Spontaneous peritoneal dialysis catheter expulsion: A case series

Abstract

There are a significant number of mechanical complications related the peritoneal dialysis (PD) catheter and early diagnosis is key. Between them, the spontaneous extrusion of the catheter is quite uncommon but it should be included among the possible complications of the technique, given that it can be related to frequent situations in PD patients like abdominal wall issues, past immunosuppression or PD fluid leakage. In 25 years our Unit had three cases, probably related to peritoneal fluid leakage, exit site infection and past steroid treatment. Careful examination of abdominal wall preimplantation, adequate break-in period to allow maturation and early detection of exit site infection to start treatment in a timely fashion could help to prevent this rare complication. The complete extrusion of the catheter does not preclude from continuing PD treatment.

Introduction

Mechanical complications of peritoneal dialysis (PD) are a significant cause of technique failure. Amongst them, complete expulsion of the peritoneal catheter is an extremely rare complication. Spontaneous loss or complete spontaneous extrusion of the peritoneal catheter is, fortunately, a rare occurrence. The fall of the catheter has been scarcely published^1,2 and is thought to be related to poor implantation technique, poor healing around the cuffs due to infection of the exit site-tunnel or pericatheter leakage. In polyurethane catheters, the use of topical mupirocin at the exit site has been associated with structural changes in the catheter and cuffs resulting, in some cases, in catheter extrusion.^1,3 We report our Unit experience with three cases of spontaneous PD catheter extrusion, aiming to describe patient’s characteristics, discuss the possible causes that could be responsible of this complication and potential prevention measures.

Case description

CASE 1: A female in her sixties with end-stage chronic kidney disease (ESKD) secondary to non-biopsied glomerulonephritis that had never been treated with immunosuppression. Relevant past medical history included one pregnancy with normal delivery, 11 months on hemodialysis and a deceased donor kidney transplant in 1996 with immediate graft loss due to venous thrombosis. She had a straight Tenckhoff catheter (2 cuffs) inserted using the minilaparotomy technique in paramedian area, with deep cuff under abdominal rectus muscle and external cuff 2 cm from the exit site, which was placed downwardly emerging below a high-lying belt line laterally. Continuous ambulatory peritoneal dialysis (CAPD) was started 5 weeks after PD catheter insertion and she had two episodes of exit site infection secondary to Corynebacterium sp, both fully resolved with topic vancomycin eye drops for 2 weeks each time with sterile control cultures after finishing treatment. Five years later she presented with a pericatheter leak that resolved with 2 months of PD rest and transient transfer to hemodialysis. Five months after restarting PD without any issues she was admitted due to complete spontaneous fall of PD catheter. She had no signs of exit site infection, no peritonitis signs nor symptoms and no topical medications had been used. No abdominal wall ultrasound was performed. A contralateral peritoneal catheter was inserted without further complications.

CASE 2: A female in her sixties presented an acute kidney injury secondary to anti-glomerular basement membrane disease in 2002, without recovery of renal function despite adequate treatment (high-dose corticosteroid therapy). In her past medical history she had one normal pregnancy delivery and she was non obese. In June 2002 a straight Tenckhoff catheter (2 cuffs) was placed using the minilaparotomy technique with paramedian location, with deep cuff under abdominal rectus muscle and external cuff 2 cm from the exit site. Exit site was placed lateral and downwardly, emerging below a high belt line. She started CAPD afterward and 2 months later she presented with her PD catheter spontaneously out of the subcutaneous tunnel. There was evidence of exit site infection at that point (but not in previous controls) with Candida parapsylopsis that was treated with clotrimazole 1% spray for 21 days with sterile cultures after finishing treatment. No signs or symptoms of peritonitis were present, with normal cell count of PD fluid analyzed after at least 2 h of permanency and negative PD fluid cultures. No abdominal wall ultrasound was performed for her exit site infection as per our Unit’s protocol, as no tunnel infection signs were present. After its resolution a contralateral PD catheter was placed without further complications.

CASE 3: A woman in her sixties with ESKD secondary to type 2 cardio renal syndrome refractory to diuretic treatment and poor fluid management was started on CAPD in May 2017 after placing a straight Tenckhoff PD catheter (2 cuffs) using the minilaparotomy technique with paramedian location, with deep cuff under abdominal rectus muscle, exit site placed lateral and downwardly surfacing below a high belt line and external cuff approximately 2–3 cm away from it. She had given birth twice and had no previous immunosuppression nor abdominal surgeries. She presented four episodes of peritonitis with bacteriemia and cloudy peritoneal effluent, identifying Pseudomonas aeruginosa in both blood and peritoneal fluid cultures. Intraperitoneal ceftazidime and tobramycin were started after definitive antibiogram results with good clinical outcome, but after 3 weeks antibiotics were stopped and she relapsed every time until taurolidine PD catheter lock was started (nine doses given). ⁴ Five months later she had an episode of exit site infection by Pseudomonas aeruginosa, solved after a course of directed antibiotics (tobramycin and ceftazidime drops as topic treatment) as well as cuff shaving and PD catheter lock with taurolidine. She did not present any signs nor symptoms of peritonitis at that point, with clear peritoneal effluent and normal cell count on PD fluid. No abdominal wall ultrasound was performed for her exit site infection as per our Unit’s protocol. Four months afterward she developed another exit site infection by the same bacteria, but despite treatment with topic tobramycin drops for 3 weeks she continued to have positive exit site cultures without peritonitis nor tunnel infection suspicion. Finally, 1 month later she presented a spontaneous expulsion of the PD catheter. A contralateral peritoneal catheter was placed without further infectious or mechanical complications.

