Comparing Peritonitis in Continuous Ambulatory Peritoneal Dialysis Patients Versus Automated Peritoneal Dialysis Patients
Alberto J. Locatelli, Gustavo M. Marcos, Marta G. Gómez, Susana A. Alvarez, Luis C. DeBenedetti

The purpose of our study was to compare the incidence of peritonitis between continuous ambulatory peritoneal dialysis (CAPD) treatment (Group I) and automated peritoneal dialysis (APD) treatment (Group II) taking into account the same population. We compared 20 patients with a followup of 215 patient-months on CAPD and 252 patient-months on APD.

Demographic data, diagnosis, peritoneal equilibration test (PET) results, adequacy, and peritonitis rate were analyzed.

Diagnoses included glomerulopathy 35%, autosomal dominant polycystic kidney disease (ADPKD) 20%, Type II diabetes 10%, systemic lupus erythematosus 5%, interstitial nephritis 5%, nephrolitiasis 5%, and unknown 20%. PET results showed that the group consisted of 30% high transporters, 45% high-average transporters, and 25% low-average transporters. Kt/V for Group I was 1.3 ± 0.3, and for Group II, 1.83 ± 0.48. Creatinine clearance for Group I was 43.64 ± 7.31 L/week/1.73 m2, and for Group II, 52.42 ± 13.47 L/week/1.73 m2. Group I presented a peritonitis rate of 8.3 episodes/patient-month, and Group II presented a rate of 18.9 episodes/patient-month. Gram-positive organisms were responsible for 49.8% of episodes of peritonitis in Group I (S. aureus 26.6%, S. epidermidis 16.6%, others 10%) and 83% of peritonitis episodes in Group II (S. epidermidis 46.6%, S. aureus 20%). Gram-negative organisms were responsible for 16.5% of episodes of peritonitis in Group I. No gram-negative peritonitis was seen in Group II. APD patients developed two cases of candida peritonitis.

Our preliminary results show that Group II exhibited a decrease in peritonitis rate while achieving better adequacy. In CAPD and APD peritonitis, gram-positive organisms predominated. In APD, we observed an increase in S. epidermidis incidence. No gram-negative organisms were observed in APD. It seems that APD is a safer treatment owing to the lower peritonitis incidence.

Key words

Peritonitis, cycler therapy

From:

Nefrología Olivos, Buenos Aires, Argentina.

Introduction

Peritonitis is still an important complication of peritoneal dialysis (PD). Automated peritoneal dialysis (APD) is a relatively new modality of PD, especially in our country, and some aspects of peritonitis are apparently different in APD than in continuous ambulatory peritoneal dialysis (CAPD) (1). With the purpose of describing the peritonitis characteristics of our PD program, we compared a population of patients who consecutively performed CAPD and APD treatments in our unit.

Patients and methods

Twenty patients [12 males and 8 females, average age 57.2 years (range: 29 - 74 years) at the beginning of APD modality] who consecutively performed CAPD (Group I) and APD (Group II) in our unit were included in the analysis. Diagnoses included glomerulopathy 35%, autosomal dominant polycystic kidney disease (ADPKD) 20%, Type II diabetes 10%, systemic lupus erythematosus 5%, interstitial nephritis 5%, nephrolitiasis 5%, and unknown 20%. Patients with at least 3 months on APD treatment were incorporated. Mean follow-up was 10.75 months (range: 3 - 26 months) in Group I and 12.6 months (range: 3 - 26 months) in Group II. Total duration of the observation period was 215 patient-months for CAPD and 252 patient-months for APD.

Our APD program started in February 1996 and the clinical follow-up ended in September 1998. Patients used UltraBag (Baxter Healthcare Corporation, Deerfield, Illinois, U.S.A.) while on CAPD and HomeChoice Cycler (Baxter Healthcare Corporation) for APD. For APD, 6 patients were on nightly intermittent peritoneal dialysis (NIPD) therapy, 12 were on continuous cycling peritoneal dialysis (CCPD), and 2 were on CCPD plus one daytime exchange with UltraBag.

A peritoneal equilibration test (PET) was performed before APD was started. PET results showed that the group consisted of 30% high transporters, 45% high-average transporters, and 25% low-average transporters. Data about adequacy corresponded to the last adequacy recorded on CAPD and APD in each patient.

Patients on APD were instructed always to observe the initial drain of the cycler and, if having abdominal pain during the day or incomplete drains while connected to the machine, to perform a manual exchange and to drain it after 2 - 4 hours for cell count and culture of the effluent. Peritonitis was defined as the presence of cloudy dialysate effluent with > 100 white blood cells/mm3, and a white blood cell differential count of > 50% polymorphonuclear cells.

