Fungal Peritonitis in Continuous Ambulatory Peritoneal DialysisThe Auckland Experience Ramesh Nagappan, MD, John F. Collins, FRACP, and Wan Tin Lee, MB, BS • Fungal infection is an uncommon cause of peritonitis in patients on continuous ambulatory peritoneal dialysis (CAPO). We report our center's experience with 38 episodes of fungal peritonitis occurring in 33 patients, out of a total of 503 patients managed on CAPO over 11 112 years, and review the relevant literature. Our usual management philosophy has been one of early peritoneal catheter removal without antifungal therapy. In those with worsening clinical features, and in those with persistence of signs and symptoms beyond 48 hours after catheter removal, antifungal drugs were administered. Only five patients received antifungal therapy initially, followed by later catheter removal. Seventy-six percent of patients treated by catheter removal alone (N = 21) and 64% of patients treated by catheter removal followed by antifungal therapy (N = 11) were successfully reestablished on CAPO. A policy of early catheter removal, usually alone, but followed by antifungal therapy in select cases, can be associated with a mortality rate of less than 15% and a high rate of return to effective peritoneal dialysis. © 1992 by the National Kidney Foundation, Inc. INDEX WORDS: Fungi; peritoneal dialysis; peritonitis; continuous ambulatory peritoneal dialysis technique survival.
F
UNGAL PERITONITIS is an uncommon complication of continuous peritoneal dialysis (PD). While it accounts for less than 10% of episodes of PD-related peritonitis, it contributes significantly to morbidity and mortality. Management strategies have included early peritoneal catheter removal with or without antifungal therapy, or antifungal therapy with subsequent catheter removal, if necessary. While many investigators favor early catheter removal and antifungal therapy,I-9 most series are small with significant complications, including a low rate of return to PD. Our management philosophy has been one of early catheter removal without antifungal therapy in most cases of fungal peritonitis. We report here on our center experience of 38 episodes of fungal peritonitis emphasizing management strategy and outcome. Our results are compared with published reports in the medical literature. 1-21
METHODS Patient Population The renal service at Auckland provides renal replacement therapy to a population of 1.2 million. Between December 1979 and April 1991, a total of 503 patients in the renal service were treated for end-stage renal disease (ESRD) using contin-
From the Division ofNephrology, Department ofMedicine, Auckland Hospital, Auckland, New Zealand. Received April 21, 1992; accepted in revised form June 30, 1992. Address reprint requests to John F. Collins, FRACP, Consultant Nephrologist, Department of Medicine, Auckland Hospital, Auckland-I, New Zealand. © 1992 by the National Kidney Foundation, Inc. 0272-6386/92/2005-0008$3.00/0 492
uous ambulatory peritoneal dialysis (CAPO). Our diagnostic criteria for fungal peritonitis were ( 1) CAPO effluent cell count of 100 or more white blood cells (WBC) per microliter, (2) a differential cell count of the dialysate showing greater than 50% polymorphonuclear cells, and (3) isolation of fungi. Patients from the literature were identified by searching the English-language literature through the Index Medicus and via the MEDLINE computer database under the headings of Peritoneal dialysis, Peritonitis, and Fungal peritonitis, and by searching bibliographies of selected reports. I-21 For the purpose of our discussion, reports of three or more cases of fungal peritonitis were included.
RESULTS The management policy for fungal peritonitis in our center revolves around early removal of the Tenckhoffperitoneal catheter. We do not use antifungal drugs except where symptoms and signs do not improve within 48 hours after catheter removal or where they are severe before that time. Thirty-eight episodes of fungal peritonitis in 33 patients (14 male, 19 female) were managed during the study period of 138 months (Table I). The average age of our patients was 47.7 years (range, 21 to 66 years). The cause of ESRD was glomerulonephritis in 33% of our patients, diabetic nephropathy in 27%, hypertensive nephrosclerosis in 12%, polycystic kidney disease in 8%, reflux nephropathy in 8%, and a variety of other causes in the remaining patients. In terms of age and sex distribution and underlying kidney disease, these 33 patients were similar to our overall CAPD population. The mean duration ofCAPD before contracting fungal peritonitis was 15.2 months (range, 2 to 69). There were 1,205 episodes of bacterial peri-
American Journal of Kidney Diseases, Vol XX, No 5 (November), 1992: pp 492-496
493
FUNGAL PERITONITIS IN CAPO-AUCKLAND EXPERIENCE Table 1. Patient and Treatment Profile in 38 Episodes of Fungal Peritonitis-Auckland CAPO Experience
Patient No.
