Advances in Peritoneal Dialysis, Vol. 27, 2011

Removal of the Peritoneal Dialysis Catheter Because of Gastrointestinal Disease in Patients on Continuous Ambulatory Peritoneal

Eriko Kojima, Tsutomu Inoue, Keita Sueyoshi, Takahiko Dialysis: A Single-Center Sato, Masahiro Tsuda, Tomohiro Kikuta, Yusuke Watanabe, Hiroshi Takane, Tsuneo Takenaka, Hiromichi Suzuki Case Series

We previously reported that peritoneal dialysis Key words (PD)–associated peritonitis is a major cause of PD Catheter removal, gastrointestinal disease, encapsulating catheter removal. Another major cause is disease of peritoneal sclerosis, contrast-enhanced computed the gastrointestinal tract, including neoplasm and tomography perforation. In the present study, we reviewed the records of patients who underwent catheter removal Introduction at our hospital for reasons other than peritoneal Catheter removal is one of the major factors in the infection—and for gastrointestinal disease in par- technical failure of peritoneal dialysis (PD), even ticular. Data were collected from the records of when adequate dialysis is performed. Previously, our patients who received continuous ambulatory PD group reported that the main cause of PD catheter (CAPD) therapy between 2004 and 2010 at the removal is PD peritonitis (1), and electron microscopy Department of Nephrology, Saitama Medical Uni- revealed biofilm formation on catheters removed be- versity. Mean duration of CAPD was 6.2 ± 4.7 years, cause of antibiotic-resistant infection (2). In our center, and mean age at onset was 64.5 ± 9.6 years. During another major cause of catheter removal is urgent or the investigation period, catheters were removed elective surgery for disease of the gastrointestinal tract, from 13 patients (4 men, 9 women) because of including neoplasm and perforation. Identification of gastrointestinal disease: gastric cancer in 3 cases, a gastrointestinal disease requiring laparotomy often colon cancer in 3 cases, perforation of the lower means conversion to hemodialysis and (long-term) gastrointestinal tract in 3 cases, and other reasons cessation of PD, because the PD catheter is removed. in 4 cases. Examination of pathology specimens The idea behind this strategy is that foreign bodies obtained from 6 patients—including 1 in whom should always be removed from infected areas except contrast-enhanced computed tomography indicated under exceptional circumstances (3). the presence of encapsulating peritoneal sclerosis After surgery, patients in these cases generally (EPS)—revealed mild fibrosis in the subserous face two unusual problems specifically related to PD: layer. No patient died of infection after a surgical complications associated with PD, and development procedure. Moreover, throughout the observation of encapsulating peritoneal sclerosis (EPS) over a period, no patient developed new EPS or postop- period of years. erative ileus. The present study suggests that CAPD One concern associated with PD is the possibility of itself seems to be free of untoward effects during the infection after a surgical procedure. To the best of our postoperative course in these patients. knowledge, only one report has presented a case series of PD patients who underwent a surgical procedure for From: Department of Nephrology, Saitama Medical Uni- gastrointestinal disease. That report summarized the versity, Saitama, Japan. data for 15 PD patients who underwent laparotomy for 78 Kojima et al. viscus perforation (4). The report noted that multiple All operations were performed in the Department organisms were found in PD fluid cultures from 12 of of Surgery at our hospital. Preoperative abdominal the 15 patients with perforation. CT imaging with or without contrast was routinely Bioincompatible glucose-containing dialysis fluid performed for all patients. No surgical deaths may also negatively affect the environment of the peri- occurred during the study period, and all patients toneal cavity. It is not known whether PD, performed recovered in the intensive care unit after surgery. until surgery, influences the postoperative course. For patients showing hypotension (systolic arterial However, it seems highly unlikely that PD could have pressure < 100 mmHg), marked tachycardia (120 bpm, a positive impact on an infection that occurs after a including atrial fibrillation), severe acidemia (pH surgical procedure. < 7.100 in arterial blood gas analysis), hyperkalemia In regard to EPS, even if there is no evidential (>7.0 mEq/L), and hypoxemia with excess fluid benefit in screening for a “pre-EPS” state using clinical with or without pulmonary edema, we usually indicators or computed tomography (CT) imaging (5), used a temporary vascular catheter to introduce we nevertheless usually evaluate changes in peritoneal continuous hemodialysis immediately after the membrane