Laparoscopic techniques enable peritoneal dialysis in the difficult abdomen.

BACKGROUND AND OBJECTIVES: Continuous ambulatory peritoneal dialysis has become an increasingly popular modality of renal replacement therapy. Laparoscopic placement of peritoneal dialysis catheters may help overcome previous barriers to peritoneal dialysis, such as previous abdominal surgical procedures or the presence of hernias, without incurring substantially greater risks.
METHODS: We performed a retrospective review of 120 consecutive patients who underwent attempted laparoscopic peritoneal dialysis catheter placement between July 2009 and June 2014 by a single surgeon. Patient and catheter characteristics and outcomes were compared between patients with and without complications, as well as between patients with a history of major abdominal surgery and those without such a history.
RESULTS: Laparoscopic peritoneal dialysis catheter placement was aborted in 4 patients because of an inability to safely achieve sufficient access to the abdominal cavity through dissection; these patients were excluded from subsequent analysis. The mean follow-up period was 18.8 ± 12.9 months. Fifty-five patients had a history of major abdominal surgery compared with 61 without such a history. No significant difference was observed with respect to age, race, sex, or body mass index between groups. Notably, more adjunctive procedures were required in patients with previous abdominal surgery, including adhesiolysis (60.0% vs 4.9%, P < .0001) and hernia repair (12.7% vs 1.6%, P = .026). Postoperative catheter complications were not significantly different between patients with and patients without a history of abdominal surgery (29.1% vs 32.8%, P = .667). Both unassisted (56.8% vs 65.0%, P = .397) and overall (72.7% vs 76.7%, P = .647) 1-year catheter survival rates were similar between patients with and patients without previous surgery, and the overall 1-year survival rate improved to 83.9% on exclusion of patients who stopped peritoneal dialysis for nonsurgical reasons.
CONCLUSIONS: Laparoscopic peritoneal dialysis catheter placement offers a chance to establish peritoneal dialysis access in patients traditionally viewed as noncandidates for this modality. Despite the potential risks incurred because of additional procedures at the time of catheter placement in these complicated cases, these patients can achieve good long-term peritoneal dialysis access with an aggressive surgical approach.

INTRODUCTION

The use of peritoneal dialysis (PD) as a practical alternative to renal replacement therapy first began in 1968 with the advent of a functional indwelling catheter by Tenckhoff and Schechter.[1] By the 1980s, PD had become accepted as a major modality,[2] and it has since become a popular alternative to hemodialysis for patients with end-stage renal disease (ESRD).

PD catheters had traditionally been inserted by an open approach through a small laparotomy incision. Throughout the 1990s, advances in laparoscopy led this technique to become an accepted alternative to catheter placement.[3-9] Although multiple studies have been published comparing the 2 methods of catheter insertion, to date, there has not been a consensus on which method is superior in terms of complications and catheter survival.[10] Despite the apparent equivocalness in modalities, however, certain barriers have hampered laparoscopy's widespread adoption among surgeons.

Barriers such as prior abdominal surgery[11] or the presence of abdominal wall hernias have posed a challenge to PD. Wang et al[12] found that in select patients with a history of abdominal operations, laparoscopy can be both a diagnostic and a therapeutic modality for adhesiolysis and subsequent establishment of PD. Few other reports, however, have examined the feasibility and functionality of laparoscopic PD catheter placement in such patients with a history of abdominal operations or in cases of abdominal wall hernias.

With the advancement of laparoscopic equipment and techniques, specifically techniques involving PD catheter placement, such as omentopexy, rectus sheath tunneling, catheter fixation, and adhesiolysis, improvements have been made regarding catheter-related complications and overall function.[13-15] These practices have resulted in a new and expansive armamentarium for laparoscopic PD catheter placement and, as such, have begun to mitigate previous barriers to such an operation.

This study aims to examine the characteristics and results of 120 patients who have undergone laparoscopic PD catheter placement at our institution, as well as to determine the effect of previous abdominal surgery on catheter-related outcomes.

METHODS

Institutional review board approval from the Tulane University School of Medicine was obtained. Between July 2009 and June 2014, 120 ESRD patients underwent attempted laparoscopic PD catheter implantation procedures by a single surgeon at Tulane University Hospital and Clinic. Patients were grouped by history of major abdominal surgery, and outcomes were compared. Major abdominal surgery was defined as a laparoscopic or open intraperitoneal procedure that would typically last >30 minutes. Perioperative and follow-up data were retrospectively collected from patient records and analyzed.