We retrospectively compared adequacy, peritonitis rate, and peritonitis organisms in dialysate cultures between Group I and Group II. Additionally, we looked for differences in adequacy and age between patients on APD who had never had a peritonitis episode (Group A) and those who had at least one peritonitis episode (Group B).

Values are expressed as mean ± standard deviation (SD). Student's t-test was performed for statistical analyses.

Results

During the observation period, we recorded 30 episodes of peritonitis in 9 patients over 215 months in Group I, and 15 episodes in 8 patients over 252 months in Group II. The peritonitis rate was 8.3 episodes per patient-month (1.44 episodes per year) in Group I and 18.9 episodes per patient-month (0.63 episodes per year) in Group II (p < 0.05).

Mean Kt/V was 1.3 ± 0.3 in Group I and 1.83 ± 0.48 in Group II (p < 0.001). Mean creatinine clearance was 43.64 ± 7.31 L/week/1.73 m2 in Group I and 52.42 ± 13.47 L/week/1.73 m2 in Group II (p < 0.01) (Table I). Data about adequacy and age in the 12 patients who were free of peritonitis in APD (Group A) and the 8 patients who had at least one episode (Group B) are shown in Table II. No significant difference was seen in average age (55.62 years vs 58.25 years), creatinine clearance (53.3 L/week/1.73 m2 vs 52.56 L/week/1.73 m2), or Kt/V (1.73 vs 1.95) between Group A and Group B (Table II).

The spectrum of organisms responsible for the episodes of peritonitis are shown in Table III. Gram-positive organisms were responsible for 49.8% of the 30 episodes of peritonitis in Group I (S. epidermidis 16.6%, S. aureus 26.6%, streptococcus and enterococcus 9.6%) and 83.3% of 15 peritonitis episodes in Group II (S. epidermidis 46.6%, S. aureus 20%). Gram-negative organisms were responsible for 16.5% of peritonitis episodes in CAPD. No gram-negative bacteria peritonitis appeared in Group II. Two APD patients developed candida peritonitis. We found one polymicrobial peritonitis in Group II. Eight patients of the 20 studied had no peritonitis episodes on either CAPD or APD. Seven of the 15 episodes of peritonitis in Group II occurred in two patients who performed CCPD plus manual exchanges.

Discussion

We observed in our study that the transfer of patients from CAPD to APD resulted in an important reduction in peritonitis rate, as well as in better adequacy. Better results are reached in some units today, both for CAPD with UltraBag and for APD, although major variations are still seen from center to center. During the 1980s and 1990s, an overall average of approximately 1.1 - 1.3 peritonitis episodes per patient-year was reported for CAPD (1). In a study (2) involving 224 CAPD patients from six centers in Europe and North America in 1991, overall mean peritonitis incidence was 0.40, with inter-center variations; but in that study, 113 patients were excluded for various medical and non medical reasons, so that the studied patients represented just two thirds of the total population. The mean peritonitis episodes reported are 0.53 (0.20 - 0.93) for CCPD (1,3-11). We previously reported a peritonitis incidence of 1.2 episodes per patient-year in CAPD with the "standard system" (12).

We consider that our high peritonitis rate in CAPD with UltraBag might have been influenced by many factors, such as a greater number of connections, patient fatigue, low motivation, and poor care in technical procedures. Although not optimal, the diminished peritonitis rate in APD was consistent and was perhaps due to factors inherent to the modality, considering that the patients were the same. Additionally, we speculate that the patients entering into the new modality (APD) were retrained, more motivated, and more meticulous in their technical procedures.

Better adequacy has not been directly associated with improvement in peritonitis rates, but it is clearly related to better outcomes and less global morbidity (13). We think it conceivable that better adequacy could have contributed to the lower peritonitis rate in our population of APD patients.

The contribution of APD to diminishing the chance for touch contamination is limited by the need to perform manual exchanges to reach adequacy targets, especially in patients with a larger body surface area. We consider that manual exchanges were the main factor in our two APD patients who experienced 7 of the total 15 peritonitis episodes. They performed CCPD plus manual exchanges.

The main features of our preliminary experience in APD peritonitis were:
  1. Gram-positive organisms predominated in both Group I and Group II (49.8% and 66.6% respectively), chiefly S. aureus in Group I (26.6% vs 20%) and S. epidermidis in Group II (16.6% vs 46.6%). As we did not include exit-site infection data in our analysis, we cannot explain this finding. Other authors reported similar findings, but with a reduction of S. aureus and with similar rates of coagulase-negative staphylococcus in APD patients compared with CAPD patients (14).

