Concern about the decreasing use of peritoneal dialysis

Transcription

1 Page 1 of 8 Peritoneal Dialysis International Peritoneal Dialysis International, Vol. 30, pp. doi: /pdi /10 $ Copyright 2010 International Society for Peritoneal Dialysis TIME-DEPENDENT REASONS FOR PERITONEAL DIALYSIS TECHNIQUE FAILURE AND MORTALITY Inna Kolesnyk, 1 Friedo W. Dekker, 2 Elisabeth W. Boeschoten, 3 and Raymond T. Krediet 1 Division of Nephrology, 1 Department of Medicine, Academic Medical Centre, University of Amsterdam; Department of Clinical Epidemiology, 2 Leiden University Medical Centre, Leiden; Hans Mak Institute, 3 Naarden, The Netherlands Background: Peritoneal dialysis (PD) technique failure is high compared to hemodialysis (HD). There is a lack of data on the impact of duration of PD treatment on technique survival and on whether there is a difference in risk factors with respect to early and late failure. The aim of this study was to clarify these issues by performing a timedependent analysis of PD technique and patient survival in a large cohort of incident PD patients. Methods: We analyzed 709 incident PD patients participating in the Netherlands Cooperative Study on the Adequacy of Dialysis (NECOSAD), who started their treatment between 1997 and We compared technique and patient survival on PD in 4 periods of follow-up: within the first 3 months, and after 3 12 months, months, and months of treatment. Cox proportional hazards model was used to analyze survival on PD and technique failure. Risk factors were also identified by comparing patients that were transferred to HD with those that remained on PD. Incidence rates for every cause of dropout for each period of follow-up were calculated to establish their trends with respect to PD treatment duration. Results: There was a significant increase in transplantation rate after the first year of treatment. The rate of switching to HD was highest during the first 3 months and decreased afterward. One-, 2- and 3-year technique survival was 87%, 76%, and 66%, respectively. Age, diabetes, and cardiovascular disease appeared to be risk factors for death on PD or switch to HD: a 1-year increase in age was associated with a relative risk (RR) of PD failure of 1.04 [95% confidence interval (CI) ]; for diabetes, RR of stopping PD after 3 months of treatment increased from 1.8 (95% CI 1.1 3) during the first year to 2.2 (95% CI 1.3 4) after the second year; cardiovascular disease had a major impact in the earliest period (RR 2.5, 95% CI 1.2 5) and had a stable influence further on (RR 2, 95% CI ). Correspondence to: I. Kolesnyk, Nephrology Division, Dialysis Secretariat A01-114, AMC, Meibergdreef 9, 1105 AZ, Amsterdam, The Netherlands. i.kolesnyk@amc.uva.nl Received 19 December 2008; accepted 6 April Loss of 1 ml/minute residual glomerular filtration rate (rgfr) appeared to be a significant predictor of PD failure after 3 months of treatment, but within the first 2 years, RR was 1.1 (95% CI ). Conclusions: In The Netherlands, transplantation is a main reason to stop PD treatment. The incidence of PD technique failure is at its highest during the earliest months after treatment initiation and decreases later due to fewer catheter and abdominal complications as well as less influence of psychosocial factors. Risk factors for PD discontinuation are those responsible for patient survival: age, cardiovascular disease, diabetes, and rgfr. Perit Dial Int 2010; 30:xx xxx epub ahead of print: doi: /pdi KEY WORDS: Technique survival; mortality; time dependency. Concern about the decreasing use of peritoneal dialysis (PD) in many countries has risen over the past few years (1,2). Attempts to explain this process led to the conclusion that many factors are involved; they are complex and vary much from country to country (1). Relatively high technique failure compared to hemodialysis (HD) was always one of the main reasons for PD deprivation: some studies reported a 3-year technique survival slightly higher than 50% (3,4). A number of studies analyzed patient and technique survival trying to identify the risk factors for PD failure (3 10). It has been shown that renal center characteristics such as the number of PD patients impact patient and technique outcome (8,10). Some analyses showed a rise in PD modality success over the past years due to technical progress (4,7,8,11). There is a lack of data on the exact reasons for PD technique failure and whether they differ with respect to treatment duration. The study by Guo and Mujais, which was done in large incident cohorts, suggested that the 1 PDI in Press. Published on February 1, doi: /pdi

