Introduction
Peritoneal dialysis is a renal replacement therapy that is often preferred by patients with end-stage renal disease as it usually has less impact on their daily routine than hemodialysis [
1]. Peritoneal dialysis is also favored in the pediatric population. Although in many countries the peritoneal dialysis-first strategy is often a default policy nowadays, kidney transplantation is still considered a superior method that usually guarantees the best possible quality of life [
2]. Therefore, an increasing number of patients undergoing peritoneal dialysis eventually receive their first renal graft. An ongoing debate concerns the management of peritoneal dialysis catheters during transplantation [
3]. One of the questions that need to be answered is whether the catheter should be removed at the time or after the transplant procedure.
Current policies differ in many transplant centers. Practices range from routine catheter removal simultaneously with transplantation [
4,
5] to postponing it until good renal function is achieved or until a particular postoperative time [
6,
7]. The problem is mentioned briefly in the European Best Practice Guidelines (EBPG) for Peritoneal Dialysis from 2005 [
8]. Over a decade ago, the EBPG Expert Group on Peritoneal Dialysis stated that leaving peritoneal dialysis catheters in situ for 3–4 months is acceptable even with a functioning renal graft, although quicker removal was advised if possible. However, evidence back then was poor and the methodology used to create these guidelines is not considered robust [
9,
10].
To solve the dilemma of peritoneal dialysis catheter removal, several aspects should be considered. The first one is the choice of a proper measure to evaluate both interventions. The main argument for leaving catheters in place is having dialysis access ready to use in case of delayed graft function or graft failure [
11]. Therefore, we believe that there is a need to compare the overall need for dialysis early after transplantation between the two groups of patients, with catheters left in situ and removed simultaneously with transplantation. Furthermore, the used dialysis modality should be evaluated to confirm the feasibility of peritoneal dialysis in such cases. Additionally, complications associated with each procedure should be analyzed. Finally, such factors as patients’ quality of life and costs should be considered.
We aimed to evaluate available evidence about the peritransplant management of peritoneal dialysis catheters using a critical and systematic approach. We focused on the timing of their removal. A systematic review and random effects meta-analysis of non-randomized studies of intervention was conducted.
Discussion
In this systematic review, we summarized the current evidence on the peritransplant management of patients with peritoneal dialysis catheters. The quality of available data is poor and the clinical question of whether peritoneal dialysis catheters should be removed at the time of surgery or not cannot be definitely answered. However, experience emerging from the published articles might be useful for clinicians and patients in the decision-making process.
The lack of evidence-based guidelines leads to the presence of varying policies adopted by different centers and specialists. Some routinely leave the catheters in situ [
4,
5,
36‐
40] while others remove them as a rule during the transplant procedure [
6,
7]. There are certain situations when indwelling peritoneal catheters should be removed before or at the time of kidney implantation, including infectious (peritonitis, exit-site/tunnel/inner cuff infection), non-infectious and mechanical (broken catheter, catheter migration, pericatheter leak, flow dysfunction, peritoneal loss of function, peritoneal membrane breach, pleuri-peritoneal communication, sclerosant peritonitis), and some rare complications (allergic reaction, organ erosion, severe infusion/pressure pain, hemoperitoneum) [
41,
42]. Additionally, some authors suggest that their immediate removal ought to be considered in living donor procedures that are associated with a lower risk of delayed graft function [
3,
11,
30,
31]. Some protocols of the evaluated studies had already implemented that as a rule, which resulted in the different prevalence of living donor recipients in the experimental and the control groups. We suspect that it may be a confounding factor, but its true effect could not be formally determined. Maiorca et al. [
32] analyzed only deceased donor transplants. It is the only study with odds of needing dialysis lower for patients with peritoneal dialysis catheters left in situ (Fig.
4). However, the control group consisted of only four patients. Moreover, Malagon and Hogg [
30] described removing the catheter only in living-donor kidney recipients. None of them had delayed graft function but two developed surgical complications which finally led to the need for dialysis. Other identified indications for simultaneous catheter removal included concurrent renal and pancreas transplantation as well as having peritoneum compromised for other reasons [
3,
11,
43].
It is believed that leaving a peritoneal catheter in situ can be considered when the probability of its subsequent use is substantial. It applies mainly to deceased donor transplant recipients in whom the prevalence of delayed graft function is between 10 and 30% [
44]. The risk is even higher for organs from extended criteria donors [
35]. However, there is no consensus on how long the removal should be postponed (some studies evaluated by us report that peritoneal dialysis catheters were used mostly within the first post-transplant month [
30,
34,
40]). Extended time is associated with more complications, mainly infectious, that in this group are particularly dangerous owing to immunosuppressive therapy [
7,
27,
34,
45]. In our meta-analysis, we calculated that pooled prevalence of catheter-related infections in patients with Tenckhoff catheters left intact was 10.2%; 95% confidence interval, 6.2 to 16.1% (Fig.
6b). Owing to data incompleteness, we could not evaluate the correlation between the time for which catheters were left in place and the incidence of catheter-related infections. Some authors still argue that in many cases these conditions (peritonitis, exit-site, and tunnel infections) can be managed with antibiotics and a sensible number of patients without complications still benefit from resuming peritoneal dialysis in case of early graft failure [
33,
46]. However, no objective (medical and economic) data are available to support this opinion.
