Presented in part at the Annual Meeting of the Taiwan Society of Nephrology held 8–9 December 2007, in Taipei, Taiwan.
Impact of Renal Survival on the Course and Outcome of Systemic Lupus Erythematosus Patients Treated With Chronic Peritoneal Dialysis
Article first published online: 17 JUL 2009
© 2009 The Authors. Journal compilation © 2009 International Society for Apheresis
Therapeutic Apheresis and Dialysis
Volume 14, Issue 1, pages 35–42, February 2010
How to Cite
Liang, C.-C., Huang, C.-C., Wang, I.-K., Chang, C.-T., Chen, K.-H., Weng, C.-H., Lin, J.-L., Hung, C.-C., Yang, C.-W. and Yen, T.-H. (2010), Impact of Renal Survival on the Course and Outcome of Systemic Lupus Erythematosus Patients Treated With Chronic Peritoneal Dialysis. Therapeutic Apheresis and Dialysis, 14: 35–42. doi: 10.1111/j.1744-9987.2009.00703.x
- Issue published online: 12 FEB 2010
- Article first published online: 17 JUL 2009
- Received December 2008; revised March 2009.
- Peritoneal dialysis;
- Peritoneal solute transporter rate;
- Renal survival;
- Systemic lupus erythematosus
This longitudinal study investigated whether renal survival can affect the course and outcome of systemic lupus erythematosus (SLE) patients treated with chronic peritoneal dialysis (PD). Thirty-five SLE patients, out of 1115 end-stage renal disease (ESRD) patients treated with chronic PD, were seen between 1990 and 2007 at the Chang Gung Memorial Hospital. Patients were followed up for a mean of 38.8 ± 22.9 months. There were no significant differences between patients with short renal survival (<3 years) and long renal survival (>3 years) for the various demographic variables such as age, sex, PD duration, immunosuppressive drug administration, or exchange system (P > 0.05). Interestingly, before PD, patients with short renal survival had lower serum complement levels than patients with long renal survival (C3, 40.2 ± 14.4 vs 76.3 ± 18.5 mg/dL, P < 0.001; and C4, 14.8 ± 4.7 vs 22.4 ± 8.1 mg/dL, P < 0.05). However, the differences in complement levels between the groups disappeared after PD (C3, 76.5 ± 27.3 vs 84.2 ± 27.8 mg/dL; and C4, 26.7 ± 11.3 vs 22.6 ± 10.8 mg/dL, both P > 0.05). Patients with short renal survival were more likely to have a high peritoneal solute transporter rate (PSTR) than their long renal survival counterparts (χ2-test, P = 0.02, and AUROC = 0.744 and P = 0.040); however, there were no significant differences for other variables such as cardiothoracic ratio (CTR), Kt/V, residual renal function, exit site infection, and peritonitis (P > 0.05). Finally, Kaplan–Meier analysis revealed that the two groups did not differ in patient and technical survival (P > 0.05). Therefore it was concluded that renal survival might be associated with PSTR, but not with patient and technical survival in SLE patients treated with PD.
Renal involvement in systemic lupus erythematosus (SLE) is a common manifestation of the disease and a strong predictor of poor outcome. The general consensus is that more than 50% of SLE patients will develop clinically relevant nephritis at some time during the course of their illness (1–3). In 20%, despite appropriate treatment, kidney damage develops and progresses to end-stage renal disease (ESRD) (1–3). According to the US Renal Data System 2008 (4), Taiwan had the world's highest incidence and prevalence of ESRD. In addition, SLE patients comprise a minor but important proportion of ESRD patients receiving chronic peritoneal dialysis (PD) in our hospital. Patients with ESRD due to lupus nephritis are difficult to manage because of active lupus activity and the requirement for immunosuppressive therapy even after they enter a chronic PD program. Due to long-term exposure to steroids or cytotoxic drugs, SLE patients with ESRD have been shown to have accelerated atherosclerosis and increased risk of infectious complications (5–7). These adverse risk factors have a significant negative impact on patient and technical survival in SLE patients undergoing hemodialysis (HD) and PD (3,7). Although most studies have reported good outcomes for SLE patients after dialysis (8,9) some investigations have found different results if all patients were considered (10,11).