2 Peritoneal Dialysis International Page 2 of 8 KOLESNYK et al. MARCH 2010 VOL. 30, NO. 2 PDI first 6 months of PD therapy are critical for technique failure (7). The latest published work by Descoeudres et al. confirmed this finding and reported the importance of early dropout from PD in a single center study (12). More data are needed, however, for better understanding the reasons for technique failure and their correlation with PD duration in order to improve the outcome of PD. We aimed to perform a detailed analysis of technique and patient survival in the PD cohort of the Netherlands Cooperative Study on the Adequacy of Dialysis (NECOSAD). Our main goal was to update and extend earlier published data (3,9) and make a precise analysis of the reasons and risk factors for PD mortality and technique failure with respect to duration of treatment. METHODS PATIENTS AND FOLLOW-UP PERIOD The patients were selected from the NECOSAD database. This database contains data on patients with endstage renal disease attending 38 dialysis centers in The Netherlands. At the start of dialysis, all patients were older than 18 years and had not previously received renal replacement therapy. We were able to include 585 patients that started renal replacement therapy with PD in the period from 1 January 1997 to 1 July Patients that started with HD but were switched to PD within the first 3 months after therapy initiation (n = 124) were also included. Altogether, 709 patients were eligible for the current analysis. After being included, patients were followed for as long as they continued on PD therapy. DATA COLLECTION Demographic data as well as data on comorbidity and primary kidney disease were collected within 1 month prior to the start of dialysis treatment. During followup, data on blood pressure and residual renal function were collected at 3 and 6 months after the start of dialysis. Afterward, data were collected on a half-yearly basis. Primary kidney disease was classified according to the codes of the European Dialysis and Transplant Association European Renal Association Registry. Comorbidity was scored on the basis of Davies Comorbidity Index. Cardiovascular disease was recorded if one of the following conditions was present: angina pectoris, myocardial infarction, congestive heart failure class III IV, peripheral vascular disease, and cerebral vascular accident. Residual renal function was expressed as residual glomerular filtration rate (rgfr) and was calculated as the mean of creatinine and urea clearances, corrected for body surface area and reported as milliliters per minute per 1.73 m 2. ANALYTICAL METHODS The primary end point of the study was failure of PD technique, defined as a permanent switch to HD or death on PD. To identify the risk factors for earlier and later PD failure, we divided the study follow-up into four periods: period I, the first 3 months of PD therapy; period II, after 3 months but within the first year; period III, 2 years of therapy; and period IV, 3 years of PD therapy. We compared data on demography, primary kidney disease, comorbidity, baseline residual renal function, and mean arterial blood pressure between the patients that had switched to HD and those that remained on PD within each period of follow-up. For this comparison, we used standard descriptive statistics: Student s t-test, chi-square test. Reasons for PD dropout within the different periods of PD treatment were compared using the chi-square test. In order to evaluate differences in reasons to stop with PD therapy in a time-dependent way, we calculated the incidence rates for every cause of dropout per 1000 patient-years for every period of follow-up. Statistical analysis of PD technique survival alone and combined with patient survival on PD ( stay on PD ) was performed using several multivariate Cox proportional hazard