Surgical Technique

Intraperitoneal access was initially attempted in the periumbilical region using the Hasson technique. If peritoneal access through the umbilicus could not be attained because of scarring or adhesions, access was performed in a remote location, typically in the upper abdomen, also using the Hasson technique. Once peritoneal access was obtained, exploration was performed to assess for adhesions, abdominal wall hernias, and omental length and thickness.

Adhesions

Laparoscopic adhesiolysis was attempted in patients who had mild to moderate adhesions that could be dissected free. If dense adhesions were noted, especially in the pelvis, the procedure was aborted. Although no enterotomies occurred during the adhesiolysis procedures, our policy would have been to abort placement with a possible reattempt in about 2 weeks if an enterotomy had occurred.

Hernias

Ventral or inguinal hernias identified at the time of catheter placement were fixed concurrently. Inguinal hernias were repaired with a standard open Lichtenstein approach, whereas ventral hernias were repaired with an open approach, attempting to stay in an extraperitoneal plane. If mesh was placed, this was also placed in an extraperitoneal plane. Typically, synthetic mesh was used. If the peritoneum was entered and there was a risk of leak, a biologic mesh was placed instead. In patients who underwent concurrent herniorrhaphy, PD was performed with low volumes and delayed until 2 to 4 weeks postoperatively.

Omentum

Patients who had omentum that appeared sufficiently long to risk encasing the catheter underwent a laparoscopic omentopexy. This was performed by tacking the omentum to the falciform ligament using an Endostich device (Covidien, Mansfield, Massachusetts).

All patients in whom peritoneal access was successful had a coiled-tip, dual-cuff PD catheter placed. This was performed with an 8-mm port placed by inserting the trocar through the lower rectus sheath, thus tunneling the catheter through the sheath to provide a downward-facing entrance into the abdomen. The exit site was selected based on patient habitus and social habits. In particular, if the patient had a large pannus, wore diapers, or preferred taking baths versus showers, the exit site was placed at the subcostal or subclavicular area via a catheter extension.

Statistical Analysis

Statistical analysis was performed with the χ2 and Fisher exact tests for nominal variables and the Student t test for continuous variables. All statistical analyses were performed using SPSS software, version 19.0 (IBM, Armonk, New York).

RESULTS

One hundred twenty consecutive laparoscopic PD catheter placements were attempted by a single surgeon between July 2009 and June 2014. An inability to safely place the catheter was the only criterion for exclusion and occurred in 4 patients (3.33%) because of the presence of prohibitively dense adhesions encountered on entering the abdomen. The mean length of follow-up was 18.8 ± 12.9 months.

describes the demographic and operative characteristics of the entire cohort. The mean age of the patients was 52.4 ± 14.5 years, and the majority were male patients (56.0%) and African Americans (59.5%). The average body mass index (BMI) of the patient cohort was 30.4 kg/m2. Nearly half (47.4%) of all patients underwent previous major abdominal surgery. Revision surgery was performed in 25.0%, most of which (55.2%) occurred within the first 6 months after catheter placement. Currently, 75.0% of the catheters remain in use, with those that failed having an average time to cessation of PD of 7.3 ± 6.2 months. Ignoring the 12 patients still within 1 year from their original surgical procedure at the time of this writing, the overall 1-year catheter survival rate was 72.2% whereas the unassisted 1-year survival rate was 60.2%. As for the 27 patients who, 1 year postoperatively, were no longer undergoing PD, only 15 (13.9% of all patients) had catheter malfunctions due to operative complications. The remainder stopped PD because they received a transplant (n = 2), because of patient choice (n = 7), or because of peritoneal membrane failure (n = 3). Excluding these patients who discontinued PD for nonsurgical issues, the overall 1-year catheter survival rate was 83.9%.

