Search

Background The benefits of biocompatible peritoneal dialysis (PD) fluids, particularly for residual renal function (RRF), are controversial. Moreover, the clinical effects of a PD regimen consisting of different biocompatible PD fluids have not been fully established. Study Design Prospective, randomized, controlled, open-label study. Setting & Participants Patients with end-stage kidney disease newly started on continuous ambulatory PD therapy (N = 150). Intervention A 12-month intervention with 3 biocompatible PD fluids (a neutral-pH, low glucose degradation product, 1.5% glucose solution; a solution with 1.1% amino acid; and a fluid with 7.5% icodextrin) or conventional PD fluid. Outcomes The primary outcome was change in RRF and daily urine volume. Secondary outcomes were peritoneal transport and inflammation markers. Measurements RRF, daily urine volume, serum and dialysate cytokine levels. Results RRF(3.24 ± 1.98 vs 2.88 ± 2.43 mL/min/1.73 m2; P = 0.9) and rate of decline in RRF (−0.76 ± 1.77 vs −0.91 ± 1.92 mL/min/1.73 m2 per year; P = 0.6) did not differ between the biocompatible- and conventional-PD-fluid groups. However, patients using the biocompatible PD fluids had better preservation of daily urine volume (959 ± 515 vs 798 ± 615 mL/d in the conventional group, P = 0.02 by comparison of difference in overall change by repeated-measures analysis of variance). Their dialysate-plasma creatinine ratio at 4 hours was higher at 12 months (0.78 ± 0.13 vs 0.68 ± 0.12; P = 0.01 for comparison of the difference in overall change by repeated-measures analysis of variance). They also had significantly higher serum levels of adiponectin and overnight spent dialysate levels of cancer antigen 125, adiponectin, and interleukin 6 (IL-6). No differences between the 2 groups were observed for serum C-reactive protein and IL-6 levels. Limitations Unblinded, relatively short follow-up; no formal sample-size calculations. Conclusions Use of a combination of 3 biocompatible PD fluids for 12 months compared with conventional PD fluid did not affect RRF, but was associated with better preservation of daily urine volume. The biocompatible PD fluids also lead to changes in small-solute transport and an increase in dialysate cancer antigen 125, IL-6, adiponectin, and systemic adiponectin levels, but have no effect on systemic inflammatory response. The clinical significance of these changes, while of great interest, remains to be determined by further studies.

Objectives To present and discuss the epidemiological and clinical aspects, as well as therapeutic options and outcome of de novo renal cell carcinoma (RCC) of the native kidneys in a series of Chinese renal transplant recipients. Patients and Methods A retrospective, cohort study examining all renal transplant recipients with the diagnosis of RCC of native kidney followed up in two major regional hospitals in Hong Kong between January 2000 and December 2009. Clinical data includedage, gender, cause of renal failure, symptoms at presentation, duration of transplantation, immunosuppressive therapy, and history of acquired cystic kidney disease (ACKD). Laboratory, radiographic, operative, and pathology reports were used to assess the tumor extent. Results Among the 1,003 renal transplant recipients recruited, 12 transplant recipients had a nephrectomy for a total of 13 RCC. The prevalence of de novo RCC was 1.3%. The mean age at diagnosis of RCC was 48.4 years, and the median time from transplantation to diagnosis was 6.1 years. ACKD was found in 6 (50%) of the patients. All patients except one were asymptomatic. pT1 disease was found in ten patients with a mean tumor size of 3.2 cm. All patients were treated successfully with radical nephrectomy. After a median follow-up of 38 months, two patients (16.7%) died. One died of sepsis, and the other died of metastatic carcinoma. Conclusions With increasing data showing a better prognosis if RCC is detected early by screening, it is time to consider screening all kidney transplant recipients for ACKD and RCC. © The Author(s) 2011. This article is published with open access at Springerlink.com., published_or_final_version, Springer Open Choice, 21 Feb 2012

Aim: Nocturnal home haemodialysis (NHHD) was started in Hong Kong in 2006. The experience of 1 year of NHHD with an alternate night schedule in two local centres is reported. Methods: The clinical parameters of 14 patients who had completed 1 year of NHHD were retrospectively analyzed. All patients were receiving an alternate night schedule (3.5 sessions/week) for 6–8 h/session. Results: After 1 year of NHHD, haemoglobin levels increased from 9.6 ± 1.6 g/dL before NHHD to 11.4 ± 2.2 g/dL (P

Tacrolimus is an immunosuppressive agent that can show a wide variety of neurological side effects, including leukoencephalopathy. The prognosis is usually good with complete neurological recovery after dose reduction or cessation. However, differentiation from progressive multifocal leukoencephalopathy can sometimes be difficult. We describe the case of a 52-year-old gentleman with severe, irreversible tacrolimus-related leukoencephalopathy 4 months after kidney transplantation.

Background Post-transplant diabetes mellitus (PTDM) after renal transplantation is associated with adverse outcome on patient and graft survival. Fasting blood glucose alone will underestimate diabetes and also ignores diagnosis of impaired glucose tolerance (IGT). IGT has a strong correlation with diabetes and cardiovascular risk. Methods In this cross-sectional study, we estimate the prevalence of abnormal glucose metabolism (AGM) using oral glucose tolerance test (OGTT) and identify its predictive factors. Patients who received kidney transplantation in our centre without pre-transplant diabetes were recruited. OGTT was performed in patients with fasting glucose levels between 5.6 and 6.9 mmol/L for at least two occasions 6 months post-transplantation. Results Of 119 patients recruited, 31 had OGTT performed. The prevalence of PTDM, IGT and IFG was 21.8 (26/119)%, 6.7 (8/119)% and 3.4 (4/119)% respectively. Thus the overall prevalence of AGM was 31.9%. Age (P = 0.003), body mass index (P = 0.032), hepatitis B seropositivity status (P = 0.01), CMV infection (P = 0.02) and acute rejection (P = 0.002) were all associated with development of AGM. Using multivariate analysis, only older age at transplant (OR 1.09), history of acute rejection (OR 3.40) and hepatitis B seropositivity (OR 3.13) were significantly associated with the development of AGM. Conclusion AGM is common in our renal transplant recipients.