models. In the analysis of technique survival, the event was permanent switch to HD; transplantation and death were censored observations. In the stay on PD analysis, the events were switch to HD or death on PD; transplantation was a censored observation. In order to identify time-dependent predictors for dropout from PD we performed Cox regression analysis separately for each period of follow-up (0 3 months, 3 12 months, months, and months) and added variables to the model. We constructed the following multivariate models: crude effects of age and gender on survival; effect of diabetes, adjusted for age and gender; effect of cardiovascular disease, adjusted for age and gender; and influence of rgfr, measured at the start of every followup period and adjusted for age, gender, diabetes, and cardiovascular disease. Data on comorbidity were taken at the start of dialysis treatment. All statistical analyses were performed using SPSS statistical software, version 14.0 (SPSS Inc., Chicago, IL, USA). A p value of 0.05 or less was considered statistically significant. 2

3 Page 3 of 8 Peritoneal Dialysis International PDI MARCH 2010 VOL. 30, NO. 2 TECHNIQUE SURVIVAL AND MORTALITY ON PD RESULTS PATIENTS AND BASELINE CHARACTERISTICS All 709 NECOSAD patients that started PD within the first 3 months after initiating renal replacement therapy were included in the study. Figure 1 shows the changes in the cohort during the four predefined periods of follow-up. A comparison of data on demography and baseline clinical factors between the patients transferred to HD and those that remained on PD is shown in Table 1. Within the first 3 months of PD therapy the patients that were transferred to HD were older, more likely to be female, and had a higher comorbidity score compared to those that did not transfer. No differences were observed between the patients that transferred during the second period (3 12 months) and those that did not. In the third period, between 12 and 24 months, those that switched to HD were younger, less likely to be diabetic, and less likely to have cardiovascular disease than those that stayed on PD. After 2 years of PD treatment, the comorbidity score alone appeared to be associated with technique survival. EVENTS For the whole study period, mean follow-up was 28 months (range months); 159 patients died, Figure 1 Patient distribution over four follow-up periods, including transfer from peritoneal dialysis (PD) to hemodialysis (HD) and dropout from the study due to death or transplantation. 226 received a kidney transplant, 102 dropped out from the study for various nonmedical reasons such as refusal to participate or transfer to another center, and 186 were transferred to HD. Infectious complications such as peritonitis and exit-site and tunnel infections were the main reason for transfer to HD, occurring in 77 patients. Underdialysis with ultrafiltration failure was the reason for transfer to HD for 19 patients; abdominal and catheter-related complications occurred in 16 and 19 patients respectively; 55 patients were transferred to HD for either psychosocial (patient s preference, etc.) or unknown reasons; and 36 patients were still being treated with PD at the censoring date. TECHNIQUE FAILURE One-, 2-, and 3-year PD technique survival was 87%, 76%, and 66%, respectively. SURVIVAL ON PD AT DIFFERENT PERIODS OF FOLLOW-UP Figure 2(a) shows differences in the various reasons for PD dropout during the four periods of follow-up. During each period approximately 25% 30% of patients died. Thirteen percent of patients received a kidney transplant during the first 3 months of PD therapy; afterward, 30%, 42%, and 43% of patients were transplanted during the first, second, and third years, respectively. Infectious complications were the reason for transfer to HD for 10% 18% of patients. Catheter problems occurred for 15% and 7% of patients during the first two periods and decreased after the first year to 1% 2%. Abdominal problems were the reason to switch to HD for 5% 7% of patients during the first year and for 2% afterward. During the first two periods, 2% of patients were considered underdialyzed or had problems with ultrafiltration; 4% of patients experienced this complication in the second