The rationale for this study was based on an important, but as yet unanswered, question that arose for many SLE patients treated with chronic PD in our hospital. Some SLE patients suffered severe lupus nephritis and progressed to ESRD in a relatively short period of time (<3 years) whereas other SLE patients suffered mild lupus nephritis and progressed to ESRD over a longer period (>3 years). Therefore, this raises the question of what the impact of renal survival before dialysis on the outcome of SLE patients treated with chronic PD is. Although there have been a few studies of SLE patients who underwent PD (3,7,8,12,13), investigation of the impact of renal survival on the outcome of SLE patients who underwent PD is lacking. Therefore, we analyzed our SLE patients with regard to renal survival, that is, the time from diagnosis of SLE to the development of ESRD and initiation of renal replacement therapy. This study might help provide a solution to the unanswered question, which for a long time has worried many of our SLE patients.
PATIENTS AND METHODS
This clinical study followed the principles of the Declaration of Helsinki and was compatible with the policies of the Medical Ethics Committee of the Chang Gung Memorial Hospital.
This longitudinally observational study included 35 patients with SLE out of 1115 patients who underwent chronic PD as renal replacement therapy at Chang Gung Memorial Hospital between 1990 and 2007. All 35 patients fulfilled at least four of the American College of Rheumatology criteria for the diagnosis of SLE (14). Patients were excluded if they had malignancies or active infectious diseases, were hospitalized or underwent major surgery within the past 3 months, had ever received HD for >3 months before PD, had undergone PD for <3 months (3), or had ever had lead (15) or cadmium (16) intoxication.
Systemic lupus erythematosus was diagnosed according to the American College of Rheumatology criteria (14), whereas lupus nephritis was defined clinically or pathologically. Renal survival was defined as survival without renal replacement therapy. Short or long renal survival was defined as survival without renal replacement therapy for less than or more than 3 years, respectively. ESRD was defined as the time of initiation of chronic renal replacement therapy. Renal pathology was categorized according to World Health Organization (WHO) criteria, that is:
- • Class I—minimal mesangial lupus glomerulonephritis;
- • Class II—mesangial proliferative lupus glomerulonephritis;
- • Class III—focal lupus glomerulonephritis;
- • Class IV—diffuse lupus glomerulonephritis;
- • Class V—membranous lupus glomerulonephritis; and
- • Class VI—advanced sclerotic lupus glomerulonephritis.
Loss of residual renal function was defined as urine output of <100 mL per day. Peritonitis was diagnosed if two of the following three criteria were satisfied: (i) abdominal pain or tenderness; (ii) turbid effluent containing white blood cells at >100/mL, with >50% polymorphonuclear leukocytes; and (iii) a pathogen identified from effluent culture or smear (7). Exit-site infection was diagnosed by the presence of local tenderness, erythema, and discharge at the Tenckhoff catheter exit site (7). The incidence rates of peritonitis and exit-site infection were calculated as the number of infectious episodes divided by the duration of follow-up, and are expressed as episodes per 100 patient-months. The cardiothoracic ratio (CTR) was calculated as the ratio of the maximal transverse diameter of the cardiac silhouette (“heart diameter”) to the distance between the internal margins of the ribs at the level of the right hemidiaphragm (“transverse thoracic diameter”) (17). Cardiomegaly was defined as CTR > 0.5 (17).
All laboratory data were obtained with standard laboratory procedures using an automatic analyzer. Hematological studies included hemoglobin, white blood cell count, and platelet count. Serum biochemical studies comprised serum creatinine, albumin, glucose, calcium, phosphate, sodium, potassium, chloride, aspartate aminotransferase (AST), alanine transaminase (ALT), alkaline phosphatase, total bilirubin, uric acid, triglyceride, and cholesterol. The corrected serum calcium level was calculated as: calcium (mg/dL) + (0.8 [4.0 − albumin (g/dL)]). Lupus activity was regularly monitored by blood testing for serum complement 3 (C3) and complement 4 (C4) levels. Dialysis clearance of urea was expressed as Kt/Vurea, as reported by Daugirdas (18). The peritoneal equilibration test (PET) was performed as described by Twardowski et al. (19).