SUMMARY: Aims: To evaluate the efficacy and safety of a tacrolimus-based immunosuppressive regimen with and without induction therapy using daclizumab in first cadaveric renal transplant recipients. Methods: Since January 2001, we studied the effect of daclizumab in a non-randomized and prospective study of 36 sequential first cadaveric renal transplant recipients. They were compared with a historical control group of 21 sequential first cadaveric renal transplant recipients without induction therapy. All patients received tacrolimus, azathioprine and corticosteroids as concomitant immunosuppressive therapy. Daclizumab was given at 1 mg/kg infusion 2 h before transplantation and then every 14 days for four more doses. Outcomes measured included incidence of acute rejection, patient survival, graft survival, annualized change in creatinine clearance (CrCl), cardiovascular risk profile, infection and malignancy. Results: Fewer biopsy proven acute rejections were observed in the induction treatment group: 11.1% (4/36) versus 19% (4/21) but the rejection free survival was similar (P = 0.37). The patient survival and graft survival were comparable. The renal function was similar in both groups. There were also no significant difference in infection, malignancy and cardiovascular risk profile in both groups. Conclusion: Adding daclizumab to a tacrolimus-based therapy is safe but cannot further improve clinical efficacy.

To evaluate the prevalence of chronic kidney disease (CKD) in Chinese HIV-infected population.This was a cross-sectional point prevalence study. All Chinese HIV-infected patients who were followed up in a tertiary referral center in Hong Kong were recruited. Spot urine was saved for each patient to calculate urine protein-to-creatinine ratio (urine P/Cr). Those with urine P/Cr0.3 would have 24-h urine collection to determine the exact amount of proteinuria. Glomerular filtration rate (GFR) was estimated using MDRD formula. CKD was defined as GFR60 ml/min/1.73 m2 and/or urine P/Cr0.3. Baseline demographic and clinical data were extracted from patients' records.In total 322 patients were recruited. The mean age was 45.2 +/- 11.7 years. The duration of follow up was 6.0 +/- 4.0 years. There were 264 male and 58 female patients. The prevalence of hypertension, diabetes mellitus and CKD were 7.4%, 10.6% and 16.8%, respectively. Eighteen patients (5.6%) had GFR60 ml/min/1.73 m2 while 44 patients (13.7%) had spot urine P/Cr0.3. Among those with urine P/Cr0.3, 38 patients had 24-h urine collection. Using univariate analysis, CKD was found to be significantly (P0.05) associated with age, hypertension, diabetes, use of indinavir, lower CD4 count and peak viral load. Multivariate logistic regression revealed older age (P0.001), lower CD4 count (P = 0.02) and use of indinavir therapy (P = 0.04) were associated with development of CKD.CKD is prevalent in Chinese HIV-infected patients. Patients with CKD were more likely to be older, associated with use of indinavir and CD4 nadir less than 100 cells/mul.

Sir,Epstein–Barr virus (EBV) associated leiomyosarcoma is an uncommon neoplasm that occurs exclusively in the settings of human immunodeficiency virus (HIV) infection,1–3 transplant recipients,4–9 ...

Background The impact of small solute clearance on patient survival in continuous ambulatory peritoneal dialysis (CAPD) is not yet solidified. Previously, we demonstrated that CAPD using small volume (6 L) daily exchanges provides adequate dialysis for most Asian patients. Methods We conducted a prospective, long-term observational study to determine the optimal dialysis adequacy that may provide better patient survival for Asian patients who receive small-volume CAPD. We recruited 294 patients. The initial CAPD regime was 3 x 2-L exchanges daily. The same regime was maintained unless there was significant loss of ultrafiltration or fluid retention despite the use of hypertonic dialysate. Results Median study period was 38.9 (range 5 – 76.5) months, with 81% and 27% of patients remaining in the study at 24 and 48 months respectively. The overall survival rates at 2 and 4 years were 94.0% and 74.8% respectively. Our long-term data revealed that survival rate was related to Kt/V values. Survival rates were significantly higher for patients with total Kt/V > 2.0 than for patients with Kt/V < 1.7 ( p = 0.02). The former group had lower body mass index and higher residual renal function and peritoneal Kt/V than the latter group. On analysis using Cox proportional hazards regression models, cardiovascular disease (CVD), lower urine volume, and higher body mass index were independent predictors of mortality. Patients with higher renal Kt/V had a significantly lower risk of mortality (RR = 0.018, p = 0.01) after adjusting for the effects of CVD and diabetes mellitus. Conclusion Our data recommend that 1.7 be the minimal target for total Kt/V in patients on long-term CAPD. Patients with high body mass index, low residual urine volume, and significant CVD need close monitoring.

Background and MethodsMeasurement of whole blood tacrolimus level by high performance liquid chromatography tandem mass spectrometry (HPLC-MS/MS) was compared with that measured by microparticle enzyme immunoassay (MEIA) in 30 renal transplant recipients.ResultsWhole blood tacrolimus concentrations measured by HPLC-MS/MS were significantly lower than those measured by MEIA, with a median difference (interquartile range, IQR) of −0.40 μg/L (2.03 μg/L) (p < 0.0005). MEIA overestimated tacrolimus concentrations by a median (IQR) of 5.04% (19.5%). There was good correlation between the two methods (r = 0.94; p < 0.0005). The Passing-Bablok regression equation was: HPLC-MS/MS (μg/L) = 0.96 (95% confidence interval, CI, 0.91-1.00) × MEIA (μg/L) −0.02 (95% CI, −0.40-0.46). Bland-Altman analysis showed that the 95% limits of agreement were 2.98 to −4.10 μg/L. The 12-hour area under the concentration curve (AUC12) of tacrolimus derived using the two-point sampling equation with tacrolimus concentrations measured by HPLC-MS/MS was compared with that measured by MEIA. The AUC12 values calculated by the two methods were highly correlated (r = 0.90; p < 0.0005). The mean difference between the AUC12 values was 3.4 ± 11.6 hr.μg/L, and the mean percentage difference was 2.6 ± 11.4%, both of which were not statistically significant.ConclusionFor tacrolimus concentrations within the recommended therapeutic range, the concentration measured by HPLC-MS/MS was statistically significantly lower than that measured by MEIA, but the difference was not clinically significant. Introduction of the more specific HPLC-MS/MS method does not require adjustment in the recommended tacrolimus trough concentration or the AUC12 level estimated by our abbreviated regression equation.