Age/Sex
Duration of CAPO (mo)
1 2 3
47/F 61/F 61/F
2 27 15
+ +
36 6 36
4
53/M
22
+
50
5
25/F
11
+
5
6 7 8 9 10 11 12
30/F 34/F 41/M 26/M 35/F 62/F 62/F 65/F 66/F 22/M 53/F 49/M 41/F 21/F 23/F 41/M 53/F 48/M 66/M 41/F 61/F 30/F 48/F 48/M 58/M 55/M 46/F 46/F 46/F 57/M 61/M 58/F 40/M
14 25 2 13 2 12 10 40 52 4
+ + + + + +
13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29
30 31 32 33
Preceding Antibiotics
Coexistent Fungal Infections Nail
Oral
+ +
17 15 41 47 69 11 35 9 21 11 15 27 9 24 10 33 4 8 12 9
17 18 7
+
+ + +
Skin/nail
+ + + + + +
Hospital Stay (d)
85 12 5 12 7 4 7 3 7 29 6 8 3 14 8 18 9 13 5 7 4 43 8 10 3 12 10 66 3 12 4 9 3
Initial Treatment Amphotericin + 5FC (IV) Miconazale + 5FC (IP) Miconazole (IP) + amphotericin (IV) Fluconazole (IP) + amphotericin (IV) Amphotericin (IP) + fluconazole PC removal PC removal PC removal PC removal PC removal PC removal PC removal PC removal PC removal PC removal PC removal PC removal PC removal PC removal PC removal PC removal PC removal PC removal PC removal PC removal PC removal PC removal PC removal PC removal PC removal PC removal PC removal PC removal PC removal PC removal PC removal PC removal Fluconazole (IV)
Further Treatment
Outcome
PC removal PC removal PC removal
HD HD HD
PC removal
HD
PC removal
HD
Amphotericin
CAPO CAPO CAPO CAPO CAPO Died" CAPO CAPO HD HD CAPO Died" CAPO CAPO Oiedt HO HO CAPO CAPO CAPO CAPO Died:f: CAPO CAPO CAPO HO CAPO CAPO CAPO CAPO CAPO CAPO Oiedt
Oral ketoconazole
Fluconazole Amphotericin
Fluconazole Amphotericin Amphotericin Amphotericin
Ketoconazole Amphotericin Amphotericin PC removal
Abbreviations: PC. peritoneal catheter; HO. hemodialysis; 5FC. 5-flucytosine; IV. intravenous; IP. intraperitoneal. " Elective withdrawal of treatment. t Candidal sepsis. :f: Cerebrovascular accident/elective withdrawal of treatment.
tonitis in 503 patients treated during the study period. The peritonitis rate in the 470 patients who did not contract fungal peritonitis was 1.56 episodes per patient-year. This compares with a bacterial peritonitis rate of 2.51 episodes per patient-year in the 33 patients who did contract fungal peritonitis (data includes only the time up to the first episode of fungal peritonitis). Table 2 outlines the causative fungi in our patients. Candida remains the most frequent or-
ganism; this conforms with the published experience. 1- 21 Eighteen patients had an episode of bacterial peritonitis in the month preceding their fungal peritonitis; these patients, and four others who had undergone abdominal surgery in the previous week, all received intraperitoneal antibiotics preceding their fungal peritonitis. This registered an antibiotic exposure rate of 57.8% compared with 69.3% (N = 176) of cases from a recent review. 1
NAGAPPAN, COLLINS, AND LEE
494 Table 2. Causative Fungi in Our 38 Episodes Causative Fungal Organisms Other
Candida Calbicans C parapsilosis C guilliermondii tropicalis (not speciated)
13 (34%) 8(21%) 1 (2.6%) 1 (2.6%) 3(8.0%)
Rhodotorula Torulopsis Fusarium Saccharomyces Paecilomyces variotii Trichosporon cutaneum
5 (13%) 3(8.6%) 1 (2.6%) 1 (2.6%) 1 (2.6%) 1 (2.6%)
Three patients suffered coexistent fungal infections (patient 1, nail; patient 6, oral; patient 25, skin and nail). Nineteen of our 33 patients are women; none of them had documented preceding or coexistent vulvovaginal candidiasis. However, we did not perform routine vaginal examinations in asymptomatic individuals. Three patients had discrete recurrences offungal peritonitis, each episode being separated from the other by 2 or more months. Case 12 suffered fungal peritonitis in the 10th, 40th, and 52nd months of CAPO. Case 29 suffered three separate episodes of Candida parapsilosis peritonitis; following her third episode, she commenced oral ketoconazole prophylaxis 100 mg weekly and has had no recurrence for more than 30 months. Therapy and Outcome Antifungal therapy with subsequent catheter removal. Six episodes were managed in this way, five in the early part of our PO experience (Table 1) and one recently (a patient who refused any surgery for two weeks). Agents used included amphotericin, fluconazole, miconazole, and 5flucytosine. None of these patients were able to be reestablished on PO and one died of candidal
sepsis. This experience prompted a policy change to early catheter removal, usually without antifungal therapy. Peritoneal catheter removal-no antifungal therapy. Twenty-one episodes were managed with catheter removal alone, and 76% of them were successfully reestablished on PO, 14% remained permanently on hemodialysis, and 10% (N = 2) died. Peritoneal catheter removal with additional antifungal therapy. Eleven episodes were managed with early catheter removal with subsequent antifungal therapy. In general, these patients had more severe signs and symptoms than those not receiving antifungal therapy, and their hospital stay averaged 18 days compared with 11 days in those requiring catheter removal alone. Sixty-four percent were successfully re-established on PO, 18% remained on hemodialysis and 18% (N = 2) died. Deaths Five (13%) of the 38 episodes resulted in death occurring in close proximity to fungal peritonitis (Table 3). Of these, patient 11 and patient 15 had elective withdrawal of treatment; patients 17 and 33 died of candidal sepsis, while patient 24, whose catheter removal was followed by antifungal drugs, suffered an unrelated cerebrovascular accident leading to withdrawal of treatment, and died soon after. DISCUSSION
Fungal peritonitis is an uncommon, yet potentially life-threatening complication of peritoneal dialysis. Three percent of all CAPO-associated peritonitis episodes in our center, over
Table 3. Summary of Treatment and Outcome in Our Study
Treatment No.