Data were expressed as mean ± SD unless otherwise stated. All statistical analyses were performed using Statistical Packages for Social Sciences (SPSS) version 13.0 for Windows (Chicago, IL, USA). Data were routinely tested for normality of distribution and equality of standard deviations before analysis. For comparison of the different variables between patients with short or long renal survival, unpaired continuous variables were compared by the Student's t-test or Mann–Whitney U-test, and paired continuous variables were compared by the Student's t-test or Wilcoxon test, as appropriate. The incidences of exit site infection and peritonitis were compared by the Poisson test. Categorical parameters were compared with the χ2-test or Fisher exact test. The Kaplan–Meier method was used to calculate the cumulative incidence of patient and technical survival, and the difference was determined by the log-rank test. The criterion for significance was the 95% confidence interval (CI) to reject the null hypothesis.
Demographic features of the 35 patients are shown in Table 1. The data for 10 patients with short renal survival (<3 years) was compared with that of 25 patients with long renal survival (>3 years). Patients with short renal survival had significantly shorter renal survival than patients with long renal survival (22.60 ± 12.01 vs 104.28 ± 64.25 months, P < 0.001). There was a female predominance (80%) and a mean age of 29.7 ± 9.39 years in the short renal survival group. There was also a female predominance (88%) and a mean age of 32.48 ± 10.52 years in the long renal survival group. Nevertheless, there were no significant differences between the groups in terms of gender, age, duration of PD, or use of immunosuppressive drugs (P > 0.05). There were only 13 patients (13/35 or 37.1%) who received a percutaneous renal biopsy at the time of initial clinical presentation. Patients with WHO class IV or VI lupus nephritis accounted for most of the cases and there was no significant difference between the groups in renal pathology (P > 0.05). Finally, there was no difference between the groups in the types of exchange PD system used (P > 0.05).
|Variables||Renal survival <3 years (N = 10)||Renal survival >3 years (N = 25)||P|
|Female||8 (80%)||22 (88%)||NS|
|Age (years)||29.70 ± 9.39||32.48 ± 10.52||NS|
|Renal survival (months)||22.60 ± 12.01||104.28 ± 64.25||***|
|Renal pathology (N = 13)|
|Duration of PD (months)||39.8 ± 23.4||38.32 ± 23.29||NS|
|Use of immunosuppressive drugs before PD||9 (90%)||22 (88%)||NS|
|Use of immunosuppressive drugs after PD||4 (40%)||16 (64%)||NS|
|O set||1 (10%)||1 (4%)||NS|
|Ultra set||1 (10%)||0 (0%)||NS|
|Ultraviolet flash||1 (10%)||1 (4%)||NS|
|Twin bag||5 (50%)||18 (72%)||NS|
|Automated PD||2 (20%)||5 (20%)||NS|
Hematological, biochemical and immunologic parameters
Table 2 compares the two groups with regard to the various hematological, biochemical, and immunologic variables. There were no significant differences between the groups in hemoglobin, white blood cell count, and platelet count before or after PD (P > 0.05). The serum creatinine level was lower in the short renal survival group before PD (P < 0.05), but no significant difference was noted after PD (P > 0.05). Patients with short renal survival had a trend of lower predialysis serum albumin and phosphate levels compared with patients who had long renal survival before PD, but the differences did not reach statistical significance (P = 0.061 and P = 0.056, respectively). For other biochemical variables, such as fasting glucose, calcium, sodium (data not shown), potassium (data not shown), AST (data not shown), ALT (data not shown), alkaline phosphatase (data not shown), total bilirubin (data not shown), uric acid (data not shown), cholesterol, and triglyceride, there were no differences between the groups before or after PD (P > 0.05). In addition, patients with short renal survival had lower serum C3 and C4 levels than patients with long renal survival before PD (P < 0.001 and P = 0.025, respectively). The serum C3 and C4 levels were similar between the two groups after PD (P > 0.05).