Background Osteopenia and bone loss are major complications after renal transplantation. Clinical data in the Asian population are lacking. Methods We prospectively studied the changes in bone mineral density (BMD) of 31 cadaveric renal allograft recipients using dual energy X-ray absorptiometry (DEXA) within 2 weeks after transplantation and then at 1 year post transplantation. Results One year after transplantation, the mean bone mineral reduction at the lumbar spine was 6.3 ± 4.9% (osteopenia in 32.2% of patients and osteoporosis in 6.5%). For the femoral neck, there was a 4.3 ± 5.5% decrease in BMD (osteopenia in 38.7% of patients and osteoporosis in 3.2%). For total hip, there was a 4.8 ± 3.6% decrease in BMD (osteopenia in 25.8% of patients and osteoporosis in 3.2%). Decrease in BMD at the femoral neck and total hip correlated with the cumulative dosage of steroid (femoral neck: r = −0.36, p = 0.048; total hip: r = −0.4, p = 0.026). Compared with patients without rejection, patients with a history of acute rejection received higher cumulative doses of steroid, and sustained a greater decrease in BMD at the femoral neck and total hip: median percentage (interquartile range) decrease in femoral neck BMD was 7.7% (−12.8 to −6.3) versus 4.4% (−6.5 to −0.065), p = 0.016; decrease in total hip BMD was 8.2% (−11.22 to −6.7) versus 3.4% (−6.92 to −1.4), p = 0.03. There was no correlation between BMD changes and cumulative dosage of cyclosporine A, azathioprine, tacrolimus, duration of dialysis, changes in body mass index, creatinine or parathyroid hormone levels. Conclusion Reduction of BMD after renal transplantation is common. The cumulative dosage of prednisolone correlates with bone loss at the femoral neck and total hip. Patients with a history of acute rejection suffered from more bone loss after 1 year compared to patients without rejection.

We report the case of a female patient with human immunodeficiency virus (HIV) receiving highly active antiretroviral therapy (HAART) who developed end-stage renal failure requiring continuous ambulatory peritoneal dialysis (CAPD). She experienced polymicrobial CAPD-associated peritonitis and died about 1 month after initiation of CAPD. Although CAPD is an effective dialytic therapy, infection remains its Achilles' heel. The potential risk of HIV transmission by peritoneal dialysis effluent should be noted. HAART dose should be adjusted according to renal function to minimize adverse drug effects.

Background Usage of cyclosporin (the Hong Kong Hospital Authority's single largest item of drug expenditure) continues to increase, mainly due to increasing numbers of renal allograft patients taking it as long-term antirejection therapy. Diltiazem, an antihypertensive agent, interferes with the first pass extraction of oral cyclosporin, thus serving to conserve its dosage. Aims In renal transplant patients, to assess whether diltiazem co-treatment could achieve worthwhile dosage conservation of Neoral® (a relatively new microemulsified cyclosporin formulation), safely. Methods A randomized, placebo-controlled, double-blind clinical trial was undertaken at three local hospitals. Renal transplant recipients receiving Neoral® as prophylactic immunosuppression were randomized to two treatment arms. Active treatment consisted of diltiazem tablets 30 or 60 mg twice daily for patients weighing

We report herein two patients who presented with bilateral hydronephrosis attributed to lupus cystitis with two totally different clinical consequences. Both had systemic lupus erythematosus complicated with nephritis and nephrotic range of proteinuria. During the investigation of their proteinuria, they were noted to have a dilated collecting system. The site of obstruction was localized at the vesicoureteric obstruction by excretory urogram. The hydronephrosis resolved spontaneously in one patient before the commencement of any medical therapy, whereas in the other patient, there was no response to steroid therapy. Augmentation enterocystoplasty was thus performed. Reversible hydronephrosis caused by vesicoureteric junction obstruction in patients with systemic lupus erythematosus has been reported. Immune-mediated vasculitis involving the ureter and bladder mucosa was the postulated pathogenic mechanism. The condition usually showed good clinical response upon immunosuppressive therapy. We conclude that systemic lupus erythematosus-associated interstitial cystitis can have a spectrum of different prognoses.

Summary Alcoholic ketoacidosis is an important cause of double gap metabolic acidosis. The differential diagnosis includes toxic alcohol (such as methanol and ethylene glycol) ingestion. We report a patient with alcoholic ketoacidosis. He had a history of excessive alcohol intake and presented with repeated vomiting, loss of appetite and hypovolemic shock. Laboratory results revealed high anion gap metabolic acidosis, increased serum osmolal gap, ketonemia, lactic acidosis and acute renal failure. He was successfully treated with hemodialysis, and hydration with glucose and thiamine. 酒精性代謝酮酸中毒是雙重間隙代謝性酸中毒的重要成因，其鑑別診斷為有毒酒精(如甲醇、乙二醇)的攝入。以下個案是一位酒精性代謝酮酸中毒患者，具過度攝入酒精的習慣，入院時呈現重複嘔吐、無食慾、及低血容積性休克。化驗結果顯示高陰離子間隙之代謝性酸中毒、高血清滲透間隙、酮血症、乳酸中毒、及急性腎衰竭。在接受血液透析、合併葡萄糖與硫胺素補充療法後，病人反應良好。

Objective Superior patient survival on continuous ambulatory peritoneal dialysis (CAPD) with 3 x 2-L exchanges has been reported from Hong Kong. This study examined the relationship between indices of dialysis adequacy and nutrition and patient survival on CAPD in Hong Kong. Design A cross-sectional study on prevalent CAPD patients. Patients were assessed for indices of dialysis adequacy and nutritional status with a composite nutritional index (CNI). Patients were then followed for 24 months. Survival data were analyzed according to adequacy indices and nutritional status. Setting All prevalent CAPD patients in nine dialysis centers in Hong Kong as of 1 April 1996. Main Outcome Measure Mortality. Results 937 patients were assessed: 68.2% were using 3 x 2-L exchanges per day; mean age was 54.6 ± 13 years. Mean total Kt/V was 1.83 ± 0.42 and total creatinine clearance was 55.6 ± 19.5 L/week/1.73 m2. 19% of patients were moderately to severely malnourished according to the CNI. There was no significant correlation between indices of adequacy and serum albumin or CNI. The 1- and 2-year patient survival from the time of assessment was 90.9% and 79.8%. There was a trend toward better survival in patients with Kt/V greater than 2.0, but it was not statistically significant. Peritoneal Kt/V did not impact survival in anuric patients. Malnourished patients had poorer survival than patients who were better nourished ( p = 0.0259). After adjusting for age and diabetes, CNI was predictive of mortality but Kt/V and creatinine clearance were not. Conclusions This study demonstrates the importance of nutritional status over adequacy indices in predicting patient survival. There was a lack of correlation between nutritional status and conventional indices of dialysis adequacy.