Complications occurred in 30.0% of all patients without a predilection for past surgical status. The majority included adhesions, infections (7 cases of peritonitis or intra-abdominal infections, 6 superficial), catheter blockage, and catheter malposition. Adhesions were characterized as a complication post hoc once reoperation for obstructed catheters confirmed adhesions as the implicated etiology. Two new incisional hernias at trocar sites separate from the point of catheter entry occurred. The remaining minority of complications included a pleural effusion due to a Morgagni hernia, a burnt catheter, a broken catheter, 2 cases of extruded cuffs, and 1 recurrent ventral hernia. Most of these complications (52.9%) were managed nonoperatively. Any complication, however, was significantly associated with an increased need for revision within 6 months (47.1% vs 0%, P < .0001), shorter time to dialysis failure (4.9 months vs 11 months, P = .009), and worse overall 1-year survival rate (58.1% vs 87.1%, P = .002).

details the 105 patients at least 1 year removed from their original surgical procedure; in 27 of these patients (25.7%), the catheters were no longer in use at that time. These “failed” catheters were almost evenly split between operative (n = 15) and nonoperative (n = 12) causes, including peritoneal membrane failure, renal transplant, and patient choice. Operative complications were not significantly associated with catheter failures, nor were the need for adjunct procedures during the initial surgical procedure, prior major surgery, BMI, or age. Interestingly, male sex was significantly associated with catheter failures (P = .0252) via multivariate analysis. The development of postoperative adhesions was not included herein because these were only diagnosed as implicative in those patients in whom revision or removal surgery was required due to catheter failure. It stands to reason that adhesions nonetheless developed in many patients who did not require additional procedures.

Finally, catheter outcomes in patients with or without a history of major abdominal surgery were assessed (). Fifty-five patients had a history of major abdominal surgery before catheter placement compared with 61 without such a history. Notably, analysis showed that no significant demographic differences were present. As expected, previous major abdominal surgery was directly related to performance of selective adhesiolysis (P < .0001) and hernia repair (P = .026) and was inversely related to selective omentopexy (P < .0001). However, no significant differences were found regarding rate of revisions (25.5% vs 24.6%, P = .915) or complications (29.1% vs 29.5%, P = .961) when comparing groups with and without a history of major abdominal surgery. In addition, no significant difference was observed in either unassisted (P = .397) or overall (P = .647) catheter survival. When catheters did fail, the average length of time until catheter failure was also similar between those with and those without a history of abdominal surgery (7.7 ± 7.2 months vs 6.9 ± 5.3 months, P = .745). At a mean of 875 days after surgery, 75% of all catheters remain in functional use.

DISCUSSION

Although definitive proof of laparoscopy's advantage over open techniques for placing PD catheters has yet to be established via randomized, controlled trials,[10] many single-center observational and prospective, randomized reports support the safety, utility,[11,16-19] and cost-effectiveness[20] of minimally invasive surgery in this field. As laparoscopy becomes more popular, however, literature reinforcing its use in potentially problematic patient populations is lacking. Prior abdominal surgical procedures, in particular, have not been studied as to their risk of perioperative complications or catheter survival.

The potential for compartmentalization due to adhesions has put laparoscopy at the forefront of PD catheter placement for its ability to perform adhesiolysis as well as more precisely direct the catheter coil to its target area in the pelvis.[21] Coupled with selective hernia repair, these techniques used in our practice appeared to mitigate operative difficulties and, in turn, increased risks incurred by major prior abdominal surgery. It is also worth noting that selective omentopexy, performed in the manner described by Öğünç[22] in 1999 and subsequently reported by other authors,[11,14,15] was far less frequently performed in patients with prior abdominal surgical procedures, likely because of the action of postoperative adhesions in preventing the omentum's natural gravity-dependent extension into the lower abdomen.

Using advanced laparoscopic techniques common to most experienced minimally invasive surgeons, our cohort demonstrates the safety and feasibility of performing laparoscopic PD catheter placement in patients with complex operative fields and obese BMIs when compared with patients with virgin abdomens. This is a novel assertion to make, in that all but 1 reviewed study[21] failed to expand on outcomes related to surgical history.

Admittedly, ours is a single-center study describing a single surgeon's work, and this may not be completely reproducible. Nonetheless, our overall complication rate of 29.3% is within the reported complication rates of 3.5% to 37.4% for laparoscopic PD catheter insertion[11,14,17-19]—figures reported from studies with universally lower percentages of prior abdominal surgical procedures and unreported patient BMIs. Our study is also limited in the lack of differentiation of patients with prior surgical procedures into separate open and minimally invasive categories. This comparison was performed yet failed to provide significant data on account of being underpowered, although the increased incidence of postoperative hernias and adhesion-related complications after open versus laparoscopic abdominal surgery has been proved in larger analyses.[23] Exploring the implications of prior surgical approach on PD catheter outcomes is certainly warranted by future studies.