2
3
Mode of Treatment
Peritoneal catheter removal alone Peritoneal catheter removal followed by antifungal drugs Antifungal drugs followed by peritoneal catheter removal
No. of Episodes
Average Duration of Hospital Stay (d)
Average Duration of Interim Hemodialysis (d)
CAPD
HD
Death
21
11
56
76%
14%
10%
11
18
56
64%
18%
18%
6
23
83%
17%
Outcome
FUNGAL PERITONITIS IN CAPO-AUCKLAND EXPERIENCE
111/2 years, were due to fungal infection. This compares with the 2% to 10.2% incidence reported in the literature. 1-21 Many investigators recommend early catheter removal and antifungal therapy. 1-21 However, some have administered antifungal therapy in an attempt to avoid catheter removal. Others have removed the peritoneal catheter without any follow-up antifungal therapy. Where antifungal therapy has been used, there have been a number of different agents employed for different durations by different routes of administration (intravenous, intraperitoneal, or oral). Indeed, no clear consensus emerges on the use of antifungal therapy in this condition. The treatment objectives for fungal peritonitis are to limit morbidity, to prevent mortality, and return the patient, if possible, to CAPO. After our early unsuccessful experience in managing patients with antifungal therapy without initial catheter removal, we moved to a policy of early catheter removal. From our review of the literature, we found that many centers have administered antifungal drugs without catheter removal and a variety of regimens were used. I -4,15-19 Among 41 cases reported, 41.4% continued on peritoneal dialysis, while 21.9% required transfer to hemodialysis. Among those treated with antifungal drugs as the solitary mode of therapy, 36.6% of the patients died due to fungal peritonitis. In our group managed with peritoneal catheter removal without subsequent antifungal therapy (Table 3), 76% of patients were able to be reestablished on PD. These results compare favorably with an analysis of the literature on patients managed with catheter removal alone (N = 53)2,4,6,12,14,20 showing only 18.8% reestablished on peritoneal dialysis, 45.2% remained on hemodialysis, and a mortality rate of 18.8% (Table 4). The final outcome was unavailable in 16.9% of patients, half of them reported only as "cured."
495
We did not use antifungal therapy unless symptoms were severe and persistent. In our series of 38 episodes, 11 patients were managed with peritoneal catheter removal followed by antifungal drugs (Table 3) and 64% of them returned to CAPO. These results also compare favorably with an analysis of the literature on this management strategy (N = 40) (Table 4), which shows only 25% being reestablished on PO and 45% remaining on hemodialysis. Fifteen percent died due to fungal peritonitis and the outcome was unavailable in another 15%. Various antifungal drug regimens were used by those reporting this strategyl-9; most used amphotericin alone or in combination with/followed by an azole derivative. Strict comparisons of drug regimens and outcome are difficult because of the wide array of doses and routes of administration. It is widely accepted that the peritoneal catheter provides a conducive site for microbial colonization. Electron microscopy of removed peritoneal catheter shows organisms embedded in an amorphous matrix on the surface of the catheter3,22; this renders effective antibiosis difficult. Where fungal colonization of the peritoneal catheter has occurred, further difficulties arise because of the more complex structure offungi,23 rendering antimicrobial therapy less effective. 24 It is logical to remove the catheter as early as possible, thus removing the main locus of sequestral colonization and allowing the normal defense mechanisms of the peritoneum to clear the remaining infection. In 21 episodes with mild symptoms, we did not use antifungal therapy on this basis. Our results show that 16 patients in this group were able to be successfully recommenced on PO after a period on hemodialysis and that only two patients died, both as a result of an elective decision to withdraw treatment for psychosocial reasons.
Table 4. Summary of a Review of the Literature on Fungal Peritonitis in CAP01- 21 Outcome
Mode of Treatment
Catheter removal alone Catheter removal followed by drugs Drugs (various routes) followed by catheter removal Antifungal drugs alone
No. of Episodes
CAPD
HD
Death
Not Available
53 40
18.8% 25%
45.2% 45%
18.8% 15%
16.9% 15%
61 41
18% 41.4%
54% 21.9%
15% 36.6%
13%
496
NAGAPPAN, COLLINS, AND LEE
We believe this outcome vindicates our policy. However, an analysis of the literature l - 21 (Table 4), shows similar death rates with all of the treatments that included catheter removal, ranging from 1.5 to 2 times the mortality rate we encountered with catheter removal alone in mild cases. We cannot offer an explanation for this discrepancy, except to state that many of these series l -2 1 were small, reflecting limited experience with this problem. The majority of our patients were able to be reestablished on PO after a rest period on hemodialysis. The rate of our reestablishment on PO was much higher than that found in the literature and makes the point that most patients who suffer fungal peritonitis can be completely cured.