We report a gentleman suffering from purpuric skin rash mimicking vasculitis associated with impaired renal function. He was found to have positive cytoplasmic antineutrophil cytoplasmic antibodies (c-ANCA) in the serum, but the test for anti-proteinase 3 antibodies (PR3-ANCA) was negative. Blood culture and echocardiogram confirmed the diagnosis of Streptococcal Oralis infective endocarditis. The renal biopsy did not show any features of vasculitis or crescentic glomerulonephritis. His skin rash subsided and renal function normalized after treatment of the infective endocarditis with appropriate antibiotics. Therefore, we conclude that for those patients presenting with features of vasculitis and a positive ANCA test, other causes, such as infective endocarditis, should be considered.

The effects of pH of conventional lactate-based peritoneal dialysis solutions (LPDS) on the net ultrafiltration and solute removal in intermittent peritoneal dialysis (IPD) were studied. Conventional LPDS is rendered acidic to minimize the caramelization of glucose. Fifteen stable uremic patients were put on weekly IPD with a conventional 1.5% LPDS (hourly 2 L cycle for 60 L). pH of the LPDS was raised to around 7.4 by the addition of 1.4 mmol sodium hydroxide (1 M solution) to each litre of LPDS under an aseptic condition before the commencement of dialysis. A similar study was performed with the same 15 patients but the sodium hydroxide was replaced by a similar amount of sodium chloride. The net ultrafiltration, removal of urea, creatinine, sodium, and potassium were determined and compared. Serum levels of urea, creatinine, sodium, and potassium were also measured before and after the IPD. No peritonitis was encountered during the study. The net ultrafiltration volume in the euvolemic LPDS group was greater than the acidic LPDS group although this was not statistically significant [(mean ±SD) 2526 ±910 mL and 2186 ±750 mL, respectively, p = 0.21]. The clearances of urea, creatinine, sodium, and potassium showed no differences in both groups. We concluded that the ultrafiltration, clearances of urea, creatinine, sodium, and potassium in IPD were not affected by the pH of a LPDS from our results and a larger study is warranted to see the effect of dialysate pH on the peritoneal membrane.

We report a renal transplant patient who suffered from disseminated nocardiosis after empirical tacrolimus rescue therapy for chronic allograft rejection. The nocardiosis presented initially as only mildly tender subcutaneous calf nodules without any other signs of inflammation nor constitutional upset, which later spread to the lung and brain causing bronchopneumonia and brain abscesses. The risk factors for nocardial infection in our patient include the use of potent immunosuppressive agents such as tacrolimus, poorly controlled diabetes mellitus and kidney dysfunction. She responded well to combination antibiotic therapy comprising parenteral meropenem, cefotaxime and oral minocycline. We conclude that in transplant recipients, especially those receiving newer and more potent immunosuppressive agents like tacrolimus, nocardial infection can present as apparently ‘cold’ subcutaneous nodules without any systemic upset. An associated brain lesion should be excluded even in patients without neurological symptoms.

A renal allograft recipient receiving triple immunosuppressive therapy developed spontaneous allograft rupture 5 days after her second cadaveric renal transplantation. Renal biopsy showed interstitial edema with severe acute tubular necrosis (ATN). There was no evidence of acute rejection or renal vein thrombosis. The ruptured renal graft was salvaged by an aggressive fluid resuscitation therapy and surgical hemostasis. The renal function was satisfactory on discharge. We conclude that renal allograft rupture can be the result of interstitial edema solely attributed to ATN in the absence of graft rejection. The ruptured graft kidney is potentially salvageable for those patients whose hemodynamic status can be stabilized by appropriate supportive therapy.

Objective To compare the therapeutic efficacy of daily oral levofloxacin plus intermittent intraperitoneal (IP) vancomycin (group 1) versus daily IP netromycin and intermittent IP vancomycin (group 2) in the primary treatment of peritonitis complicating continuous ambulatory peritoneal dialysis (CAPD). Design A randomized multicenter prospective openlabel comparative clinical study. Setting University and Hospital Authority hospitals in Hong Kong. Patients All CAPD patients who developed bacterial or culture-negative peritonitis beyond 28 days of a previous episode and without evidence of septicemia, associated tunnel infection, or known sensitivity to trial medications were accepted into the clinical trial. Results A total of 101 patients entered the trial. The primary cure rate was 74.5% for group 1 and 73.6% for group 2. Baseline culture results appeared to influence the clinical outcome: the primary cure rate for culture negative, gram-positive, and gram-negative episodes was 83.3%, 78.6%, and 42.9% for group 1 and 69.1 %, 76.9%, and 71.3% for group 2, respectively. The primary cure rate also varied considerably among individual centers and was particularly noticeable in group 1. In the latter group, it correlated closely with in vitro levofloxacin resistance which in turn correlated closely with previous exposure to fiuoroquinolones. Conclusion Oral levofloxacin in combination with intermittent IP vancomycin has comparable efficacy to IP netromycin combined with intermittent IP vancomycin as primary treatment in CAPD peritonitis, but is simpler and more cost-effective to administer. It may be recommended as primary therapy in centers with relatively low exposure and, therefore, low background resistance to fiuoroquinolones.

In 22 hemodialysis patients, during a dialysis session, the solute removal index (SRI) for urea obtained from the use of a partial spent dialysate collection method was compared with that derived from the use of a total spent dialysate collection technique. The partial spent dialysate collection method was used to harvest a small representative sample of the total spent dialysate. The volumes of spent dialysate collected by the partial and the total spent dialysate collection methods were 1.7 +/- 0.4 L and 129.6 +/- 15.3 L, respectively. The total amount of urea nitrogen removed by dialysis as estimated by the partial spent dialysate collection method was similar to that determined by the total spent dialysate collection approach. As a result, the SRI value for urea obtained by the partial spent dialysate collection method (namely, 63% +/- 8%) correlated very well (r = 0.95, P < 0.001) with that derived by the total spent dialysate collection technique (namely, 62% +/- 8%). Our data suggest that it is feasible to use a simple partial spent dialysate collection method to obtain SRI results in patients treated with hemodialysis.