|Variables||Renal survival <3 years (N = 10)||Renal survival >3 years (N = 25)||P|
|Predialysis||8.00 ± 1.99||7.89 ± 2.15||NS|
|Postdialysis||8.46 ± 1.72||9.06 ± 1.56||NS|
|White cell count (/µL)|
|Predialysis||6.81 ± 1.39||6.34 ± 2.73||NS|
|Postdialysis||7.72 ± 2.04||7.52 ± 2.50||NS|
|Platelet count (1000/µL)|
|Predialysis||176.88 ± 62.17||162.67 ± 73.33||NS|
|Postdialysis||259 ± 90.06||236.88 ± 88.97||NS|
|Predialysis||7.17 ± 3.17||10.31 ± 4.31||0.036*|
|Postdialysis||12.43 ± 4.49||11.58 ± 2.72||NS|
|Predialysis||2.71 ± 0.83||3.34 ± 0.83||0.061|
|Postdialysis||3.38 ± 0.68||3.50 ± 0.60||NS|
|Fasting glucose (mg/dL)|
|Predialysis||99.78 ± 25.06||93.95 ± 21.98||NS|
|Postdialysis||96.44 ± 20.63||96.67 ± 19.31||NS|
|Predialysis||8.24 ± 0.77||8.07 ± 1.15||NS|
|Postdialysis||9.40 ± 1.41||9.45 ± 0.57||NS|
|Predialysis||4.69 ± 2.03||6.39 ± 2.22||0.056|
|Postdialysis||5.43 ± 1.58||5.32 ± 1.17||NS|
|Predialysis||218.33 ± 46.90||236.05 ± 64.15||NS|
|Postdialysis||194.67 ± 39.68||209.38 ± 46.52||NS|
|Predialysis||231.14 ± 107.29||211.35 ± 147.51||NS|
|Postdialysis||213.80 ± 134.32||245.43 ± 182.81||NS|
|Predialysis||38.01 ± 12.41||76.27 ± 18.47||***|
|Postdialysis||83.86 ± 33.79||84.18 ± 27.86||NS|
|Predialysis||14.12 ± 4.48||22.44 ± 8.10||0.025*|
|Postdialysis||25.83 ± 10.79||22.60 ± 10.78||NS|
Clinical course after PD
As shown in Table 3, there were no significant differences between the two groups in fluid control (measured by CTR), adequacy of dialysis (measured by Kt/V), or loss of residual renal function (P > 0.05). Besides this, there were no differences in the incidence of exit-site infection or peritonitis between two groups (P > 0.05); however, the two groups were different with regards to the PET. First, patients with short renal survival had higher high- or high-average peritoneal transporters than those with long renal survival (7/10 vs 6/25, P = 0.02). Second, patients with short renal survival had higher dialysate/plasma creatinine ratios than the long renal survival patients (0.69 ± 0.12 vs 0.62 ± 0.11, P = 0.05). Third, patients with short renal survival drained out less volume than those with long renal survival (2264 ± 164.53 vs 2441.2 ± 181.57 mL, P < 0.01). Area under the receiver operating characteristic curve (AUROC) analysis was used to assess the discriminative powers of renal survival, predialysis serum albumin, and C3 against PSTR (20). As revealed in Figure 1, renal survival can fairly discriminate high from low PSTR (AUROC = 0.744, P = 0.040) in these patients. The predialysis serum albumin had a trend of association with PSTR, but did not reach statistical significance (AUROC = 0.731, P = 0.051).