transport characteristics or other factors, operation. Otherwise, patients underwent intermittent and we sometimes ask patients, especially long-term hemodialysis within 2 days after surgery, also using a PD patients, to wash the peritoneal cavity for a certain temporary vascular catheter. When patients were able to period to prevent EPS. Even after PD is discontinued, take nourishment by mouth, they were admitted to the EPS may progress, particularly in patients whose CT kidney center for conversion to hemodialysis using an findings demonstrate abdominal signs or some of the arteriovenous fistula. We collected data on prognosis at features of EPS. In cases of antibiotic-resistant infec- 1 October 2010. Informed consent was obtained from tion, catheter removal is probably the best method to each patient. The mean period of observation was 3.7 prevent EPS progression. By contrast, it is uncertain ± 2.3 years (range: 7.1 – 0.5 years). whether accidental cessation of PD will subsequently lead to an undesirable clinical course. Results In the present study, we summarize data from During the study period, catheters were removed 13 patients who simultaneously underwent laparo- from a total of 13 patients (4 men, 9 women) because tomy and PD catheter removal, not for antibiotic- of gastrointestinal disease. Table I shows the clinical resistant peritonitis, but for gastrointestinal disease, characteristics of these patients. The mean age at dis- including neoplasm. ease onset was 64.5 ± 9.6 years. The underlying cause of end-stage renal disease was nephrosclerosis in 7 Methods patients. Mean duration of PD was 6.2 ± 4.7 years. The From January 2004 to September 2010, all patients gastrointestinal diseases were gastric cancer (n = 3), undergoing continuous ambulatory PD (CAPD) at colon cancer (n = 3), perforation of the lower gastro- our kidney center whose PD catheters were removed intestinal tract (n = 3), and others (n = 4). In patient 7, because of gastrointestinal disease were eligible for peritoneal calcification and very mild, but not incon- the study. All patients received outpatient treatment at siderable, adhesion of the intestine was revealed by least once monthly at our outpatient clinic throughout the preoperative contrast-enhanced CT imaging, the period during which they were on PD. The PD and thus EPS was diagnosed. In all cases, excised dose (fluid volume and exchange frequency) was specimens underwent pathology examination. Vari- adjusted to maintain a weekly creatinine clearance of ous degrees of subserous fibrosis accompanied by approximately 60 L. If the weekly creatinine clearance chronic inflammatory changes and occasionally fibrin fell below 45 L despite a full dose of PD, once-weekly deposition were observed in peritoneum in 6 patients, hemodialysis was introduced (6), and those patients including patient 7. were excluded from the study. Daily dietary protein During the study period, 2 of the 13 patients intake was approximately 1 g per kilogram body died. Patient 1, a 63-year-old woman, died from weight, and daily energy intake was more than 30 kcal severe acute necrotizing pancreatitis not associated to 35 kcal per kilogram body weight. Daily salt intake with PD. Patient 2, an 80-year-old woman, already was restricted to between 7 g and 9 g (6). had advanced cancer when PD was initiated because Removal of PD Catheter for GI Disease 79 b b b b b b b b b b b Death Death Alive Alive Alive Alive Alive Alive Alive Alive Alive Alive Alive Outcome findings Pathology of peritoneum Almost normal Almost normal Almost normal Almost normal Almost normal Almost normal Almost normal fibrin deposition Subserous fibrosis Subserous fibrosis Subserous fibrosis Subserous fibrosis Subserous fibrosis Subserous fibrosis, Total Type of Type surgery Hernioplasty Laparoscopic sigmoidectomy with colostomy Sigmoidectomy cholecystectomy Cholecystectomy Abscess drainage Distal gastrectomy Distal gastrectomy Distal gastrectomy, nephroureterectomy Hartmann procedure Low anterior resection Low anterior resection Segmental rectal resection abscess Sigmoid Diagnosis perforation pancreatitis Necrotizing colon cancer Rectal cancer Rectal cancer Gastric cancer Gastric cancer Gastric cancer Inguinal hernia Intra-abdominal Rectal perforation Rectal perforation Renal pelvic cancer Sigmoid diverticulum 4 5 6 4 2 12 19 4.5 0.5 7.5 6.5 5.5 4.5 (years) PD duration A. Renal Diabetic Diabetic diagnosis Polycystic nephropathy nephropathy kidney disease Lupus nephritis Lupus nephritis Nephrosclerosis Nephrosclerosis Nephrosclerosis Nephrosclerosis Nephrosclerosis Nephrosclerosis Nephrosclerosis IgA nephropathy IgA = immunoglobulin Sex Male Male Male Male Female Female Female Female Female Female Female Female Female a 63 80 67 56 59 43 72 78 62 62 61 69 66 Age Clinical characteristics of the patients (years)