OBJECTIVE: Although peritoneal dialysis (PD) is a widely accepted form of renal replacement therapy, concerns remain regarding the bioincompatible nature of standard PD fluid (PDF). Short-term studies of new biocompatible PDFs low in glucose degradation products (GDPs) reveal divergent results with respect to peritoneal integrity. METHODS: We studied 125 patients on maintenance PD who were assigned, by simple randomization, to receive either conventional or low-GDP PDF at PD initiation. Parameters of dialysis adequacy and peritoneal transport of small solutes were determined at initiation and after a period of maintenance PD at the time when serum and overnight effluent dialysate were simultaneously collected and assayed for various cytokines, chemokines, adipokines, and cardiac biomarkers. All patients were further followed prospectively for an average of 15 months from the day of serum and effluent collection to determine patient survival and cardiovascular events. RESULTS: Patients treated with conventional or low-GDP PDF were matched for sex, age, duration of dialysis, dialysis adequacy, and incidence of cardiovascular disease or diabetes. After an average of 2.3 years of PD treatment, the weekly total and peritoneal creatinine clearance, and the total and peritoneal Kt/V were comparable in the groups. However, urine output was higher in patients using low-GDP PDF despite there having been no difference between the groups at PD initiation. Patients using low-GDP PDF also experienced a slower rate of decline of residual glomerular filtration and urine output than did patients on conventional PDF. Compared with serum concentrations, effluent concentrations of tumor necrosis factor alpha, hepatocyte growth factor, macrophage migration inhibitory factor, interleukins 8 and 6, C-reactive protein, and leptin were found to be higher in both groups of patients after long-term PD, suggesting that the peritoneal cavity was the major source of those mediators. Compared with patients on low-GDP PDF, patients on conventional fluid showed elevated leptin and reduced adiponectin levels in serum and effluent. The effluent concentration of interleukin 8 was significantly lower in patients using low-GDP PDF. The survival rate and incidence of cardiovascular complications did not differ between these groups after maintenance PD for an average of 3.6 years. CONCLUSIONS: It appears that low-GDP PDF results in an improvement of local peritoneal homeostasis through a reduction of chronic inflammatory status in the peritoneum., link_to_OA_fulltext

We report the first case of Streptococcus gordonii-related continuous ambulatory peritoneal dialysis (CAPD) peritonitis. He is a 69-year-old man with end-stage renal failure due to chronic glomerulonephritis who had been put on CAPD for 1 year. He was successfully treated with a 2-week course of cefazolin. This case highlights the emerging threat that S. gordonii can be the source of infection in patients on CAPD.

Nocturnal home haemodialysis (NHHD) was started in Hong Kong in 2006. The experience of 1 year of NHHD with an alternate night schedule in two local centres is reported.The clinical parameters of 14 patients who had completed 1 year of NHHD were retrospectively analyzed. All patients were receiving an alternate night schedule (3.5 sessions/week) for 6-8 h/session.After 1 year of NHHD, haemoglobin levels increased from 9.6±1.6 g/dL before NHHD to 11.4±2.2 g/dL (P0.05) despite a reduction in erythropoietin dose requirement from 120.6±44.3 to 59.4±74.6 U/kg/week (P0.05). Four patients (29%) were able to stop taking erythropoietin after NHHD. Serum phosphate levels reduced from 2.33±0.41 to 1.59±0.29 mmol/L (P0.01) and calcium phosphate product decreased from 5.29±0.96 to 3.74±0.90 mmol2/L2 (P0.01). Phosphate binder dose was greatly reduced and eight patients (67%) were able to stop taking phosphate binders. The number of antihypertensive medications tended to reduced from 2.5±1.3 to 1.6±1.5 (P=0.067) with four patients (29%) able to stop antihypertensives. Left ventricular mass index decreased from 186±62 to 168±60 g/m2 (P=0.463) although this was not statistically significant. Weekly spKt/V during conventional haemodialysis was 3.63±0.95 while that during NHHD was three times higher at 11.09±6.44 (P0.01). The quality of life indexes also showed improvement.This 1 year experience of alternate night NHHD demonstrates benefits in terms of anaemia control, erythropoietin requirement, serum phosphate and calcium phosphate product reduction, blood pressure control, haemodialysis adequacy and quality of life. NHHD with an alternate night schedule is a promising dialytic therapy for patients receiving chronic haemodialysis in this locality.

Aim: There is limited data concerning the impact of recipient body mass index (BMI) on graft outcome in Asian renal transplant recipients. The aim of this study is to identify whether obesity (BMI 25 kg/m 2 ) and overweight (BMI 23 kg/m 2 ) can predict graft outcome. Methods: This is a single-centre retrospective study. All patients who received kidney transplantation between 1997 and 2005 were recruited. Patients were categorized according to two different designated BMI cut-off values. Results: One hundred and thirty-one patients were recruited with a median follow-up duration of 73 months. If a BMI cut-off value of 25 kg/m 2 was used, 86.3% patients were classified as non-obese and 13.7% as obese. Obesity was significantly associated with poor renal graft function and decreased patient and graft survival. On the other hand, 34.3% patients were classified as overweight and 65.7% patients as normal if a BMI cut-off value of 23 kg/m 2 was used. Overweight was significantly associated with a lower glomerular filtration rate only. Cox regression analysis showed that obesity (odds ratio (OR) = 3.09), acute rejection (OR = 5.68), pre-transplant diabetes mellitus (OR = 3.21) and age of recipient (OR = 1.06) were all significant independent risk factors associated with graft failure. Conclusion: Recipient BMI 25 kg/m 2 is a significant predictive factor for long-term renal graft outcome in the Asian population. With the introduction of new immunosuppressive agents, the risk of acute rejection in renal transplantation has been significantly reduced. Much of the focus nowadays has shifted to prolong graft survival. Obesity had been linked with an increased incidence of proteinuria, hypertension, hyperlipidaemia, diabetes mellitus (DM) and focal segmental glomerulosclerosis (FSGS) in the general population. 1 On the other hand, the impact of recipient obesity on patient and renal allograft survival is controversial. Higher body mass index (BMI) has been shown to be associated with increased risk for graft failure and patient death among white patients with end-stage renal disease who undergo renal transplantation. 2 The strong association between BMI and transplant outcome was supported in a retrospective analysis of 51 927 adult renal transplant recipients. Meier-Kriesche et al. showed that both abnormally low and abnormally high BMI are risk factors for decreased patient and graft survival, independent of most of the known

We report a female patient suffering from cytomegalovirus (CMV)-induced tubulointerstitial nephritis in a renal allograft 70 days after a cadaveric renal transplantation. CMV-induced renal allograft injury reported in the literature mainly related to immune-mediated mechanisms. In our patient, acute tubulointerstitial nephritis, associated with histological evidence of CMV infection, was demonstrated in the renal allograft biopsy. There were no histological features of allograft rejection, cyclosporin nephrotoxicity nor ‘CMV glomerulopathy’. She was successfully treated by foscarnet therapy and a reduction in immunosuppression. Her renal function returned to baseline afterwards.