|Variables||Renal survival <3 years (N = 10)||Renal survival >3 years (N = 25)||P|
|Cardiomegaly (CTR > 0.5)||4 (40%)||14 (46%)||NS|
|Adequacy of dialysis (Kt/V)||2.09 ± 0.28||2.11 ± 0.40||NS|
|Peritoneal equilibrium test|
|High or high-average transporters||7 (70%)||6 (24%)||0.02*|
|Dialysate/plasma creatinine ratio||0.69 ± 0.12||0.62 ± 0.11||0.05*|
|Drained out volume (mL)||2264.00 ± 164.53||2441.20 ± 181.57||**|
|Loss of residual renal function||7 (70%)||15 (60%)||NS|
|Exit-site infection (episodes/100 months)||2.1 ± 2.079||1.2 ± 1.732||NS|
|Peritonitis (episodes/100 months)||1.600 ± 1.174||1.480 ± 2.163||NS|
Clinical outcomes after PD
Table 4 lists the clinical outcomes after PD. At the end of the study period, there were 3 (30%) patients in the short renal survival group and 12 (48%) patients in the long renal survival group who were alive and continued to receive PD. Seven (70%) patients in the short renal survival group and 13 (52%) in the long renal survival group withdrew from PD due to various reasons. Three patients (30%) with short renal survival and 3 patients (12%) with long renal survival died after PD. The most important causes of mortality were cardiovascular and infection. As shown in Figures 2 and 3, the Kaplan–Meier analysis did not find any significant differences between the two groups after PD in terms of patient (P = 0.2174, Fig. 2) or technical (P = 0.8214, Fig. 3) survival. The data revealed that 5-year patient survival was 77.8% and 81.4% for patients with short and long renal survival, respectively. For all patients, the 5-year patient survival was 80.4%.
|Variables||Renal survival <3 years (N = 10)||Renal survival >3 years (N = 25)||P|
|1) Remain on PD||3 (30%)||12 (48%)||NS|
|2) Drop out from PD||7 (70%)||13 (52%)||NS|
|All-cause||3 (30%)||3 (12%)||NS|
|Infectious-cause||2 (20%)||1 (4%)||NS|
|Cardiovascular disease-cause||1 (10%)||2 (8%)||NS|
|b) Transfer to HD due to PD technical failure|
|Infection-related failure||3 (30%)||4 (16%)||NS|
|Mechanical failure||1 (10%)||3 (12%)||NS|
|c) Renal transplantation||0||2 (8%)||NS|
|d) Spontaneous renal recovery||0||1 (4%)||NS|
As mentioned above, there are few studies (3,7,8,12,13) in the literature describing the outcome of SLE patients undergoing chronic PD. The main reason for the limited number of studies was the relatively small number of SLE patients entering chronic PD programs. In addition, lupus patients are predominantly female and of a younger age, which creates difficulties for cohort studies (8,9). In this study, instead of using intermittent measurements of the SLE Disease Activity Index (SLEDAI) or serological lupus activity, renal survival was used to represent predialysis lupus severity. During the course of lupus nephritis in progression to ESRD, both immunologic and non-immunologic insults play important roles. Non-immunologic insults include sepsis, exposure to nephrotoxic agents, hypertension, proteinuria, and hyperperfusion of the remaining glomeruli (8). Previous studies (7,12,13) have almost focused on the impact of immunologic insults on SLE patients after dialysis. Nevertheless, the non-immunologic impacts are also important. Therefore, renal survival might be more representative of predialysis lupus severity than either the SLEDAI or serological lupus activity.
According to Nossent et al. (3), most of their patients had a slower progression in the deterioration of renal function before the initiation of PD. Only one third (34.5%) began chronic PD within 3 years, that is, they were in the short renal survival group. Similarly, patients with short renal survival accounted for approximately one third of patients (28.6%) in our study. Although these SLE patients had a rapid deterioration of renal function before dialysis, our data indicated that renal survival did not have any impact on the outcome after chronic PD. A similar result was also mentioned by Nossent et al. (3), although most patients in their study underwent chronic HD; therefore, we believe that the rate of progression in lupus nephritis did not affect the outcome of SLE patients undergoing chronic PD. On the other hand, the primary causes of death in this study were infection and cardiovascular disease. The data are also compatible with the studies by Jarrett et al. (10) and Coplon et al. (11), who claimed that death in SLE patients after 3 months of dialysis was mostly caused by infection and cardiovascular disease, rather than by active lupus activity.