Discussion

PD technique is the renal replacement therapy of choice in up to 15%–20% of patients with ESKD and it has well-known mechanical complications usually related to abdominal pressure and wall issues. Spontaneous extrusion of the PD catheter, although an exceptional complication of PD, might present. Related risk factors for this complication are poor PD catheter insertion technique, suboptimal healing, exit-site infection and mupirocin ointment use in polyurethane catheters. ³

From 1996 until 2021 our Unit has had 427 patients on PD treatment, with up to 60% of PD catheters implanted by an experienced interventional nephrologist. In addition, all catheters inserted were straight Tenckhoff type with double cuffs. In this time period, we only encountered the three cases of spontaneous PD catheter extrusion presented in this paper, which is concordant with scarce previous reports.^1,2 Our cases were all female, had a planned start on PD defined as PD start >4 weeks after PD catheter insertion, which complies with the ISPD recommendations for the break-in period, ⁷ and they were all on CAPD, which is the most frequent technique type related to abdominal wall issues.^5,6 In all three, the PD catheter was implanted by an interventional nephrologist using the mini-laparotomy technique (open surgical dissection under local anesthesia) and following ISPD guideline on PD access recommendations, ⁷ such as exit site placement, superficial Dacron distance from it and a paramedian insertion of the catheter through the rectus muscle with deep catheter cuff below posterior rectus muscle sheath. Our Unit’s pre operative checklist ⁸ includes bowel voidance (enema prescription) the night before and same morning of the catheter implantation, bladder voidance just before insertion, shower and prophylactic antibiotic dose. All blood tests are checked upon admission the night before procedure and the pre operative checklist must be correctly filled. ⁸

Usually, we perform an abdominal wall ultrasound when patients are obese (above grade I-II) to assess patients wall thickness and any other issues before placing the PD catheter, which was not required in any of the above cases as per our protocol. Furthermore, none of them had signs of tunnel infection nor PD leakage diagnosed at the time extrusion occurred, thus an ultrasound of exit site or tunnel was not performed. Exit site infection in case 2 was diagnosed at the same time the patient was admitted with catheter extrusion. Cuff shaving was performed in case 3 following guidelines ⁷ due to recurrent infection by Pseudomonas aeruginosa, but it did not prevent further exit site infection episodes.

When admitted due to catheter extrusion our patients had spontaneous peritoneal fluid drainage through the tunnel at the exit site. Moreover, once expelled, the only visible change in the catheter was an almost complete disappearance of the fibrosis around both Dacron cuffs, looking like a new catheter. Splicing could not even be considered as patients had a complete catheter extrusion on admission, all of them received a prophylactic course of intravenous cefazolin and did not suffer from peritonitis.

Known risk factors for abdominal wall complications that could justify the lack of catheter anchoring in PD patients, such as advanced age, polycystic kidney disease or high body mass index, were not present in our series. ⁴ Furthermore, none of our patients had a history of abdominal surgeries or any type of damage of the abdominal wall, thus it was presumed to be normal. Possible triggers for catheter expulsion in our patients may be related to poor abdominal wall basal characteristics due to past pregnancies, chronic inflammation and catabolic state. Above that, peritoneal fluid leakage in case 1, exit site infection in cases 2 -3 and previous steroid treatment (cases 1 and 2) may have played a role in poor pericatheter healing. All these factors could have conditioned a poor anchorage of the cuffs to the subcutaneous cellular tissue, but despite the extrusion episode all patients could have another PD catheter inserted and continued on PD therapy without any further issues.

Although spontaneous extrusion is an exceptional complication of PD catheters we believe it should be included as a possible complication of the technique, and it does not preclude from continuing on PD after another catheter is placed. Careful examination of abdominal wall previous to implantation, adequate break-in period to allow maturation and early detection of exit site infection to start treatment in a timely fashion could help to prevent this rare complication, above all in patients with previous abdominal wall stresses and steroid treatment.

Conclusions

Our cases represent an atypical PD complication, possibly associated with frequent situations for our PD patients, such as exit site infection and pericatheter leak. Despite the dramatic onset of the event, patients were able to continue on PD technique. Spontaneous expulsion of the catheter should be included among the possible complications of the technique and it does not preclude from continuing PD treatment. Careful examination of abdominal wall preimplantation, adequate break-in period to allow maturation and early detection of exit site infection to start treatment in a timely fashion could help to prevent this rare complication.