It is likely that in many centers an attempt is not made to reestablish the patient on PO; in the light of our experience, some groups may wish to review their policies in this matter. In conclusion, we state that early peritoneal catheter removal, with antifungal medications only in patients with severe or persistent symptoms, can be associated with a low mortality and a high rate of successful reestablishment of CAPO. ACKNOWLEDGMENT The authors acknowledge the assistance of Dr Bremner and the staff of the clinical microbiology laboratory at Auckland hospital for fungal identification, and Dr Ellis-Pegler, Infectious Diseases Specialist at Auckland Hospital, for valuable comments.
REFERENCES I. Cheng IKP, Fang GX, Chan TM, et aI: Fungal peritonitis complicating peritoneal dialysis: Report of27 cases and review of treatment. Q J Med New Series 71(265):407-416, 1989 2. Fabris A, Biasioli S, Borin D, et al: Fungal peritonitis in peritoneal dialysis: Our experience and review of treatments. Perit Dial Bull 4:75-77, 1984 3. Kerr CM, Perfect JR, Craven PIC, et aI: Fungal peritonitis in patients on continuous ambulatory peritoneal dialysis. Ann Intern Med 99:334-337, 1983 4. Powell D, San Luis E, Calvin S, et al: Peritonitis in children undergoing CAPD. Am J Dis Child 139:29-32, 1985 5. Vargemezis V, Papadopoulou ZL, Liamos H, et al: Management of fungal peritonitis during CAPD. Perit Dial Bull 6:17-20, 1986 6. Forwell MA, Smith WG, Tsakiris D, et aI: Morbidity of fungal peritonitis. Contrib NephroI57:11O-113, 1987 7. Tapson JS, Mansy H, Fr eeman R, et aI: The high morbidity of CAPD fungal peritonitis-Description of 10 cases and review of treatment strategies. Q J Med New Series 61(235):1047-1053, 1986 8. Morford DW, Currie J: Eight cases offungal peritonitis during 765 patient months of CAPO. III International symposium on Peritoneal Dialysis, Washington, DC, June 17-20, 1984 9. Enriquez JL, Kalia A, Travis LB: Fungal peritonitis in children on peritoneal dialysis. J Pediatr 117:830-832, 1990 10. Steinberg SM, Cutler SJ, Nolph KD, et al: Acomprehensive report on the experience of patients on CAPD for the treatment of end-stage renal disease. Am J Kidney Dis 4:233241 , 1984 II. Bayer AS, Blumenkrantz MJ, Montgomerie JZ, et aI: Fungal peritonitis-Report of22 cases and review of the English literature. Am J Med 61 :832-839, 1976 12. Khanna R, Oreopoulos DG, Vas S, et al: Fungal peritonitis in patients undergoing chronic intermittent or continuous ambulatory peritoneal dialysis. Proc Eur Dial Transplant Assoc 17:291-296, 1980
13. Keogh JAB, Carr ME, Murray F, et al: Treatment of fungal peritonitis in CAPD patients using peritoneal lavage. Perit Dial Bull 5:67-69, 1985 14. Rodriguez-Perez JC: Fungal peritonitis in CAPDWhich treatment is best? Contrib Nephrol 57: 114-121, 1987 15. Johnson RJ, Ramsey PG, Gallagher N, et al: Fungal peritonitis in patients on peritoneal dialysis: Incidence, clinical features and prognosis. Am J Nephrol 5: 169-175, 1985 16. Benevent D, Peyronnet P, Lagarde C, et al: Fungal peritonitis in patients on CAPD. Three recoveries in 5 cases without catheter removal. Nephron 41 :203-206, 1985 17. Zaruba K, Peters J, Jungbluth H: Successful prophylaxis of fungal peritonitis in patients on CAPD: Six years' experience. Am J Kidney Dis 17:43-46, 1991 18. Cecchin E, De Marchi S, Panarello G , et al: Chemotherapy and/or removal of peritoneal catheter in the management of Fungal peritonitis complicating CAPD. Nephron 40:251-252, 1985 19. Struijk D, Krediet RJ, Boeschoten EW, et al: Antifungal treatment of candida peritonitis in CAPD patients. Am J Kidney Dis 9:66-70, 1987 20. Eisenberg ES, Leviton I, Soeiro R: Fungal peritonitis in patients receiving peritoneal dialysis. Experience with II patients and review of the literature. Rev Infect Dis 8:309321 , 1986 21. Pollock CA, Ibels LS, Caterson RJ, et aI: CAPO-Eight years of experience at a single center. Medicine (Baltimore) 68:293-307, 1989 22. Marrie TJ, Noble MA, Costerton JW: Examination of the morphology of bacteria adhering to peritoneal dialysis catheters by scanning and transmission electron microscopy. J Clin MicrobioI18:1388-1398, 1983 23. Huppert M, Macpherson D, Cazin J : Pathogenesis of Candida albicans infection following antibiotic therapy. The effect of antibiotics on the growth of C albicans. J Bacteriol 65:171-177, 1953 24. Edwards JE Jr: Invasive Candida infections-Evolution ofa fungal pathogen. N Eng! J Med 324: 1060-1062, 1991