SUMMARY: nep_1155 758..763 Aim: The first prospective, randomized trial with paired kidney analysis was conducted to compare the efficacy and safety of tacrolimus with cyclosporine-based immunosuppressive therapy in renal transplant recipients. This paper reports the long-term follow-up results of the authors’ previously published study, with the main focus on graft survival and renal function. Methods: Chinese patients transplanted in our centre between June 1998 and June 2005 with their first deceased renal transplant were included. Patients were included if both kidneys were received by the authors’ centre, thus allowing a paired analysis. Patients were randomized to receive triple immunosuppressive therapy with either tacrolimus or Neoral cyclosporine, concomitantly with prednisolone and azathioprine therapy. Results: Seventy-six patients received cadaveric kidneys from 38 donors. Each pair of kidneys was randomly assigned to a separate group (38 subjects/group). The mean follow-up duration was 6.1 1 1.8 years. The mean calculated creatinine clearance was significantly higher in patients receiving tacrolimus-based therapy. The rate of biopsy-proven acute rejection was lower in the tacrolimus group (18.4% vs 42.1%, P = 0.03). The patient and graft survival were comparable in both treatment arms. Significantly fewer patients on tacrolimus-based therapy developed hypercholesterolaemia (P = 0.05). However, there was no significant difference in the development of post-transplant diabetes mellitus, hypertension, opportunistic infection and malignancy between both groups. Conclusion: Using the immunosuppressive regimen, tacrolimus-based therapy provided adequate immunosuppression with better renal function and less acute rejection, as compared with cyclosporine-based therapy.

Dried blood spot (DBS) sampling and high-performance liquid chromatography tandem-mass spectrometry have been developed in monitoring tacrolimus levels. Our center favors the use of limited sampling strategy and abbreviated formula to estimate the area under concentration-time curve (AUC(0-12)). However, it is inconvenient for patients because they have to wait in the center for blood sampling. We investigated the application of DBS method in tacrolimus level monitoring using limited sampling strategy and abbreviated AUC estimation approach. Duplicate venous samples were obtained at each time point (C(0), C(2), and C(4)). To determine the stability of blood samples, one venous sample was sent to our laboratory immediately. The other duplicate venous samples, together with simultaneous fingerprick blood samples, were sent to the University of Maastricht in the Netherlands. Thirty six patients were recruited and 108 sets of blood samples were collected. There was a highly significant relationship between AUC(0-12), estimated from venous blood samples, and fingerprick blood samples (r(2) = 0.96, P < 0.0001). Moreover, there was an excellent correlation between whole blood venous tacrolimus levels in the two centers (r(2) = 0.97; P < 0.0001). The blood samples were stable after long-distance transport. DBS sampling can be used in centers using limited sampling and abbreviated AUC(0-12) strategy as drug monitoring.

We report an unusual case of veno-caliceal fistula that developed because of high ureteric pressure caused by graft ureteric stricture after kidney transplantation in a 60-year-old patient. We further confirmed its presence with radiological images. Recirculation of creatinine and other uremic toxins resulted in a biochemical picture of renal failure in the presence of normal kidney function, confirmed by normal scintigraphy findings. Drainage of the pelvi-caliceal system could not be assessed accurately by means of diuretic renogram using technetium-99m diethylenetriaminepentaacetic acid with frusemide because of the rapid clearance of tracer activity from the system in the presence of a veno-caliceal fistula. The patient's renal function improved rapidly after interrupting urine recirculation by using percutaneous drainage, confirming "pseudo renal failure" as the cause of his persistently increased serum creatinine concentration. The ureter was re-implanted later, and the veno-caliceal fistula was not seen in the nephrostogram after the operation. He remains well with stable renal function at 3 years' follow-up. Clinicians should exercise judgment when evaluating patients with allograft dysfunction, especially when the investigation and clinical findings show contradicting results.

Few studies used paired kidneys for comparison between tacrolimus and cyclosporine in renal transplantation. Most of the published data used whole blood trough levels for drug monitoring. However, the use of limited sampling strategy and abbreviated formula to estimate the 12-h area under concentration-time curve (AUC(0-12)) allowed better prediction of drug exposure. Sixty-six first cadaveric renal transplant recipients receiving paired kidneys were randomized to receive either tacrolimus-based (n = 33) or cyclosporine microemulsion (Neoral)-based therapies (n = 33). Abbreviated AUC(0-12) was used for drug monitoring and dose titration. Mean follow-up duration was 2.8 +/- 2 years. The patient and graft survival were comparable. Fewer incidence of acute rejection was observed in tacrolimus group (15% vs. 27.3%) though the difference was not significant (P = 0.23). The absolute value and the rate of decline of creatinine clearance were both significantly better in tacrolimus-treated patients. Prevalence of hypertension, post-transplant diabetes mellitus, infection, and malignancy were similar in both groups. Prevalence of hypercholesterolemia (11/33 vs. 4/33) and gum hypertrophy (6/33 vs. 1/33) was more common in cyclosporine-treated patients (P = 0.04 in both parameters). This was the first prospective, randomized study with paired kidney analysis showing the renal function was significantly better in tacrolimus-treated patients than in cyclosporine-treated patients.

Tacrolimus has a narrow therapeutic window and a wide interindividual variation in its pharmacokinetics. The cytochrome P450 3A (CYP3A) and the ATP-binding cassette B1 (ABCB1) genes play an important role in the tacrolimus disposition. Therefore, the aim of this study was to evaluate whether CYP3A and ABCB1 polymorphisms are associated with the area under the time concentration curve (AUC0–12) calculated using a two time point sample strategy. The CYP3A and ABCB1 genotypes were determined by real-time polymerase chain reaction (RT-PCR) fluorescence resonance energy transfer (FRET) assays in 103 Chinese renal transplant recipients and consequently related to their dose-normalized (dn)AUC0–12. A significant allele-dependent effect (Kruskal–Wallis; p 0–12. Multiple regression analysis showed that the CYP3A5*3 polymorphism is the most significant independent variable and explained 35% of the dose requirement variability in relation to tacrolimus use. Regarding the ABCB1 G2677T/A and C3435T polymorphisms, a trend was observed between the different genotypes and the dnAUC0–12. In conclusion, the CYP3A5*3 polymorphism may be an important factor in determining the dose requirement for tacrolimus and genotyping can help determine the initial daily dose required by individual patients for adequate immunosuppression.