  2. We found no gram-negative infections in patients on APD, whereas gram-negative infections represented 16.5% of total peritonitis in CAPD. Other authors found an increase in the incidence of gram-negative peritonitis in patients on the HomeChoice Cycler system, but without statistical significance when compared with patients using UltraBag (14). Two APD patients developed candida peritonitis. One of these cases originated from an evident touch contamination; the other patient never reached adequacy goals and had been treated with multiple antibiotics.

In summary, controversy remains about peritonitis germ distribution in APD. It is of major concern to highlight bacteriologic features of this modality of treatment by prospective and multicenter studies.

Conclusions

Our preliminary results show that Group II exhibited a decrease in peritonitis rate while achieving better adequacy. In CAPD and APD, gram-positive organisms predominated. In APD, we observed an increase in S. epidermidis incidence. No gram-negative organisms were observed in APD. It seems that APD is a safer treatment, owing to the reduction of peritonitis incidence.

References

  1. Keane WF, Alexander SR, Bailie G, et al. Peritoneal dialysis-related peritonitis treatment recommendations: 1996 update. Perit Dial Int 1996; 16:557-73.
  2. Young GA, Kopple JD, Lindholm B, et al. Nutritional assessment of continuous ambulatory peritoneal dialysis patients: An international study. Am J Kidney Dis 1991; 17:462-71.
  3. Rottemberg J, Brouard R, Isaad B, et al. Prevention of peritonitis during continuous ambulatory dialysis. Value of disconnectable system. Presse Med 1988; 17:1349-53.
  4. de Fijter CWH, Oe PL, Nauta JJP, et al. A prospective, randomized study comparing the peritonitis incidence of CAPD and Y connector (CAPD-Y) with continuous cyclic peritoneal dialysis (CCPD). In: Khanna R, Nolph KD, Prowant BF, Twardowski ZJ, Oreopoulos DG, eds. Advances in peritoneal dialysis. Toronto: Peritoneal Dialysis Bulletin, 1991; 7:186-9.
  5. Holley JL, Bernardini J, Piraino B. Continuous cycling peritoneal dialysis is associated with lower rates of catheter infection than continuous ambulatory peritoneal dialysis. Am J Kidney Dis 1990; 16:133-6.
  6. Price C, Suki W. New modification of peritoneal dialysis: Option in the treatment of patients with renal failure. Am J Nephrol 1981; 1:97-104.
  7. Cavoretto L, Jackson F. A decrease in peritonitis with CCPD: One unit's experience. Nephrol Nurse 1983; 5:33-7.
  8. Walls J, Smith BA, Feehally J, Tavernel D, Turgan C. CCPD —An improvement on CAPD. In: Gahl GM, Nolph KD, eds. Advances in Peritoneal Dialysis. Amsterdam: Excerpta Medica 1981:141-3.
  9. Diaz-Buxo JA, Walker PJ, Burgess WP, Chandler JT, Farmer CD, Hot KL. Current status on CCPD in the prevention of peritonitis. In: Khanna R, Nolph KD, Prowant BF, Twardowski ZJ, Oreopoulos DG, eds. Advances in CAPD. Toronto: Peritoneal Dialysis Bulletin, 1986; 2:145-8.
  10. Levy MN, Balfe JW, Geary DF, et al. Factors predisposing and contributing to peritonitis during chronic peritoneal dialysis in children. A 10 years experience. Perit Dial Int 1990; 10:263-9.
  11. Diaz-Buxo JA. Management of peritonitis in automated peritoneal dialysis patients. In: Khanna R, ed. Advances in peritoneal dialysis. Toronto: Peritoneal Dialysis Publications, 1998; 14:131-6.
  12. Barone R, Alvarez Quiroga M, Ferraro J, Locatelli A, De Benedetti L. Exit site infection and peritonitis. In: Otta K, et al, eds. Current concepts in peritoneal dialysis. Proceedings of the Fifth Congress of the International Society for Peritoneal Dialysis; 21 July 1990; Kyoto, Japan. Kyoto: Elsevier Science Publishers, 1990:413-18.
  13. Churchill D. The multicenter infection study project. Request for proposals. Perit Dial Int 1996; 16:550.
  14. Troidle LK, Gorban-Brennan N, Kliger AS, Finkelstein FO. Continuous cycler therapy, manual peritoneal dialysis therapy and peritonitis. In: Khanna R, ed. Advances in peritoneal dialysis. Toronto: Peritoneal Dialysis Publications, 1998; 14:137-41.

Corresponding author:

Alberto Locatelli, (1640) Prilidiano Pueyrredón 2047, Martínez, Provincia de Buenos Aires, República Argentina.