In conclusion, using advanced laparoscopic techniques in the placement of PD catheters has the potential to imbue better outcomes and expectations for surgeons and patients alike, regardless of demographic characteristics previously seen as barriers to a minimally invasive approach. This is particularly encouraging in light of the survival advantage for patients with ESRD in whom PD is used over hemodialysis as a primary modality for renal replacement therapy.[24,25] Adding PD's relative cost-effectiveness and improved quality of life, we believe that ongoing developments in patient education and surgical technique should further enable this underused dialysis method. Nevertheless, evidence reported and reviewed duly recognizes the risks inherent to intra-abdominal implantation of a prosthetic device in medically compromised patients. It is unlikely that perioperative complications in this patient demographic will ever be fully mitigated, no matter the techniques used. Thus our confidence in the safety and efficacy of laparoscopic PD catheter placement in patients with prior major abdominal surgical procedures is tempered by the need for prompt follow-up, multidisciplinary care, and early operative intervention as needed for catheter malfunction.

Table 1.

Descriptive Characteristics of Study Population (N = 116)

	Data
Age, y	52.4 ± 14.5
Male, %	56.0
Race, %
White	33.6
African American	59.5
Hispanic	2.6
Other	4.3
BMI,[a] kg/m2	30.4 ± 6.9
Previous major abdominal surgery, %	47.4
LOA,[a] %	31.0
Omentopexy, %	61.2
Hernia repair, %	6.9
Substernal exit site, %	32.8
Currently in use, %	74.3
Overall 1-y survival, %	72.2
Unassisted 1-y survival, %	60.2
Catheter failure within 1 y, %	25.7
Time to dialysis failure, mo	7.3 ± 6.2
Operative cause, %	13.9
Nonoperative cause, %	11.1
Revision, %	25.0
Revision within 6 mo, %	13.8
Complication, %	29.3
Adhesion, n	11
Infection, n	13
Catheter blockage, n	3
Malposition, n	4
Other, n	6

BMI = body mass index; LOA = lysis of adhesions.

Table 2.

Descriptive Characteristics in Relation to Catheter Failure at 1 Year

	Catheter Failure (n = 27)	Catheter Functional (n = 78)	P Value
			UVA[a]	MVA[ab]
Age, y	53.15	51.27	.7678
Male, %	88.9	48.7	.0049d	.0252
BMI,[a] kg/m2	29	30	.859
Previous major surgery, %	48.1	41.0	.5198
Adjunct surgery,[c] %	70.4	74.4	.6865
Peritonitis, %	7.4	5.1	.858
Catheter infection, %	11.1	5.1	.2941

BMI = body mass index; MVS, multivariate analysis; UVA, univariate analysis.

Unlisted variables were excluded for lack of significance.

Adjunct surgical procedures include lysis of adhesions, omentopexy, or hernia repair during original surgery.

Table 3.

Descriptive Characteristics in Relation to Abdominal Surgery

	No Abdominal Surgery (n = 61)	Abdominal Surgery (n = 55)	P Value
Age, y	50.5 ± 13.0	54.5 ± 15.9	.137
Male, %	60.7	50.9	.291
Race, % African American	60.7	58.2	.228
BMI,[a] kg/m2	29.9 ± 6.2	31.0 ± 7.5	.414
Revision, %	24.6	25.5	.915
Revision within 6 mo, %	13.1	14.5	.823
No. of revisions per patient	1.3 ± 0.6	1.3 ± 0.8	.861
Time to dialysis failure, mo	6.9 ± 5.3	7.7 ± 7.2	.745
Unassisted 1-y survival,[b] %	66.7	59.0	.397
Overall 1-y survival,[b] %	77.8	76.9	.647
LOA,[a] %	4.9	60.0	< .0001
Omentopexy, %	77.0	43.6	< .0001
Hernia repair, %	1.6	12.7	.026
Substernal exit site, %	29.5	36.4	.432
Complication, %	29.5	29.1	.961

BMI = body mass index; LOA = lysis of adhesions.

One-year survival calculations exclude patients presently <1 year removed from their original date of surgery (n = 105).