F
UNGAL PERITONITIS is an uncommon complication of continuous peritoneal dialysis (PD). While it accounts for less than 10% of episodes of PD-related peritonitis, it contributes significantly to morbidity and mortality. Management strategies have included early peritoneal catheter removal with or without antifungal therapy, or antifungal therapy with subsequent catheter removal, if necessary. While many investigators favor early catheter removal and antifungal therapy,I-9 most series are small with significant complications, including a low rate of return to PD. Our management philosophy has been one of early catheter removal without antifungal therapy in most cases of fungal peritonitis. We report here on our center experience of 38 episodes of fungal peritonitis emphasizing management strategy and outcome. Our results are compared with published reports in the medical literature. 1-21
METHODS Patient Population The renal service at Auckland provides renal replacement therapy to a population of 1.2 million. Between December 1979 and April 1991, a total of 503 patients in the renal service were treated for end-stage renal disease (ESRD) using contin-
From the Division ofNephrology, Department ofMedicine, Auckland Hospital, Auckland, New Zealand. Received April 21, 1992; accepted in revised form June 30, 1992. Address reprint requests to John F. Collins, FRACP, Consultant Nephrologist, Department of Medicine, Auckland Hospital, Auckland-I, New Zealand. © 1992 by the National Kidney Foundation, Inc. 0272-6386/92/2005-0008$3.00/0 492
uous ambulatory peritoneal dialysis (CAPO). Our diagnostic criteria for fungal peritonitis were ( 1) CAPO effluent cell count of 100 or more white blood cells (WBC) per microliter, (2) a differential cell count of the dialysate showing greater than 50% polymorphonuclear cells, and (3) isolation of fungi. Patients from the literature were identified by searching the English-language literature through the Index Medicus and via the MEDLINE computer database under the headings of Peritoneal dialysis, Peritonitis, and Fungal peritonitis, and by searching bibliographies of selected reports. I-21 For the purpose of our discussion, reports of three or more cases of fungal peritonitis were included.
RESULTS The management policy for fungal peritonitis in our center revolves around early removal of the Tenckhoffperitoneal catheter. We do not use antifungal drugs except where symptoms and signs do not improve within 48 hours after catheter removal or where they are severe before that time. Thirty-eight episodes of fungal peritonitis in 33 patients (14 male, 19 female) were managed during the study period of 138 months (Table I). The average age of our patients was 47.7 years (range, 21 to 66 years). The cause of ESRD was glomerulonephritis in 33% of our patients, diabetic nephropathy in 27%, hypertensive nephrosclerosis in 12%, polycystic kidney disease in 8%, reflux nephropathy in 8%, and a variety of other causes in the remaining patients. In terms of age and sex distribution and underlying kidney disease, these 33 patients were similar to our overall CAPD population. The mean duration ofCAPD before contracting fungal peritonitis was 15.2 months (range, 2 to 69). There were 1,205 episodes of bacterial peri-
American Journal of Kidney Diseases, Vol XX, No 5 (November), 1992: pp 492-496
493
FUNGAL PERITONITIS IN CAPO-AUCKLAND EXPERIENCE Table 1. Patient and Treatment Profile in 38 Episodes of Fungal Peritonitis-Auckland CAPO Experience
Patient No.