year and 6% after 2 years of PD. Psychosocial or unknown reasons for PD dropout varied from 20% in first 3-month period to 7% 9% later on. In comparisons of reasons for PD dropout among the four periods of follow-up, the contribution of transplantations increased markedly during the follow-up period. The number of people that received a kidney graft during the first 3 months was significantly lower than in later periods (p = <0.0001). Also, the number of patients transplanted after 3 months but within the first year of PD was lower than that for the next two periods (p 0.04). Similar numbers of transplantations were performed during the second and third years of PD treatment. With respect to problems with peritoneal access 3

4 Peritoneal Dialysis International Page 4 of 8 KOLESNYK et al. MARCH 2010 VOL. 30, NO. 2 PDI TABLE 1 Baseline Characteristics: Patients Are Compared Within the Group for Each Period I (0 3 months) II (3 12 months) III (12 24 months) IV (24 36 months) Switched Stayed Switched Stayed Switched Stayed Switched Stayed to HD on PD to HD on PD to HD on PD to HD on PD Patients (n) Age (years) 59 a (29 77) 53 (18 86) 55 (27 78) 52 (18 86) 53 a (19 86) 57 (18 69) 51 (21 82) 51 (19 77) Gender (% male) 50 a Primary kidney disease (%) Diabetes Glomerulonephritis Renovascular disease Other Comorbidity (%) Diabetes a Cardiovascular disease a Davies comorbidity score No comorbidity 42 a a a 71 Moderate Severe rgfr at baseline (ml/minute) 4.9 (0 15) 5.5 (0 16) 3.2 (0 11) 4.1 (0 16) 2.7 (0 13) 1.9 (0 9) 2.1 (0 8) 2.1 (0 13) Mean BP at baseline (mmhg) 105±15 107±15 107±17 107±15 101±13 102±17 99±12 100±13 Patients on APD (%) NA b NA rgfr = residual glomerular filtration rate; BP = blood pressure; APD = automated peritoneal dialysis; HD = hemodialysis; NA = not available; PD = peritoneal dialysis. a p b Data on type of PD modality are not available for the first 3 months of treatment. Data are presented as % or median (range). Values for mean arterial blood pressure are given as mean±sd. Values for rgfr and mean arterial blood pressure were taken from the beginning of each period. leading to a switch to HD, we observed a reverse trend: the number of these complications decreased significantly. There was no difference between the periods before and after 3 months of PD but the number of catheter complications decreased very much after 1 year (p 0.002). There was no difference in other reasons for PD dropout between the periods of follow-up. Figure 2(b) graphically displays the incidence rates for reasons for PD dropout during the four treatment periods. Incidence rates of mortality, kidney transplantation, and underdialysis/ultrafiltration failure increased over periods of follow-up while other causes for PD failure followed a decreasing pattern. The incidence rates for every cause of PD dropout per period of followup are given in Table 2. The results of multivariate Cox analysis of PD technique survival showed that age had a borderline effect on switching to HD during the first 3 months of treatment (HR 1.03, 95% CI ) and residual renal function had an impact on technique survival during second year of treatment (HR 1.17, 95% CI ). Furthermore, none of the chosen factors appeared to be a significant predictor for switching from PD therapy to HD, regardless of treatment duration (data not shown). Table 3 shows the results of Cox analysis for probability to stay on PD treatment when events are a switch to HD and death on PD. Age appeared to have an effect on staying on PD regardless of treatment duration. Diabetes seemed to have a somewhat increasing influence with time. Cardiovascular disease had a major effect during the first 3 months and remained a significant predictor for PD failure afterward. Residual GFR at baseline for treatment periods II and III was a significant risk factor for discontinuation of PD. Transplantation was the main reason for dropout for patients that stayed on PD for more than 3 years: 49% of remaining patients received a kidney graft. Also, none of these patients had catheter problems as a reason for switching to HD. Otherwise, for this group of patients, reasons to stop PD were not different from the previous patients (data not shown). We did not find a difference between continuous ambulatory PD and automated PD patients in the probability of staying on PD (data not shown). Also, no difference was found on adjustment for center effect (data not shown). 4

