Incidence and outcomes of encapsulating peritoneal sclerosis (EPS) and factors associated with severe EPS

Background Encapsulating peritoneal sclerosis (EPS) is a rare but serious complication of long-term peritoneal dialysis (PD). However, previous studies reported large variations in its mortality rates that may associate with a different degree of EPS severity. This study reports the incidence and outcomes of EPS and identifies the risk factors associated with severe EPS. Methods We retrospectively analyzed clinical data of EPS patients from 3 medical centers in Taiwan from January 1982 to September 2015, and classified patients as having mild/moderate or severe EPS. Patients with intractable intestinal obstruction/gut-related sepsis that needed surgical intervention or resulted in mortality were in severe EPS group. Follow-up for outcome was through December 31, 2015. Clinical characteristics, peritoneal dialysis (PD)related parameters, biochemical and imaging results were analyzed and compared between groups. Results Fifty-eight of 3202 patients undergoing PD during the study period had EPS (prevalence 1.8%). The incidence of EPS increased for patients on PD for >6–8 years (6 yrs. vs. >6–8 yrs., 0.0% vs. 1.8%, p = 0.001). Relative to those on PD for >6–8 years, the risk of EPS significantly increased with PD duration longer than 10 years (>10–12 years vs. >6–8 years: OR: 5.5, 95% CI: 1.7–17.1, p < 0.01). Twenty-three patients fulfilled the criteria for severe EPS. The overall mortality rate of EPS was 35% (20/58), and was 74% (17/23) in the severe EPS group. The average serum levels of C-reactive protein (CRP) and intact-parathyroid hormone (i-PTH), which were checked every 3~6 months within one year before diagnosis of EPS, were higher in severe EPS group than in mild/moderate group (p = 0.02, p = 0.08, respectively). Multivariate analysis revealed severe EPS was independently associated with bowel tethering (based on CT), presentation with bloody ascites, diagnosis of EPS after withdrawal from PD, and i-PTH 384 pg/mL. Receiver operating characteristic analysis indicated that presentation with 2 or more of the 5 risk factors (EPS diagnosis after PD withdrawal, bloody ascites, bowel tethering, CRP 29 mg/L, and i-PTH 384 pg/mL) had a good accuracy (AUC = 0.80, p = 0.001) for prediction of severe EPS. Conclusions The incidence of EPS increases with PD duration. Severe EPS has high mortality rate and is associated with bowel tethering, presentation of bloody ascites, diagnosis after PD withdrawal, and higher serum levels of i-PTH before EPS diagnosis. Having 2 or more of the 5 risk factors can provide a good accuracy for prediction of severe EPS.