i

At the time of writing. At the time of surgery. In all patients, the catheter was removed at the time of surgery for the primary gastrointestinal disease. In all patients, the catheter was removed at time of surgery At the time of surgery. 1 2 6 11 3 ID 13 8 5 9 Pt 4 7 12 10

a b PD = peritoneal dialysis; IgA t a b l e 80 Kojima et al. of symptoms of chronic uremia. Six months later, is associated with high morbidity related to bowel she was rushed to hospital for massive melena, and obstruction and malnutrition. The reported mortality emergency surgery was required. She eventually died is around 50%, usually within 12 months of diagnosis, from pneumonia with malnutrition. although not all deaths are attributable to EPS itself To date, no new EPS has developed after removal (10–12). Microscopically, EPS is characterized by of the PD catheter. No patient has required further PD, chronic inflammation and fibrosis of the peritoneum. and all patients continue to receive hemodialysis at our Thus, there is nothing surprising in the notion that kidney center or other outpatient clinics. the presence of both inflammation (for example, peritonitis) and long-term PD markedly enhances the Discussion odds of subsequent EPS development. From January 2004 to September 2010, the period of By contrast, it should be emphasized that all this study, 216 patients were newly introduced to PD current data indicate that most patients receiving at our kidney center, and about 250 patients regularly long-term PD do not develop EPS (5). Most PD attended our outpatient clinic. Surgeons at our hospital patients have some degree of peritoneal sclerosing are familiar with PD patients, and we maintain close inflammation, and our study showed that obvious coordination between the Department of Surgery peritoneal alterations were present in approximately and the kidney center. It should be noted that, in our half the study patients. As expected, such microscopic experience, PD seems to have no negative impact peritoneal alterations do not indicate EPS progression, on the postoperative course. When patients undergo and similarly, the presence of gastrointestinal disease laparoscopic surgery at our hospital, the PD catheter does not affect the incidence of EPS. is removed, without exception, in consideration of the potential for infection after a surgical procedure. To Conclusions our knowledge, only 1 case report has described lap- We present data for 13 patients who experienced aroscopic therapy in a CAPD patient with perforated catheter removal because of gastrointestinal disease appendicitis in which the peritoneal catheter was not requiring surgical intervention. Our results suggest removed (3). It is significant, however, to consider that a history of PD presents no disadvantages in the the actual situation with respect to PD, because the postoperative clinical course. Moreover, as reported modality is usually eschewed as a renal replacement previously, many PD patients show peritoneal altera- therapy if the subject has a history of laparotomy. On tions characterized by subserous fibrosis and fibrin that basis, our decision to remove the PD catheter is deposition without any increased risk of EPS. thought to be reasonable. The age of new dialysis patients is rapidly increas- The present study also demonstrates that subserous ing, with one third of patients in Japan being 75 years fibrosis with chronic inflammation with or without of age or older. Increased age has been associated with fibrin deposition is not necessarily an indication of the an increased incidence of malignant gastrointestinal future progression of EPS in PD patients. Many reports tumors and viscus perforation, which would imply have discussed peritoneal morphology changes during that aging is associated with an increased need for PD, demonstrating that mesothelial cells have an surgery. However, there is no evidence to support the active role in the structural and functional alterations withholding of PD as a treatment option through fear to the peritoneum during PD (7). Significant structural of the risk of postoperative complications. changes have been shown to become worse with CAPD We conclude that PD does not, in itself, complicate treatment, and alterations have also been observed the postoperative course and prognosis for survival. in the peritoneal membrane of uremic patients (8). In addition, with the exception of 1 patient in whom The average peritoneal thickness (resulting from contrast-enhanced CT had already revealed EPS, none subserous fibrosis with chronic inflammation and of 5 other patients showing changes related to subse- fibrin deposition) is greater in uremic patients and rous fibrosis and chronic inflammation in excised progressively thickens with increase in the duration specimens have developed sclerogenic peritonitis of PD (9). Thus, PD treatment itself is thought to to date. Therefore, the foregoing pathology findings have a very strong impact on the progression of do not always indicate future progression of EPS in peritoneal sclerosis. Encapsulating peritoneal sclerosis PD patients. Removal of PD Catheter for GI Disease 81