Age/Sex
Duration of CAPO (mo)
1 2 3
47/F 61/F 61/F
2 27 15
+ +
36 6 36
4
53/M
22
+
50
5
25/F
11
+
5
6 7 8 9 10 11 12
30/F 34/F 41/M 26/M 35/F 62/F 62/F 65/F 66/F 22/M 53/F 49/M 41/F 21/F 23/F 41/M 53/F 48/M 66/M 41/F 61/F 30/F 48/F 48/M 58/M 55/M 46/F 46/F 46/F 57/M 61/M 58/F 40/M
14 25 2 13 2 12 10 40 52 4
+ + + + + +
13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29
30 31 32 33
Preceding Antibiotics
Coexistent Fungal Infections Nail
Oral
+ +
17 15 41 47 69 11 35 9 21 11 15 27 9 24 10 33 4 8 12 9
17 18 7
+
+ + +
Skin/nail
+ + + + + +
Hospital Stay (d)
85 12 5 12 7 4 7 3 7 29 6 8 3 14 8 18 9 13 5 7 4 43 8 10 3 12 10 66 3 12 4 9 3
Initial Treatment Amphotericin + 5FC (IV) Miconazale + 5FC (IP) Miconazole (IP) + amphotericin (IV) Fluconazole (IP) + amphotericin (IV) Amphotericin (IP) + fluconazole PC removal PC removal PC removal PC removal PC removal PC removal PC removal PC removal PC removal PC removal PC removal PC removal PC removal PC removal PC removal PC removal PC removal PC removal PC removal PC removal PC removal PC removal PC removal PC removal PC removal PC removal PC removal PC removal PC removal PC removal PC removal PC removal Fluconazole (IV)
Further Treatment
Outcome
PC removal PC removal PC removal
HD HD HD
PC removal
HD
PC removal
HD
Amphotericin
CAPO CAPO CAPO CAPO CAPO Died" CAPO CAPO HD HD CAPO Died" CAPO CAPO Oiedt HO HO CAPO CAPO CAPO CAPO Died:f: CAPO CAPO CAPO HO CAPO CAPO CAPO CAPO CAPO CAPO Oiedt
Oral ketoconazole
Fluconazole Amphotericin
Fluconazole Amphotericin Amphotericin Amphotericin
Ketoconazole Amphotericin Amphotericin PC removal
Abbreviations: PC. peritoneal catheter; HO. hemodialysis; 5FC. 5-flucytosine; IV. intravenous; IP. intraperitoneal. " Elective withdrawal of treatment. t Candidal sepsis. :f: Cerebrovascular accident/elective withdrawal of treatment.
tonitis in 503 patients treated during the study period. The peritonitis rate in the 470 patients who did not contract fungal peritonitis was 1.56 episodes per patient-year. This compares with a bacterial peritonitis rate of 2.51 episodes per patient-year in the 33 patients who did contract fungal peritonitis (data includes only the time up to the first episode of fungal peritonitis). Table 2 outlines the causative fungi in our patients. Candida remains the most frequent or-
ganism; this conforms with the published experience. 1- 21 Eighteen patients had an episode of bacterial peritonitis in the month preceding their fungal peritonitis; these patients, and four others who had undergone abdominal surgery in the previous week, all received intraperitoneal antibiotics preceding their fungal peritonitis. This registered an antibiotic exposure rate of 57.8% compared with 69.3% (N = 176) of cases from a recent review. 1
NAGAPPAN, COLLINS, AND LEE
494 Table 2. Causative Fungi in Our 38 Episodes Causative Fungal Organisms Other
Candida Calbicans C parapsilosis C guilliermondii tropicalis (not speciated)
13 (34%) 8(21%) 1 (2.6%) 1 (2.6%) 3(8.0%)
Rhodotorula Torulopsis Fusarium Saccharomyces Paecilomyces variotii Trichosporon cutaneum
5 (13%) 3(8.6%) 1 (2.6%) 1 (2.6%) 1 (2.6%) 1 (2.6%)
Three patients suffered coexistent fungal infections (patient 1, nail; patient 6, oral; patient 25, skin and nail). Nineteen of our 33 patients are women; none of them had documented preceding or coexistent vulvovaginal candidiasis. However, we did not perform routine vaginal examinations in asymptomatic individuals. Three patients had discrete recurrences offungal peritonitis, each episode being separated from the other by 2 or more months. Case 12 suffered fungal peritonitis in the 10th, 40th, and 52nd months of CAPO. Case 29 suffered three separate episodes of Candida parapsilosis peritonitis; following her third episode, she commenced oral ketoconazole prophylaxis 100 mg weekly and has had no recurrence for more than 30 months. Therapy and Outcome Antifungal therapy with subsequent catheter removal. Six episodes were managed in this way, five in the early part of our PO experience (Table 1) and one recently (a patient who refused any surgery for two weeks). Agents used included amphotericin, fluconazole, miconazole, and 5flucytosine. None of these patients were able to be reestablished on PO and one died of candidal
sepsis. This experience prompted a policy change to early catheter removal, usually without antifungal therapy. Peritoneal catheter removal-no antifungal therapy. Twenty-one episodes were managed with catheter removal alone, and 76% of them were successfully reestablished on PO, 14% remained permanently on hemodialysis, and 10% (N = 2) died. Peritoneal catheter removal with additional antifungal therapy. Eleven episodes were managed with early catheter removal with subsequent antifungal therapy. In general, these patients had more severe signs and symptoms than those not receiving antifungal therapy, and their hospital stay averaged 18 days compared with 11 days in those requiring catheter removal alone. Sixty-four percent were successfully re-established on PO, 18% remained on hemodialysis and 18% (N = 2) died. Deaths Five (13%) of the 38 episodes resulted in death occurring in close proximity to fungal peritonitis (Table 3). Of these, patient 11 and patient 15 had elective withdrawal of treatment; patients 17 and 33 died of candidal sepsis, while patient 24, whose catheter removal was followed by antifungal drugs, suffered an unrelated cerebrovascular accident leading to withdrawal of treatment, and died soon after. DISCUSSION
Fungal peritonitis is an uncommon, yet potentially life-threatening complication of peritoneal dialysis. Three percent of all CAPO-associated peritonitis episodes in our center, over
Table 3. Summary of Treatment and Outcome in Our Study
Treatment No.