Background: This prospective randomized study aims to assess the effectiveness of intradermal (ID) hepatitis B virus (HBV) vaccination in patients on continuous ambulatory peritoneal dialysis (CAPD) therapy. Methods: Sixty patients were randomly divided into 2 groups. The ID group was treated with 5 μg of recombinant HBV vaccine intradermally every week for a total of 10 doses, and the intramuscular (IM) group, with 20 μg intramuscularly at 0, 1, and 6 months. Results: ID HBV vaccination was associated with a greater seroconversion rate (81.5% versus 62.1%), although the difference did not reach statistical significance ( P = 0.14). The cumulative seroconversion rate was significantly greater with ID vaccination by 6 months after the first vaccine dose ( P = 0.03). There was no difference between the 2 groups in time required to convert, peak antibody to HBV surface antigen (anti-HBs), and proportion of patients with anti-HBs levels maintained at greater than 10 mIU/mL or 100 mIU/mL in the 2-year observation period. Although the ID group achieved a peak anti-HBs titer significantly earlier than the IM group ( P = 0.001), we found a significant trend for the ID group to achieve a lower peak anti-HBs titer (chi-square test for trend, P = 0.005). The incidence of local reactions was significantly greater with ID immunization; however, reactions were mild and transient. Conclusion: ID HBV vaccination is associated with significant improvement in seroconversion rate in CAPD patients at 6 months, but this difference diminishes at 2 years. Larger studies are warranted to confirm this finding.

Usage of cyclosporin (the Hong Kong Hospital Authority's single largest item of drug expenditure) continues to increase, mainly due to increasing numbers of renal allograft patients taking it as long-term antirejection therapy. Diltiazem, an antihypertensive agent, interferes with the first pass extraction of oral cyclosporin, thus serving to conserve its dosage.In renal transplant patients, to assess whether diltiazem co-treatment could achieve worthwhile dosage conservation of Neoral (a relatively new microemulsified cyclosporin formulation), safely.A randomized, placebo-controlled, double-blind clinical trial was undertaken at three local hospitals. Renal transplant recipients receiving Neoral as prophylactic immunosuppression were randomized to two treatment arms. Active treatment consisted of diltiazem tablets 30 or 60 mg twice daily for patients weighing60 oror= 60 kg, respectively. One hundred and ten eligible patients gave their informed consent, and were followed up for at least six months. The mean difference in the dollar cost in the sixth month was the primary outcome. Secondary/ancillary outcomes included changes in cyclosporin dosage and blood level, and untoward clinical events including rejection. Outcomes were evaluated by intention to treat analyses.During weeks 23-26 (sixth month) post randomization, diltiazem co-treatment yielded an estimated average cost saving per patient on drugs of 15%[the 95% confidence interval (CI) of the difference being HK dollars 609 +/- 517 or pound 50 +/- 42], with no apparent excess of untoward or adverse events, complications, hospitalization, outpatient visits, or inferior quality of life.This diltiazem co-treatment regime applied to the nearly 1800 surviving renal allograft patients followed up in Hospital Authority hospitals could have saved approximately HK dollars 14.3 million ( pound 1.17 million) annually, without adverse sequelae.

Effect of Kt/V on survival and clinical outcome in CAPD patients in a randomized prospective study.BackgroundThere has been a lack of randomized control study on the effect of Kt/V on patient outcome. This interventional study was designed to examine the effect of Kt/V on continuous ambulatory peritoneal dialysis (CAPD) patients' clinical outcome and nutritional status in a randomized prospective manner.MethodA total of 320 new CAPD patients with baseline renal Kt/V 2.0. Kt/V and nutritional status were assessed every 6months and dialysis prescription adjusted accordingly. Nutritional assessment included serum albumin and composite nutritional index (CNI). Patients were allowed to withdraw at the discretion of their physicians or themselves.ResultsTotal Kt/V were significantly different between groups (P = 0.000) and the difference was contributed by peritoneal Kt/V only. The overall 2-year patient survival was 84.9%. There was no statistical difference in patient survival among the three groups (2-year survival in group A, 87.3%; group B, 86.1%; and group C, 81.5%). However, there were more patients withdrawn by physicians in group A (group A, 16; group B, 7; and group C, 6; P = 0.023). Total Kt/V or Kt did not significantly affect survival after adjustment to age and diabetes. There was no difference in serum albumin, CNI scores, and hospitalization rate, but there were more patients in group A requiring erythropoietin (EPO) treatment after 1year.ConclusionPatients with total Kt/V maintained below 1.7 had significantly more clinical problems and severe anemia but there was no difference in outcome demonstrated for patients with Kt/V maintained above 2.0 and between 1.7 and 2.0. We recommended that the minimal target of total Kt/V should be above 1.7.

Frasier syndrome is a rare human developmental disorder classically affecting 46,XY females and leading to male pseudohermaphroditism and chronic renal failure. We describe a family with both 46,XX and 46,XY females affected by the syndrome due to WT1 splice site mutations. The diagnosis of Frasier syndrome in 1 of the children led to the discovery of the syndrome in 2 other siblings, of whom 1 is asymptomatic. Since the mutation was not found in either parents, gonadal mosaicism was suggested. The implication of family screening for WT1 gene mutation in asymptomatic members is also discussed.

This study was done to evaluate the outcome after brachytherapy (BT) given to prevent restenosis after stent insertion for central venous stenosis in patients with ipsilateral hemodialysis arteriovenous fistulas (AVF). Angioplasty and stenting were performed on 9 primary central venous stenoses in 8 patients with AVF followed by BT, delivering Iridium-192 radiation using an afterloading technique. BT was also administered to three patients with five recurrent stenoses at the stent margins. There was no residual stenosis after angioplasty and stenting. Venographic follow-up (77-644 days, mean 272 days) showed no restenosis in seven primary stenoses. New strictures (45%-100%) developed at the stent margin in six veins (five patients). Angioplasty or stenting was performed for five margin stenoses in three patients, followed by a second BT. Residual stenosis before BT was 0-30%. In our venographic follow-up (140-329 days, mean 215 days), three restenoses occurred (35%-100%). All progressed to complete occlusion on later venographic follow-up irrespective of whether BT was given to the stent margin or not. The mean primary and assisted primary patency of the central veins were 359 days and 639 days, respectively. Endovascular irradiation with a noncentering source does not prolong the patency after angioplasty and stenting of central venous stenosis in hemodialysis patients.

This report describes a patient with mixed normal anion gap hyperchloremic metabolic and respiratory acidosis associated with hypokalemia attributed to cough mixture abuse. Metabolic acidosis was likely related to an overdose of ammonium chloride, whereas respiratory acidosis was probably related to the effect of hypokalemia on respiratory muscles, causing hypoventilation. Hypokalemia was caused by a transcellular shift of potassium induced by ephedrine and pseudoephedrine. Both ammonium chloride and ephedrine were probably present in the cough mixture obtained by our patient as an over-the-counter medication. Physicians should be aware of the potential for cough mixture abuse to cause major electrolyte disturbances that may carry the risk for major cardiac arrhythmias, particularly in youth. [copy ] 2001 by the National Kidney Foundation, Inc .