In a recent commentary on a case–control study by Gardezi et al. [
47], Issa and Lakhani [
48] proposed an algorithm for the management of peritoneal dialysis catheter in patients undergoing renal transplantation. These experts suggest leaving peritoneal dialysis catheters in place only when (1) delayed graft function is highly expected and (2) peritoneum breach as well as catheter-related infection is absent. Otherwise, it should be removed at the time of transplant or before hospital discharge. Nevertheless, the latter exposes the patient to the risk of additional surgical procedures. Moreover, the methods to predict delayed graft function in individual patients are still a matter of debate. Issa and Lakhani propose the use of Kidney Donor Profile Index (KDPI) along with several other predictive factors (such as long cold ischemia time and recipient’s history of obesity, diabetes mellitus, prior allosensitization, long dialysis vintage, and waiting time) [
48]. However, predictive models for delayed graft function are not well clinically validated. Furthermore, KDPI was not meant and developed to predict delayed graft function. The available evidence shows that these two factors are not even well correlated [
49,
50]. The discussed algorithm should be precisely investigated in a prospective randomized clinical trial.
Using random effects meta-analysis, we calculated that patients requiring post-transplant dialysis more likely had their dialysis catheters left intact during transplantation than those not needing it (pooled unadjusted odds ratio, 2.21; 95% confidence interval, 1.03 to 4.73, Fig.
4). This finding suggests that protocols followed by the centers, from which the data were obtained, are effective to a certain extent. More individuals with peritoneal dialysis catheters left in situ needed dialysis (pooled prevalence, 20.9%; 95% confidence interval, 13.6 to 30.7% vs. 12.4%; 95% confidence interval, 5.6 to 25.2%, Fig.
5), as expected.
When dialysis was needed early after transplantation and a peritoneal dialysis catheter had not been removed at the time of surgery, some studies report that in up to 80–100% of cases peritoneal dialysis was feasible [
11,
27,
31,
32]. However, in the group studied by Warren et al. [
27], in five of such patients, the peritonitis occurred, and three others had to be converted to hemodialysis. In a report by Rizzi et al. [
51] who analyzed 313 patients whose catheters were left during transplantation, almost 16% of them required dialysis. Among them, only 33% could benefit from peritoneal dialysis access being available — others were referred to hemodialysis. Taking into account that only for a few studies are these data known, the high prevalence of these catheters’ utilization might be overestimated. In one study (excluded from the final analysis due to lack of a control group), patients requiring dialysis right after transplantation were electively hemodialyzed even with a peritoneal dialysis catheter left in situ [
35]. Only those with primary non-function were transferred to peritoneal dialysis immediately (2.5%, 3/120). Such an approach stands against the fact that peritoneal dialysis catheters might be safely used early after renal transplantation [
45,
52].
Patients whose peritoneal catheters had been removed at the time of transplantation might also require dialysis. They are usually treated with hemodialysis which could possibly lead to some complications. However, no undesired events were reported in the analyzed studies [
11,
27]. Furthermore, McGregor [
3] pointed out that many patients undergoing renal transplantations have a central line placed by an anesthesiologist. In such cases, it is relatively simple and safe to replace it with a dialysis line when necessary. Some individuals may as well be managed with peritoneal dialysis. Malagon and Hogg [
30] described two living-donor graft recipients who needed dialysis because of surgical complications (kidney laceration in one case and vascular damage in the second). They had new peritoneal dialysis catheters placed during their grafts’ repair surgeries and the therapy was uneventful.
The discussion about the timing of indwelling peritoneal catheter removal involves consideration of some costs and benefits. Most patients with such dialysis access left in situ during transplantation eventually require a second intervention to take it out. There are various complications linked to the presence of a foreign body in the abdomen. It is not always feasible to perform peritoneal dialysis even with the Tenckhoff or similar catheter available, as discussed above. The benefits are avoiding potential problems associated with vascular access for hemodialysis, the possibility of quick resumption to peritoneal dialysis in case of temporal or permanent graft failure, and having a route ready to drain potential ascites. However, owing to the unpleasant experience with catheter-related complications in renal graft recipients with catheters left in place, some authors declared to switch to a routine of removing them at the time of transplantation [
3,
27]. As the certainty of the available is very low, we believe that ultrasound examination should be considered in select uncertain cases [
53].
Limitations
This study is limited by the lack of high-quality evidence. Our synthesis was based on retrospective imbalanced studies of moderate or low quality. We summarized their findings in a meta-analysis including only 338 patients. We were only able to obtain odds ratios that were not adjusted for any of the previously identified candidate confounders. We also found relatively old articles which might not well represent current standards of care. Our report proves that further evidence is urgently needed to answer the clinical dilemma of peritransplant peritoneal dialysis catheter management. The medical community might benefit from our experience when designing future trials. As we demonstrated, there are numerous potential confounders to be controlled for when conducting retrospective studies about this problem. Taking into account that 45% of the excluded papers were lacking a control group, it is important to avoid this approach, if not justified. However, it is highly desirable in this case to design and conduct high-quality and adequately powered randomized controlled trial in both the adult and pediatric populations.
Publisher's note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.