2
3
Mode of Treatment
Peritoneal catheter removal alone Peritoneal catheter removal followed by antifungal drugs Antifungal drugs followed by peritoneal catheter removal
No. of Episodes
Average Duration of Hospital Stay (d)
Average Duration of Interim Hemodialysis (d)
CAPD
HD
Death
21
11
56
76%
14%
10%
11
18
56
64%
18%
18%
6
23
83%
17%
Outcome
FUNGAL PERITONITIS IN CAPO-AUCKLAND EXPERIENCE
111/2 years, were due to fungal infection. This compares with the 2% to 10.2% incidence reported in the literature. 1-21 Many investigators recommend early catheter removal and antifungal therapy. 1-21 However, some have administered antifungal therapy in an attempt to avoid catheter removal. Others have removed the peritoneal catheter without any follow-up antifungal therapy. Where antifungal therapy has been used, there have been a number of different agents employed for different durations by different routes of administration (intravenous, intraperitoneal, or oral). Indeed, no clear consensus emerges on the use of antifungal therapy in this condition. The treatment objectives for fungal peritonitis are to limit morbidity, to prevent mortality, and return the patient, if possible, to CAPO. After our early unsuccessful experience in managing patients with antifungal therapy without initial catheter removal, we moved to a policy of early catheter removal. From our review of the literature, we found that many centers have administered antifungal drugs without catheter removal and a variety of regimens were used. I -4,15-19 Among 41 cases reported, 41.4% continued on peritoneal dialysis, while 21.9% required transfer to hemodialysis. Among those treated with antifungal drugs as the solitary mode of therapy, 36.6% of the patients died due to fungal peritonitis. In our group managed with peritoneal catheter removal without subsequent antifungal therapy (Table 3), 76% of patients were able to be reestablished on PD. These results compare favorably with an analysis of the literature on patients managed with catheter removal alone (N = 53)2,4,6,12,14,20 showing only 18.8% reestablished on peritoneal dialysis, 45.2% remained on hemodialysis, and a mortality rate of 18.8% (Table 4). The final outcome was unavailable in 16.9% of patients, half of them reported only as "cured."
495
We did not use antifungal therapy unless symptoms were severe and persistent. In our series of 38 episodes, 11 patients were managed with peritoneal catheter removal followed by antifungal drugs (Table 3) and 64% of them returned to CAPO. These results also compare favorably with an analysis of the literature on this management strategy (N = 40) (Table 4), which shows only 25% being reestablished on PO and 45% remaining on hemodialysis. Fifteen percent died due to fungal peritonitis and the outcome was unavailable in another 15%. Various antifungal drug regimens were used by those reporting this strategyl-9; most used amphotericin alone or in combination with/followed by an azole derivative. Strict comparisons of drug regimens and outcome are difficult because of the wide array of doses and routes of administration. It is widely accepted that the peritoneal catheter provides a conducive site for microbial colonization. Electron microscopy of removed peritoneal catheter shows organisms embedded in an amorphous matrix on the surface of the catheter3,22; this renders effective antibiosis difficult. Where fungal colonization of the peritoneal catheter has occurred, further difficulties arise because of the more complex structure offungi,23 rendering antimicrobial therapy less effective. 24 It is logical to remove the catheter as early as possible, thus removing the main locus of sequestral colonization and allowing the normal defense mechanisms of the peritoneum to clear the remaining infection. In 21 episodes with mild symptoms, we did not use antifungal therapy on this basis. Our results show that 16 patients in this group were able to be successfully recommenced on PO after a period on hemodialysis and that only two patients died, both as a result of an elective decision to withdraw treatment for psychosocial reasons.
Table 4. Summary of a Review of the Literature on Fungal Peritonitis in CAP01- 21 Outcome
Mode of Treatment
Catheter removal alone Catheter removal followed by drugs Drugs (various routes) followed by catheter removal Antifungal drugs alone
No. of Episodes
CAPD
HD
Death
Not Available
53 40
18.8% 25%
45.2% 45%
18.8% 15%
16.9% 15%
61 41
18% 41.4%
54% 21.9%
15% 36.6%
13%
496
NAGAPPAN, COLLINS, AND LEE
We believe this outcome vindicates our policy. However, an analysis of the literature l - 21 (Table 4), shows similar death rates with all of the treatments that included catheter removal, ranging from 1.5 to 2 times the mortality rate we encountered with catheter removal alone in mild cases. We cannot offer an explanation for this discrepancy, except to state that many of these series l -2 1 were small, reflecting limited experience with this problem. The majority of our patients were able to be reestablished on PO after a rest period on hemodialysis. The rate of our reestablishment on PO was much higher than that found in the literature and makes the point that most patients who suffer fungal peritonitis can be completely cured.