Disclosures of mesothelial cells. N Engl J Med 2003;348:403–13. The authors have no competing interests to declare. 8 Savidaki I, Karavias D, Sotsiou F, et al. Histologic change of peritoneal membrane in relation to ad- References equacy of dialysis in continuous ambulatory perito- 1 Uchida K, Shoda J, Sugahara S, et al. Comparison neal dialysis patients. Perit Dial Int 2003;23(suppl and survival of patients receiving hemodialysis and 2):S26–30. peritoneal dialysis in a single center. Adv Perit Dial 9 Honda K, Hamada C, Nakayama M, et al. Impact of 2007;23:144–9. uremia, diabetes, and peritoneal dialysis itself on the 2 Nodaira Y, Ikeda N, Kobayashi K, et al. Risk factors pathogenesis of peritoneal sclerosis: a quantitative and cause of removal of peritoneal dialysis catheter in study of peritoneal membrane morphology. Clin J Am patients on continuous ambulatory peritoneal dialysis. Soc Nephrol 2008;3:720–8. Adv Perit Dial 2008;24:65–8. 10 Rigby RJ, Hawley CM. Sclerosing peritonitis: the 3 Wesseling KM, Pierik RG, Offerman JJ. Laparo- experience in Australia. Nephrol Dial Transplant scopic therapy in a CAPD patient with perforated ap- 1998;13:154–9. pendicitis without removal of the peritoneal catheter. 11 Balasubramaniam G, Brown EA, Davenport A, et al. Nephrol Dial Transplant 2003;18:1929–30. The Pan-Thames EPS study: treatment and outcomes 4 Wakeen MJ, Zimmerman SW, Bidwell D. Viscus per- of encapsulating peritoneal sclerosis. Nephrol Dial foration in peritoneal dialysis patients: diagnosis and Transplant 2009;24:3209–15. outcome. Perit Dial Int 1994;14:371–7. 12 Brown MC, Simpson K, Kerssens JJ, Mactier RA on 5 Brown EA, Van Biesen W, Finkelstein FO, et al. behalf of the Scottish Renal Registry. Encapsulating Length of time on peritoneal dialysis and encapsulat- peritoneal sclerosis in the new millennium: a national ing peritoneal sclerosis: position paper for ISPD. Perit cohort study. Clin J Am Soc Nephrol 2009;4:1222–9. Dial Int 2009;29:595–600. 6 Okada S, Inoue T, Nakamoto H, et al. Residual renal function plays an important role in regulating para- Corresponding author: thyroid hormone in patients on continuous ambulato- Hiromichi Suzuki, m d , 38 Morohongo, Moroyama- ry peritoneal dialysis. Adv Perit Dial 2007;23:150–4. machi, Iruma-gun, Saitama 350-0495 Japan. 7 Yáñez–Mó M, Lara–Pezzi E, Selgas R, et al. Perito- E-mail: neal dialysis and epithelial-to-mesenchymal transition [email protected]