Cefepime is a cephalosporin with a broad spectrum of activity against most gram-positive and gram-negative pathogens. In this study, we attempted to compare the safety and efficacy of cefepime monotherapy against the potentially more toxic combination of vancomycin and netilmicin in the treatment of continuous ambulatory peritoneal dialysis (CAPD)-associated bacterial peritonitis. Eighty-one consecutive CAPD patients who presented with peritonitis from January 1, 1998, to June 30, 2000, were recruited for study. Patients were randomized to be administered either intraperitoneal (IP) cefepime, 1 g once daily (group A), or intravenous vancomycin and netilmicin at conventional doses (group B) for 10 days. Bacterial growth was obtained in 52 episodes (66%), and pathogens identified included gram-positive organisms (30 episodes; 38%), gram-negative organisms (14 episodes; 18%), mixed organisms (2 episodes; 2.5%), and fungus (6 episodes; 8%). Eight patients were excluded after randomization for various reasons (6 patients, fungal peritonitis; 2 patients, wrong diagnoses). Because of the relatively low peritonitis rate after the use of a disconnect system, the sample size of this study was relatively small, giving a power of 0.45. There were no significant differences in primary response rates and cure rates (no relapse >28 days after completion of antibiotic therapy) between both groups of patients (group A versus group B, 82% [32 of 39 patients] versus 85% [29 of 34 patients] and 72% [28 of 39 patients] versus 76% [26 of 34 patients], respectively; P = not significant). No significant side effect was encountered in either group. Total peritonitis-related hospitalizations were 84 patient-days (1, 7, 8, 11, 20, and 37 patient-days) and 115 patient-days (3, 6, 9, 14, 21, 21, and 41 patient-days), whereas total costs per patient cure were estimated to be US $1,039 and US $1,371 in groups A and B, respectively. We conclude that once-daily 1-g IP cefepime monotherapy is a simple, safe, and cost-effective alternative to vancomycin and netilmicin therapy in the treatment of CAPD-associated bacterial peritonitis.

We modified the urea reduction ratio (URR) equation to correct the effects of ultrafiltration and intradialytic urea generation on the delivered dose of hemodialysis: mURR = [1 - (R/(1 + 2 * UF/BW)) + 0.01 * t] × 100% where mURR is modified URR, R is postdialysis plasma urea nitrogen (PUN) to predialysis PUN ratio, UF is ultrafiltrate volume in liters, BW is postdialysis body weight in kilograms, and t is dialysis session length in hours. The equation was validated against 145 hemodialysis treatments on 33 stable maintenance dialysis patients. The mURR values obtained closely predicted single-pool Kt/V (spKt/V) values. In contrast to conventional URR, the range of spKt/V values at each mURR value is narrow over a wide range of UF, t, and R values. Based on the mURR equation, mURR values of 64%, 70%, and 76% are mathematically equivalent to spKt/V levels of 1.0, 1.2, and 1.4, respectively. This equation can be a useful bedside tool to quantify hemodialysis dose.

The area under the concentration time curve (AUC) for oral tacrolimus (FK) may provide a more precise model for FK monitoring after renal transplantation. The purpose of this study is to identify a simple, cost-effective method for predicting FK AUC. FK concentrations were measured at 0, 1, 2, 4, 6, 8, and 12 hours after the morning dose. The predicted AUCs (AUC(p)s) derived from regression equations were used to estimate the actual 12-hour AUCs (AUC(12)s). The relationship between AUC(p) and AUC(12) was validated by determining the coefficient of multiple determination (R(2)), percentage of prediction error (PE%), and percentage of absolute prediction error (APE%). Eighteen stable Oriental renal transplant recipients (9 men, 9 women) with a mean age of 42.6 +/- 6 years and mean body weight of 62.7 +/- 10 kg were recruited for the study. The FK AUC(12), trough, 2-hour, and 4-hour concentrations were 125 +/- 24 h. ng/mL (range, 87.7 to 181.9 h. ng/mL), 6 +/- 1.3 ng/mL, 18.1 +/- 4.7 ng/mL, and 11 +/- 2.4 ng/mL, respectively. Trough FK concentration did not have a significant correlation with AUC(12) (r = 0.34; P = 0.17). AUC(p) obtained by a two-time point regression equation using 2-hour (C2) and 4-hour (C4) FK concentrations: (AUC(P) = 16.2 + 2.4*C2 + 5.9*C4) obtained an R(2), PE%, and APE% of 0.93, -0.2% +/- 5.2% (range, -13% to 9.3%), and 3. 6% +/- 3.7% (range, 0.02% to 13%), respectively. We conclude that a two-point sampling method using C2 and C4 may be a more cost-effective FK monitoring strategy than morning FK trough levels in transplant recipients.

We performed a multicenter, single-blinded, prospective randomized study on the use of a double-bag disconnect system (B) versus a Y-set disconnect system (Y). The peritonitis rate, exit site infection, clinical outcome, and patients' acceptance to the procedure were assessed. A total of 120 new end-stage renal failure patients of three regional hospitals were randomized: 60 each to the B and the Y systems. The results of 60 patients on the B system and 51 on the Y system were analyzable. They were followed up for a median of 16 months. Peritonitis rates for the B and the Y systems were 33.5 and 29.4 patient-months per episode, respectively. Exit site infection rates for the B and Y systems were 17.4 and 16.0 patient-months per episode, respectively. Four catheters were removed in each group. Patients on the B system were hospitalized for 2.1 days per patient per year related to peritonitis and exit site infection, and those on the Y system were hospitalized for 1.2 days. There was no significant difference between the B and Y systems in the incidences of peritonitis (all causes and those due to coagulase-negative staphylococci), exit site infection, and in hospitalization days. However, there was a higher percentage of gram-positive infections in the Y system (52%) than in the B system (32%) and a lower percentage of gram-negative infections in the Y system (16%) than in the B system (32%). Patients on the B system had a better acceptance of the procedure than patients on the Y system, as assessed by a six-item, 10-point questionnaire (total score, 43.1 +/- 10.2 v 37.6 +/- 9.4; P < 0.005 at 1 month; 44.6 +/- 9.1 v 39.8 +/- 8.6; P < 0.01 at 6 months). From this study, it is concluded that the B and Y systems are similar in the incidences of peritonitis and exit site infection, although the B system is better accepted by patients. This is probably the first multicenter randomized study comparing the double-bag and Y-set disconnect system using only new patients who had never used other systems of continuous ambulatory peritoneal dialysis.