Diagnostic and prognostic role of peritoneal CA 125 in peritoneal dialysis patients presenting with acute peritonitis

Abstract

Background: Cancer antigen 125 (CA125) is made by peritoneal mesothelial cells and can be measured in spent dialysate effluent from peritoneal dialysis (PD) patients. It has been suggested that CA125 is a marker of peritoneal mesothelial cell mass and turnover. As PD CA125 increases during peritoneal inflammation, we wished to determine whether measuring PD CA125 during peritonitis provided additional information in determining outcome of peritonitis.

Methods: We prospectively measured peritoneal CA125 in 127 adult PD patients presenting with 187 acute episodes of PD peritonitis, measuring peritoneal CA125 from a sample of dialysate effluent obtained from a 4 hour 2 litre 13.6 g/l dextrose peritoneal dwell.

Results: Mean patient age 60.8 ± 17.1 years, 62.6% male, 33.7% diabetic and the median PD vintage was 22 (11-48) months. 127 patients (66.8%) presented with their first episode of peritonitis, 20% their second episode, 13.2% third or greater. Gram positive bacteria accounted for 64.7% of all peritonitis episodes and Gram negative bacteria 21.1%. Treatment was successful for 151 episodes of PD peritonitis (81.1%). The median PD effluent total WBC was 1240 (430-3660)/ml and serum CRP 67 (20-144) mg/l, with a PD CA125 of 38 (20.3-72.3) IU/l on presentation. There were positive correlations between PD effluent CA125 concentrations and total WBC on presentation (r = 0.41, p = <0.001) and dialysis vintage (r = -0.43, p < 0.001) but not with patient age, diabetic status, or serum CRP.There was no difference in PD effluent CA125 concentrations between Gram positive, and Gram negative peritonitis or between those episodes which responded to treatment, median 38 IU/ml (21-69) vs those with treatment failures 38 IU/ml (15-94).

Conclusion: We did not find any additional diagnostic or prognostic benefit for measuring effluent CA125 in PD patients presenting with acute peritonitis compared to standard investigations, including peritoneal WBC and serum CRP. As such our study would not support the routine measurement of peritoneal CA125 during episodes of peritonitis.

Figure 1

Correlation between log total peritoneal dialysate effluent white cell count (cells/ul) and CA-125. Both log values to base 10, Spearman correlation r = 0.41 p <0.001.

Figure 2

Box and whisker plot of median, interquartile and 95% confidence limits of peritoneal dialysate CA125 effluents measured on presentation of PD peritonitis according to underlying bacterial infection. P > 0.05by Kruskal Wallis anova.

Figure 3

Box and whisker plot of median, interquartile and 95% confidence limits of peritoneal dialysate CA125 effluents measured on presentation of PD peritonitis according to whether patients experienced 1 ^st , 2 ^nd , 3 ^rd , 4 ^th , 5 ^th or greater number of peritonitis episode. P for trend >0.05.