5 Page 5 of 8 Peritoneal Dialysis International PDI MARCH 2010 VOL. 30, NO. 2 TECHNIQUE SURVIVAL AND MORTALITY ON PD A 100% B 80% 60% 40% 20% 0% Periods of PD treatment (months) abdominal complications underdialysis/uff psychosocial/unknown catheter complications infections Tx death Incidence rate per 1000 patient-years Periods of PD treatment (months) Figure 2 Reasons (A) and incidence rates for each reason (B) for dropout from peritoneal dialysis (PD) during four periods of follow-up. UFF = ultrafiltration failure; Tx = transplant. TABLE 2 Incidence Rates of Each Reason for Dropout from Peritoneal Dialysis During Four Periods of Follow-up (Rates with 95% confidence intervals are given per 1000 patient-years) Time period I (0 3 months) II (3 12 months) III (12 24 months) IV (24 36 months) Patients (n) Rate of death 81 (45 136) 77 (53 108) 97 (70 132) 113 (77 161) Rate of transplantation 35 (13 76) 77 (53 108) 161 ( ) 164 ( ) Rate of infections 35 (13 76) 45 (28 70) 57 (36 85) 36 (18 67) Rate of catheter failures 40 (16 83) 18 (7 36) 7 (1 20) 3 (0 20) Rate of psychosocial/ unknown problems 52 (24 99) 18 (7 36) 35 (20 59) 29 (13 58) Rate of underdialysis/uff 5 (0 32) 5 (0 16) 14 (5 31) 25 (10 53) Rate of abdominal problems 17 (6 51) 14 (5 30) 7 (1 20) 7 (0 26) UFF = ultrafiltration failure. TABLE 3 Multivariate Cox Proportional Hazards Model for Survival on Peritoneal Dialysis at the Various Baselines I (0 3 months) II (3 12 months) III (12 24 months) IV (24 36 months) Factor RR 95% CI RR 95% CI RR 95% CI RR 95% CI Age (per 1 year) Gender (male) Diabetes Age (per 1 year) Gender Cardiovascular disease Age (per 1 year) Gender rgfr a (per 1 ml/min) Age (per 1 year) Gender Diabetes Cardiovascular disease rgfr = residual glomerular filtration rate; RR = relative risk; CI = confidence interval. a Values for rgfr were taken at the beginning of each period of follow-up. 5

6 Peritoneal Dialysis International Page 6 of 8 KOLESNYK et al. MARCH 2010 VOL. 30, NO. 2 PDI MORTALITY Probability curves for PD technique survival, patient survival, and stay on PD are shown in Figure 3. One-, 2-, and 3-year patient survival was 91%, 81%, and 74%, respectively. Causes of death were classified as cardiac (n = 44); vascular, including stroke and hemorrhages (n = 14); infectious complications not related to the treatment (n = 12); abdominal complications not related to the treatment (n = 5); PD-related complications, including peritonitis and peritoneal sclerosis (n = 6); other reasons, including malignancy and treatment refusal (n = 37); and unidentified causes (n = 41). There was no difference in cause of death among the periods of follow-up (data not shown). DISCUSSION In the present study we analyzed technique and patient survival of incident PD patients according to duration of follow-up. Most other studies published so far performed their Cox proportional hazard analyses for the total follow-up period, thereby assuming hazards do not change with time. To our knowledge, this study is the first to use a time-dependent approach to analyzing PD technique survival and mortality. With this years At risk (n): Events: Mortality Technique failure Stay on PD Figure 3 Survival curves for peritoneal dialysis (PD) technique survival (solid line), patient survival (dashed line), and probability of staying on PD (dotted line). Death is an event in the patient survival curve, whereas transplantation and switch to hemodialysis (HD) are censored observations. Switch to HD is an event in PD technique survival, while death and transplantation are censored. Death and switch to HD are events in the Stay on PD curve and transplantation is a censored observation. new approach we were able to show that there are differences in reasons to stop PD treatment at various points of follow-up and that risk factors for PD failure are also time dependent. The main reasons for discontinuation of PD were kidney transplantation and death. The latter contributed, on average, 25% 30% in all periods of follow-up. Dropout due to transplantation