It is likely that in many centers an attempt is not made to reestablish the patient on PO; in the light of our experience, some groups may wish to review their policies in this matter. In conclusion, we state that early peritoneal catheter removal, with antifungal medications only in patients with severe or persistent symptoms, can be associated with a low mortality and a high rate of successful reestablishment of CAPO. ACKNOWLEDGMENT The authors acknowledge the assistance of Dr Bremner and the staff of the clinical microbiology laboratory at Auckland hospital for fungal identification, and Dr Ellis-Pegler, Infectious Diseases Specialist at Auckland Hospital, for valuable comments.
REFERENCES I. Cheng IKP, Fang GX, Chan TM, et aI: Fungal peritonitis complicating peritoneal dialysis: Report of27 cases and review of treatment. Q J Med New Series 71(265):407-416, 1989 2. Fabris A, Biasioli S, Borin D, et al: Fungal peritonitis in peritoneal dialysis: Our experience and review of treatments. Perit Dial Bull 4:75-77, 1984 3. Kerr CM, Perfect JR, Craven PIC, et aI: Fungal peritonitis in patients on continuous ambulatory peritoneal dialysis. Ann Intern Med 99:334-337, 1983 4. Powell D, San Luis E, Calvin S, et al: Peritonitis in children undergoing CAPD. Am J Dis Child 139:29-32, 1985 5. Vargemezis V, Papadopoulou ZL, Liamos H, et al: Management of fungal peritonitis during CAPD. Perit Dial Bull 6:17-20, 1986 6. Forwell MA, Smith WG, Tsakiris D, et aI: Morbidity of fungal peritonitis. Contrib NephroI57:11O-113, 1987 7. Tapson JS, Mansy H, Fr eeman R, et aI: The high morbidity of CAPD fungal peritonitis-Description of 10 cases and review of treatment strategies. Q J Med New Series 61(235):1047-1053, 1986 8. Morford DW, Currie J: Eight cases offungal peritonitis during 765 patient months of CAPO. III International symposium on Peritoneal Dialysis, Washington, DC, June 17-20, 1984 9. Enriquez JL, Kalia A, Travis LB: Fungal peritonitis in children on peritoneal dialysis. J Pediatr 117:830-832, 1990 10. Steinberg SM, Cutler SJ, Nolph KD, et al: Acomprehensive report on the experience of patients on CAPD for the treatment of end-stage renal disease. Am J Kidney Dis 4:233241 , 1984 II. Bayer AS, Blumenkrantz MJ, Montgomerie JZ, et aI: Fungal peritonitis-Report of22 cases and review of the English literature. Am J Med 61 :832-839, 1976 12. Khanna R, Oreopoulos DG, Vas S, et al: Fungal peritonitis in patients undergoing chronic intermittent or continuous ambulatory peritoneal dialysis. Proc Eur Dial Transplant Assoc 17:291-296, 1980
13. Keogh JAB, Carr ME, Murray F, et al: Treatment of fungal peritonitis in CAPD patients using peritoneal lavage. Perit Dial Bull 5:67-69, 1985 14. Rodriguez-Perez JC: Fungal peritonitis in CAPDWhich treatment is best? Contrib Nephrol 57: 114-121, 1987 15. Johnson RJ, Ramsey PG, Gallagher N, et al: Fungal peritonitis in patients on peritoneal dialysis: Incidence, clinical features and prognosis. Am J Nephrol 5: 169-175, 1985 16. Benevent D, Peyronnet P, Lagarde C, et al: Fungal peritonitis in patients on CAPD. Three recoveries in 5 cases without catheter removal. Nephron 41 :203-206, 1985 17. Zaruba K, Peters J, Jungbluth H: Successful prophylaxis of fungal peritonitis in patients on CAPD: Six years' experience. Am J Kidney Dis 17:43-46, 1991 18. Cecchin E, De Marchi S, Panarello G , et al: Chemotherapy and/or removal of peritoneal catheter in the management of Fungal peritonitis complicating CAPD. Nephron 40:251-252, 1985 19. Struijk D, Krediet RJ, Boeschoten EW, et al: Antifungal treatment of candida peritonitis in CAPD patients. Am J Kidney Dis 9:66-70, 1987 20. Eisenberg ES, Leviton I, Soeiro R: Fungal peritonitis in patients receiving peritoneal dialysis. Experience with II patients and review of the literature. Rev Infect Dis 8:309321 , 1986 21. Pollock CA, Ibels LS, Caterson RJ, et aI: CAPO-Eight years of experience at a single center. Medicine (Baltimore) 68:293-307, 1989 22. Marrie TJ, Noble MA, Costerton JW: Examination of the morphology of bacteria adhering to peritoneal dialysis catheters by scanning and transmission electron microscopy. J Clin MicrobioI18:1388-1398, 1983 23. Huppert M, Macpherson D, Cazin J : Pathogenesis of Candida albicans infection following antibiotic therapy. The effect of antibiotics on the growth of C albicans. J Bacteriol 65:171-177, 1953 24. Edwards JE Jr: Invasive Candida infections-Evolution ofa fungal pathogen. N Eng! J Med 324: 1060-1062, 1991
