialysis, or who received a kidney
transplant was similar to the controls. Also, the waiting
time for SLE CAPD patients to receive a kidney
transplant was not different from that of CGn patients.
This discrepancy in the results is probably related to
the difference in the selection criteria of the control
groups. We have recruited a more homogenous group
of patients as controls than in previous series,
with underlying idiopathic CGn as the cause of
ESRD. Many of these patients had a history of
significant proteinuria and of steroid or cytotoxic
drug exposure, thus more closely resembling patients
with lupus nephritis. Therefore, based on our present
data, we concluded that patients on CAPD due to
SLE-related glomerulonephritis are not inferior to
patients on dialysis due to CGn in terms of CAPD
technique survival or the chance of having a kidney
transplant.
In conclusion, a significant proportion of lupus
ESRD patients still have clinically persistent lupus
disease activity after the initiation of dialysis but,
despite this, they should not be precluded from CAPD
as a group; rather, strategies to minimize the infective
complications should be employed. The use of immuno-
suppressants should be minimized and steroid with-
drawn if there is no evidence of any extrarenal lupus
activities. However, as SLE CAPD patients who
have a low pre-dialysis serum albumin level are at an
increased risk of CAPD peritonitis and all-cause
infections, and associated with a high mortality, we
would advise haemodialysis treatment rather than
CAPD in this particular subgroup, at least until the
patients have achieved a normalization of their
inflammatory and nutritional parameters. We have
also demonstrated that SLE CAPD patients have a
5-fold increase in all-cause mortality compared with
patients on CAPD secondary to CGn, mostly due to
cardiovascular events; therefore, more intense modifi-
cation of traditional cardiovascular risk factors in
SLE patients undergoing CAPD is crucial.
Conflict of interest statement. None declared.
References
1. Dooley MA, Hogan S, Jennette C, Falk R. Cyclosphosphamide
therapy for lupus nephritis: poor renal survival in black
Americans. Kidney Int 1997; 51: 1188–1195
2. Abusamieh M, Ash J. Atherosclerosis and systemic lupus
erythematosus. Cardiol Rev 2004; 12: 267–275
2802
3. Huang JW, Hung KY, Yen CJ, Wu KD, Tsai TJ.
Systemic lupus erythematosus and peritoneal dialysis:
outcomes and infectious complications. Perit Dial Int 2001; 21:
143–147
4. Mitic I, Ilic T, Strazmester-Majstorovic G et al. Long term
survival of patients with lupus nephritis. Srp Arh Celok Lek
2002; 130 [Suppl 3]: 32–37
5. Bruce IN, Hallett DC, Gladman DD, Urowitz MB. Extrarenal
disease activity in systemic lupus erythematosus is not
suppressed by chronic renal insufficiency or renal replacement
therapy. J Rheumatol 1999; 26: 1490–1494
6. Tzioufas AG, Kokori SI, Petrovas CI, Moutsopoulos HM.
Autoantibodies to human recombinant erythropoietin in patients
with systemic lupus erythematosus: correlation with anemia.
Arthritis Rheum 1997; 40: 2212–2216
7. Szeto CC, Li PK, Wong TY, Leung CB, Lui SF.
Factors associated with active systemic lupus erythe-
matosus after endstage renal disease. J Rheumatol 1998; 25:
1520–1525
8. Berend K, Nossent H. Reactivation of systemic lupus
erythematosus after transfer to peritoneal dialysis. Nephrol Dial
Transplant 1997; 12: 2808
Y. P. Siu et al.
9. Stock GG Jr, Krane NK. Treatment of end-stage renal disease
due to lupus nephritis: comparison of six patients treated with
both peritoneal and hemodialysis. Adv Perit Dial 1993; 9:
147–151
10. Altieri P, Sau G, Cao R et al. Immunosuppressive treatment
in dialysis patients. Nephrol Dial Transplant 2002; 17
[Suppl 8]: 2–9
11. Andrews PA, Warr KJ, Hicks JA, Cameron JS. Impaired
outcome of continuous ambulatory peritoneal dialysis in
immunosuppressed patients. Nephrol Dial Transplant 1996; 11:
1104–1108
12. Lee PT, Fang HC, Chen CL, Chiou YH, Chou KJ, Chung HM.
Poor prognosis of end-stage renal disease in systemic lupus
Downloaded from https://academic.oup.com/ndt/article/20/12/2797/1924397 by Joongbu University user on 14 August 2020
erythematosus: a cohort of Chinese patients. Lupus 2003; 12:
827–832
13. Krane NK, Burjak K, Archie M, O’Donovan R. Persistent
lupus activity in end-stage renal disease. Am J Kidney Dis 1999;
33: 872–879
14. Phillips AO, Jones HW, Hambley H, Hillis AN, Hendry BM.
Prevalence of lupus anticoagulant and anticardiolipin
antibodies in haemodialysis patients. Nephron 1993; 65:
350–353
Received for publication: 19.4.05
Accepted in revised form: 14.7.05