increased from 10% during the first year of PD treatment to 50% after 3 years. This increase over time corresponds with a median time on the transplant waiting list in Europe. Transfer to HD decreased during follow-up, from 40% during the first 3 months to 25% after 2 years, due mainly to fewer abdominal and catheter complications. A high prevalence of catheter-related problems in the earliest period of PD treatment has also been reported in other studies (7,12,13). After the first 3 months of PD treatment, reasons to switch to HD are mostly represented by infectious complications and psychosocial reasons, as also found by others (3,6,7,12,14,15). Unlike the rate of infections, which follows a reversed U-shaped trend, psychosocial factors appeared to be more dominating in the earliest stage of PD treatment. This might be partially explained by pitfalls in appropriate patient selection. A relatively low incidence of stopping PD due to underdialysis or ultrafiltration failure might be explained by the assumption that some of these transfers to HD could be due to another cause. For instance, if a patient with insufficient ultrafiltration or signs of underdialysis developed an infectious or catheter complication, his transfer to HD would be documented as being due to the latter reason. Some recent studies found a significant contribution of early failure to the general PD outcome: a group from Switzerland found that up to one third of all PD failures in one center happened within the first couple of months (12); the study by Guo and Mujais reported a decline in PD technique failure between the first and second years of therapy (7); and our findings suggest that the incidence rate for PD failure is highest during the first 3 months and then remains stable for the next 3 years. We could not find a single risk factor for switch from PD to HD; therefore, it is most likely that the reason for transfer is usually more complex than the one considered final. However, the risk factors for staying on PD treatment are those most responsible for patient survival: older age, diabetes, and cardiovascular disease. These were also found by others to be significant predictors of death on PD (3,7,16,17). However, residual renal function seemed to make a significant impact on staying on PD therapy after the first 3 months and within the first 2 years of treatment. 6

7 Page 7 of 8 Peritoneal Dialysis International PDI MARCH 2010 VOL. 30, NO. 2 TECHNIQUE SURVIVAL AND MORTALITY ON PD When patients that switched to HD were compared with those that remained on PD, it appeared that older patients with a higher comorbidity score tended to switch to HD during the first 3 months. They were also more likely to be female. Female gender being associated with early dropout from PD was also found by Descoeudres et al. (12). The NECOSAD group reported earlier that, when given a free choice, older women tend to choose HD over PD (18). Probably this preference may also contribute to the early switch from PD to HD. Some studies also reported that women have worse survival on PD (19,20); however, we could not confirm that finding. Generally, when compared to reports from the previous decade (3,9), PD technique survival in The Netherlands continues to improve. This was also shown in a study using the Dutch National Registry, although we could not find a marked center effect as reported by Huisman et al. (8). Patient survival on PD in The Netherlands has somewhat improved over the past 10 years (3). The rate we found corresponds with that found in another national study (21), is higher than rates reported from some American studies (4,7), and is somewhat lower than in Asian studies (22,23). To our knowledge, this is the only study that provides detailed data on PD technique survival. The fact that this analysis was performed using the database of a national multicenter study with a relatively high number of patients makes the obtained information representative. We conclude that, in The Netherlands, the chance of switching from PD to HD is highest during the first couple of months of treatment. There is a high rate of transplantation during PD treatment and therefore an integrated care approach is being widely used. However, technique