In our study, patients with short renal survival had a lower serum creatinine level before dialysis than patients with long renal survival (P = 0.036). Moreover, predialysis serum albumin and phosphate levels in the short renal survival group were also lower, although the P values did not reach statistical significance (P = 0.061 and P = 0.056, respectively). The above results imply that before dialysis, patients with short renal survival had poorer nutrition than patients with long renal survival. Nevertheless, the discrepancies in the nutritional parameters between the two groups disappeared after PD. It is well known that good nutritional status is a prerequisite for good patient survival, not only for dialysis patients but also for SLE patients (1,22). Despite not reaching statistical significance, patients with short renal survival did have lower serum albumin prior to dialysis, which is compatible with the argument that hypoalbuminemia is not only a marker of poor nutrition, but also markers of poor outcome (21,22).
Similarly, compared to patients with long renal survival, patients with short renal survival had more active lupus activity before dialysis. The difference in serological lupus activity between the two groups also disappeared after dialysis. One possible explanation is that the short renal survival group had more severe inflammation due to more active lupus activity and poorer response to immunosuppressive therapy, which might increase the catabolic rate and contribute to poor nutritional status. Hence, the similarity in nutritional status and lupus activity after chronic PD might explain why there was no difference in patient survival between the two groups after chronic PD.
Regarding the clinical course, no differences between the two groups were observed in fluid control (measured by CTR), adequacy of dialysis (measured by Kt/V), or loss of residual renal function; however, patients with short renal survival had higher dialysate/plasma creatinine ratio, lower drained out volume, and higher high/high-average transport characteristics than patients with long renal survival (P = 0.05, P < 0.01, and P = 0.02, respectively). Therefore, the short renal survival group had a higher chance of becoming high PSTR. Furthermore, AUROC analysis revealed that renal survival can fairly discriminate between high and low PSTR. In a subsequent multiple logistic regression model, renal survival was still associated with peritoneal membrane transporters (P = 0.05, OR 1.8%, data not shown) after correction of gender and age.
The AUROC analysis also confirmed an association between albumin (an acute phase protein and inflammatory marker) and PSTR, which almost reached statistical significance (P = 0.051). This raises the question of how renal survival before dialysis could influence peritoneal membrane transportation. As mentioned, lupus patients with short renal survival had more immunologic and non-immunologic insults before PD, compared with patients who had long renal survival. These insults would bring about malnutrition, inflammation and atherosclerosis (MIA) syndrome, which probably leads subsequently to an increase in peritoneal permeability (23,24). It has been proposed that there are three different types of fast transporters (25,26). Early inherent phenotype type I describes those patients who have high PSTR at the beginning of PD and are inflamed systemically and intraperitoneally. Our data implies that patients with short renal survival might belong to the early inherent phenotype type I.
The 5-year patient survival is 80.4%, probably related to the relatively young age of our patients. Most importantly, we have observed that renal survival before PD might associate with PSTR, but not patient and technical survival in SLE patients treated with chronic PD. The findings might be partly explained by a poorer nutritional status and more severe micro-inflammation in the patients with short renal survival before PD. After chronic PD, the differences in variables between the two groups disappeared, which can probably explain why renal survival before PD had no impact on patient and technical survival after chronic PD. However, a small patient population and a relatively short follow-up period after dialysis limit the certainty of our conclusion. We suggest that larger and longer observations should be conducted in the future to delineate the impact of renal survival on the course and outcome in SLE patients following chronic PD.
- 4U.S. Renal Data System. USRDS 2008 Annual Data Report: atlas of chronic kidney disease and end-stage renal disease in the United States. National Institutes of Health, National Institute of Diabetes and Digestive and Kidney Diseases, Bethesda, MD, 2008.
- 19Peritoneal equilibration test. Perit Dial Bull 1987;7:138–47., , et al.