success can still be improved when attention is paid to the major reasons for early dropout, for example, problems with peritoneal access and psychosocial factors, such as appropriate patient selection. DISCLOSURES This study was supported by Baxter Healthcare through the Renal Discoveries Extramural Grant Program. REFERENCES 1. Lameire N, Peeters P, Vanholder R, Van Biessen W. Peritoneal dialysis in Europe: an analysis of its rise and fall. Blood Purif 2006; 24: Prichard S. Will peritoneal dialysis be left behind? Semin Dial 2005; 18: Jager KJ, Merkus MP, Dekker FW, Boeschoten EW, Tijssen JGP, Stevens P, et al. Mortality and technique failure in patients starting chronic peritoneal dialysis: results of the Netherlands Cooperative Study on the Adequacy of Dialysis. Kidney Int 1999; 55: Mujais S, Story K. Peritoneal dialysis in the US: evaluation of outcomes in contemporary cohorts. Kidney Int Suppl 2006; 103:S Firanek CA, Vonesh EF, Korbet SM. Patient and technique survival among urban population of peritoneal dialysis patients: an 8-year experience. Am J Kidney Dis 1991; 18: Genestier S, Hedelin G, Schaffer P, Faller B. Prognostic factors in CAPD patients: a retrospective study of a 10-year period. Nephrol Dial Transplant 1995; 10: Guo A, Mujais S. Patient and technique survival on peritoneal dialysis in the United States: evaluation in large incident cohorts. Kidney Int Suppl 2003; 88:S Huisman RM, Niewenhuizen MG, de Charro FT. Patient-related and centre-related factors influencing technique survival of peritoneal dialysis in The Netherlands. Nephrol Dial Transplant 2002; 17: Jager KJ, Merkus MP, Boeschoten EW, Dekker FW, Tijssen JGP, Krediet RT. What happens to patients starting dialysis in The Netherlands? Neth J Med 2001; 58: Schaubel DE, Blake PG, Fenton SSA. Effect of renal center characteristics on mortality and technique failure on peritoneal dialysis. Kidney Int 2001; 60: Rodrigues AS, Matos CB, Silva F, Fonesca I, Nogubra C, Santos J, et al. Long-term peritoneal dialysis experience in Portugal. Int J Artif Organs 2006; 29: Descoeudres B, Koller MT, Garzoni D, Wolff T, Steiger J, Schaub S, et al. Contribution of early failure to outcome on peritoneal dialysis. Perit Dial Int 2008; 28: Lupo A, Tarchini R, Cancarini G, Catizone L, Cocchi R, De Vecchi A. Long-term outcome in continuous ambulatory peritoneal dialysis: a 10-year-survey by the Italian Cooperative Peritoneal Dialysis Study Group. Am J Kidney Dis 1994; 24: Troidle L, Watnick S, Wuerth DB, Gorban-Brennan N, Kliger AS, Finkelstein FO. Depression and its association with peritonitis in long-term peritoneal dialysis patients. Am J Kidney Dis 2003; 42: Woodrow G, Turney JH, Brownjohn AM. Technique failure in peritoneal dialysis and its impact on patient survival. Perit Dial Int 1997; 17: Van Biessen W, Vanholder R, Veys N, Dhondt A, Lameire N. An evaluation of an integrative care approach for endstage renal disease patients. J Am Soc Nephrol 2000; 11: Vonesh EF, Snyder JJ, Foley RN, Collins AJ. The differential impact of risk factors on mortality in hemodialysis and peritoneal dialysis. Kidney Int 2004; 66: Jager KJ, Korevaar JC, Dekker FW, Krediet RT, Boeschoten EW. The effect of contraindications and patient preference on dialysis modality selection in ESRD patients in The 7

8 Peritoneal Dialysis International Page 8 of 8 KOLESNYK et al. MARCH 2010 VOL. 30, NO. 2 PDI Netherlands. Am J Kidney Dis 2004; 43: Collins AJ, Hao W, Xia H, Ebben JP, Everson SE, Constantini EG. Mortality risks of peritoneal dialysis and hemodialysis. Am J Kidney Dis 1999; 34: Vonesh EF, Moran J. Mortality in end-stage renal disease: a reassessment of differences between patients treated with hemodialysis and peritoneal dialysis. J Am Soc Nephrol 1999; 10: Liem YS, Wong JB, Hunnik MGM, de Charro FT, Winkelmayer WC. Comparison of hemodialysis and peritoneal dialysis survival in The Netherlands. Kidney Int 2007; 71: Lo WK, Ho YW, Li CS, Wong KC, Chan TM, Yu AWY, et al. Effect of Kt/V on survival and clinical outcome in CAPD patients in a randomized prospective study. Kidney Int 2003; 64: Wong JS, Port FK, Hulbert-Shearon TE, Carrol CE, Wolfe RA, Agodoa LY. Survival advantage in Asian American endstage renal disease patients. Kidney Int 1999; 55: