A Modified Open Surgery Technique for Peritoneal Dialysis Catheter Placement Decreases Catheter Malfunction

Abstract

♦ Background: This retrospective study was conducted to evaluate the effect of a new, modified open surgery technique on catheter-related malfunction.

♦ Methods: During the period from January 1997 to June 2009, 216 patients received initial peritoneal catheters. For the present study, patients were divided into four groups according to the catheter types and the surgery techniques:

• TO-S: traditional open surgery, straight Tenckhoff catheter

• TO-C: traditional open surgery, coiled Tenckhoff catheter

• TO-SN: traditional open surgery, swan-neck catheter

• MO-S: modified open surgery, straight Tenckhoff catheter

The modified surgery was characterized by a low incision site, a short intra-abdominal catheter segment and an additional upward straight subcutaneous tunnel.

All patients were followed up for 2 years or until death. Survival rates, complications caused by catheter placement, and the probability of malfunction-free catheter survival were compared between the groups.

♦ Results: Catheter malfunction was the most frequent mechanical complication, found in 31 patients (14.4%), who experienced 38 malfunctions. Only 2 episodes of catheter malfunction were found in the MO-S group, representing a rate significantly less than those in the TO-S and TO-C groups (both p < 0.05). Kaplan-Meier curves for malfunction-free PD catheter survival showed a significantly different malfunction-free probability for the various groups (p = 0.009). After 2 years of follow-up, 136 patients (63.0%) survived with their initial PD catheter. The initial catheter survival rate was 76.8% in the MO-S group. Kaplan-Meier curves for initial catheter survival showed that the highest survival rate was found in the MO-S group (p = 0.001).

♦ Conclusions: The modified open surgery technique is a reliable method for catheter placement.

Catheter malfunction, commonly seen in peritoneal dialysis (PD), is a major cause for PD technique failure (1-4). Aggressive interventions are usually needed to restore catheter function once catheter malfunction occurs (5-8). The need for intervention not only increases medical costs, but also increases the risk for complications accompanying the secondary procedures. Moreover, catheter malfunction may frustrate physicians and patients, lowering their confidence in PD as the first choice in the treatment of uremia. Finding an optimal catheter placement technique, with minimal mechanical malfunction, is therefore critical for a successful PD program.

Ever since the first permanent silicone catheter was introduced in the 1970s, a wide variety of catheters and placement techniques have been developed to attempt to eliminate catheter malfunction (5,9-12). Today, debate about the ideal catheter and the optimal technique for catheter placement is still ongoing (11,13-21). At present, the conventional Tenckhoff PD catheter, placed by open surgery, is still the approach most widely used around the world because of its simplicity, low cost, and lack of need for advanced instrumentation (4,22-26). We have used various catheters and surgical methods for catheter placement since a PD program was established in our department in 1997. Here, we report our retrospective study on catheter-related mechanical complications by open surgery, and we demonstrate a new modified open surgery technique that is ideal for straight Tenckhoff PD catheter placement.

Methods

Patients

From 1 January 1997 to 30 June 2009, 243 patients underwent initial PD catheter placement at our center. We excluded 27 patients from the present analysis because their catheters were placed by laparoscopic surgery. Analysis of the remaining 216 patients included only their initial PD catheters. The study was approved by the Ethics Review Board of Drum Tower Hospital.

The 3 catheter types used in our center (straight Tenckhoff, coiled Tenckhoff, and swan-neck catheter) were all Quinton-type catheters (Covidien, Mansfield, MA, USA). The length between the two cuffs is 6 cm for straight Tenckhoff catheters, with a 15 cm intra-abdominal segment. Traditional and modified open surgery techniques were applied for catheter placement. Patients were categorized according to catheter type and surgical technique:

• TO-S group: traditional open surgery, straight Tenckhoff catheter (n = 53)

• TO-C group: traditional open surgery, coiled Tenckhoff catheter (n = 41)

• TO-SN group: traditional open surgery, swan-neck catheter (n = 66)

• MO-S group: modified open surgery, straight Tenckhoff catheter (n = 56)

Data Collection

We collected patient information by manually reviewing hospital records. Patient demographics, catheter and placement technique, complications, and catheter and patient outcomes were recorded. All patients were followed for 2 years after catheter placement or until loss of the initial functioning catheter or death, whichever came first.

Mechanical complications caused directly by the catheter type or placement technique were of particular interest. Among such complications, catheter malfunction was defined as a persistent inability to instill dialysate or to reliably drain effluent within 30 minutes. Catheter malfunction was classified into two types: catheter migration, defined as dislocation of catheter tip out of the true pelvis on an abdominal radiograph; and non-migration problems, mainly refractory obstruction (requiring intraluminal intervention or further surgery). Other complications observed in the present study included extrusion of the outer cuff, inflow or outflow pain, hernia, leakage, and prominent bleeding, including persistent bleeding at the incision site or exit site and persistent hemoperitoneum with blood clot in the effluent that required blood transfusion or surgical intervention.

Catheter Placement

All surgeries were performed by the same group of nephrologists and urologists in the operating room with the patient under local anesthesia. Prophylactic antibiotics, usually cephalosporin, were administered before the procedure. The details of the traditional and modified open surgical procedures are described next.

Traditional Open Surgery: Traditional open surgery procedure has been described elsewhere (27). In brief, the incision site in our patients was generally in the paramedial area, 11 cm above the pubic symphysis and 2 - 2.5 cm lateral to the midline. A 4 - 5 cm sagittal incision was made in the skin and carried down to the anterior rectus fascia. After the rectus muscle had been bluntly separated, the posterior rectus sheath was incised to expose the peritoneum. A purse-string suture with a diameter of 0.5 cm was then placed in the peritoneum. The catheter was inserted through the opening and introduced into the deep pelvis with the help of a blunt stiffening metal stylet. After removal of the stylet, the purse string was closed just below the inner cuff. Catheter function was tested with in-and-out instillation of 1000 mL dialysate to observe whether the dialysate was easily aspirated. The anterior rectus sheath was then closed with interrupted sutures, with the inner cuff staying in the abdominal muscular layer and the catheter going through the rectus sheath at the uppermost point of the incision. A subcutaneous tunnel with a downward exit site was made using a tunneler. The exit site was chosen to avoid the trouser belt area and skinfolds, and the outer cuff was placed approximately 2 cm from the exit site. The incision was then closed as usual.

Modified Open Surgery for Straight Tenckhoff Catheter: To improve initial catheter function, we developed a new open surgery method for implanting a straight Tenckhoff catheter. The major differences between the new technique and traditional open surgery are that

• the main incision site is set in the paramedial area 7 cm above the pubic symphysis and 2 - 2.5 cm lateral to the midline.

• the length of the intra-abdominal portion of the catheter is set during operation based on a real-time measurement of the distance from the opening in the peritoneum to the pouch of Douglas. Briefly, after a purse string suture is placed in the peritoneum, the catheter, with metal stylet, is inserted into the pouch of Douglas. The distance between the peritoneal opening and the inner cuff is measured using a sterile ruler. The catheter is then withdrawn from the abdominal cavity and the intra-abdominal segment is trimmed and shortened to a length that is about 1 cm longer than the measured distance between the peritoneal opening and the inner cuff. With the aid of the stylet, the trimmed catheter is then placed back into the pouch of Douglas.

• an additional 1 cm transverse skin incision, which serves as a “jump site,” is made at a point 6 cm just above the upper margin of the main incision. The catheter is then pulled through the “jump site” to create a long, straight, upward subcutaneous tunnel, with the outer cuff staying in the tunnel.

• a curved subcutaneous tunnel is made for the exit site by using a tunneler in a lateral downward direction, resulting in a 5 - 6 cm uncuffed catheter segment within the subcutaneous tunnel.

PD Regime after Surgery

The patient’s PD program was usually commenced on the second day after catheter placement, with the patient in a supine position during the first week after surgery. Dialysate volume during the first week was 1 L. Exchanges of PD fluid were performed once every 2 - 3 hours. Thereafter, a regular schedule was used for each patient (2 L and 3 - 4 exchanges daily).

Statistical Analysis

Normally distributed continuous variables are expressed as mean ± standard error, and qualitative data as absolute numbers and percentages. One-way analysis of variance was applied to compare the means of normally distributed continuous variables. For qualitative data, a Mann-Whitney test or Kruskal-Wallis test for abnormally distributed continuous variables and a chi-square test or a Fisher exact test was used, as appropriate.

The Kaplan-Meier method, based on the log-rank test, was used for the descriptive analysis of initial catheter survival probability and malfunction-free probability. Catheter survival was defined as a functional initial catheter before its loss for any catheter-related problem. Malfunction-free time was defined as the duration from catheter placement to the first observed catheter-related malfunction. Catheter removal for reasons unrelated to failure of peritoneal access (resolution of renal failure, conversion to hemodialysis, renal transplantation) and unrelated death was censored.

A value of p < 0.05 was considered statistically significant, and all tests were two-tailed. All calculations were performed with the SPSS software application (version 17.0 for Windows: SPSS Institute, Chicago, IL, USA).

Results

Patient Characteristics

Table 1 shows the baseline demographic characteristics of the four patient groups. The study included 134 men and 82 women, with an average age of 58.3 ± 15.3 years. No significant difference were found in patient age between the groups. Of the 216 patients, 91 had primary glomerulonephritis, 51 had diabetic nephropathy, 49 had hypertensive arteriosclerosis, and 25 had other diseases.

TABLE 1.

Demographic Data and Clinical Outcomes After Two Years of Follow-Up

After 2 years of follow-up, 36 patients (16.7%) had died, with no significant difference in the patient death rate between the groups. There were 132 episodes of peritonitis, for an overall rate of 0.033 episodes per patient-month. Episodes of exit-site infection numbered 47, for a rate of 0.012 episodes per patient-month. The difference in episodes of exit-site infection between groups was significant (p = 0.039).

Comparison of Surgical Techniques

Traditional open surgery was performed in 160 patients, and modified open surgery was performed in the remaining 56 patients. Figure 1(A,B) shows a radiograph and schematic representation of the traditional open surgery; Figure 1(C,D) shows the modified open surgery. The distance from the peritoneal opening to the bottom of the pouch of Douglas varied from 10.0 cm to 14.0 cm in the MO-S group. The mean intra-abdominal catheter segment for insertion in the MO-S group was 11.2 ± 0.82 cm.

Figure 1 —

Comparison of the traditional open surgery and modified open surgery techniques. (A) Schematic representation and (B) radiograph of a straight Tenckhoff catheter placement by traditional open surgery. a = Distance between the pubic symphysis and the main incision in the traditional technique (11 cm). b = Intra-abdominal segment in the traditional technique (15 cm). (C) Schematic representation and (D) radiograph of a straight Tenckhoff catheter placement by modified open surgery. c = Distance between the pubic symphysis and the main incision in the modified technique (7 cm). d = Intra-abdominal segment in the modified technique (9 - 13 cm).

Initial Catheter Survival

After 2 years of follow-up, 136 patients (63.0%) survived with their initial PD catheter. Initial catheter survival was significantly different between the groups (p = 0.008), with the highest survival rate (76.8%) observed in the MO-S group. Figure 2 shows Kaplan-Meier curves for initial catheter survival. The overall difference in initial catheter survival was statistically significant (p = 0.001).

Figure 2 —

Initial catheter survival rate in the MO-S (modified open surgery, straight Tenckhoff catheter), TO-SN (traditional open surgery, swan-neck catheter), TO-C (traditional open surgery, coiled Tenckhoff catheter), and TO-S (traditional open surgery, straight Tenckhoff catheter) groups (log-rank p = 0.001).

Mechanical Complications

During 2 years of follow-up, 87 episodes of mechanical complications occurred with the initial catheters (Table 2). Catheter malfunction was the most frequent mechanical complication, seen in 31 patients (14.4%), who experienced 38 episodes. There were also 18 episodes of leakage, 11 episodes of inflow or outflow pain, 8 episodes of hernia, 7 episodes of outer cuff extrusion, and 5 episodes of prominent bleeding. Only 1 patient in the MO-S group experienced prominent bleeding, with the primary disease being systemic lupus erythematosus. Persistent hemorrhage at the incision and blood clot in the effluent were observed after surgery. The patient’s platelet level dropped as low as 9×10⁹/L, and bleeding was stopped by platelet transfusion. Other causes for bleeding included inappropriate surgical ligation and suture in subcutis. In those instances, bleeding was stopped with additional sutures and compression bandaging.

TABLE 2.

Mechanical Complications by Patient Group

Two patients in the MO-S group experienced catheter malfunction, for a rate that was significantly lower than rates in the other groups (TO-S, 20.8%; TO-C, 24.4%; TO-SN, 12.1%; MO-S, 3.57%; overall p = 0.014). One experienced catheter tip migration, and the other experienced catheter obstruction because of a fungal infection. Figure 3 shows Kaplan-Meier curves for PD catheter malfunction-free survival. The overall difference in malfunction-free survival was statistically significant (p = 0.009).

Figure 3 —

Initial catheter malfunction-free rate in the MO-S (modified open surgery, straight Tenckhoff catheter), TO-SN (traditional open surgery, swan-neck catheter), TO-C (traditional open surgery, coiled Tenckhoff catheter), and TO-S (traditional open surgery, straight Tenckhoff catheter) groups (log-rank p = 0.009).

Review of the Literature

We also conducted a comprehensive literature review of studies of catheter malfunction using various modified open surgery techniques. Table 3 shows representative studies of double-cuffed straight catheter placement. The rate of catheter malfunction in the modified open surgery group in our study is comparable to rates reported in earlier studies (19,28-33).

TABLE 3.

Previous and Present Studies Using Modified Open Surgery for Straight Catheter Placement

Discussion

The rate of catheter malfunction with open surgery can be as high as 19.4% (3,15,34,35). Repeated operations to obtain a well-functioning PD catheter are obviously unfavorable to both physicians and patients. In the present retrospective study, we demonstrated a new modified open surgery technique for straight Tenckhoff catheter placement that effectively reduced catheter malfunction during 2 years of follow-up.

Catheter malfunction is troublesome, and recurrence of this complication inevitably compromises the efforts to increase PD penetration among uremic patients. Various catheters were designed in the early years after the invention of the first permanent silicone catheter, aiming to lower the rate of catheter malfunction. However, none of those catheters has been fully proved to be superior to others (9-11,20,23). In addition to new catheter designs, great efforts have also been made to improve catheter placement techniques (36-39). Although some new techniques were reported with promising results, there appear to be several barriers to wide uptake of those techniques, including the requirement for additional instrumentation, high cost, and a need for more specialized and experienced operators. Hence, up to now, open surgery is still the most widely used method for catheter placement all over the world (4,22,24,25).

In the present study, using a modified open surgical technique, we found that only 2 of 56 patients experienced catheter malfunction during 2 years of follow-up. The associated rate of malfunction was significantly lower for the MO-S group than those for the TO-S and TO-C groups. We speculate that several key characteristics of our procedure may contribute to the low rate of catheter malfunction.

First, the length of the intra-abdominal catheter segment used in the MO-S group (11.2 ± 0.82 cm) was significantly shorter than is common in the traditional technique. Some recent studies have showed that a low incidence of catheter malfunction can be achieved by creating a short, mobilizable intra-abdominal catheter (30,40-44). It is reasonable that the short length of the intra-abdominal segment may increase the non-deformability of the catheter tip and thus antagonize tip migration.

Second, because omental entrapment is a common cause for refractory catheter malfunction (6,45,46), we placed the catheter in a lower site, allowing the catheter to locate deep enough in the pelvis. As shown in Figure 1, the inner cuff was positioned at the level of the brim of true pelvis, which should largely reduce the chance of interaction between the omentum and the intra-abdominal segment.

Third, the length of intra-abdominal segment was adjusted for each individual patient in our study. This adjustment might effectively avoid catheter bending, which is known to increase the risk of the catheter tip rebounding out of the pouch of Douglas because of its high intrinsic tensile force (47). Although the pubic symphysis is now widely used as a body marker for catheter placement, the distance from the predetermined insertion point to the pouch of Douglas can still vary from patient to patient. For example, in the present study, the distance from the peritoneal opening to the bottom of the pouch of Douglas ranged from 10.0 cm to 14.0 cm in the MO-S group. That variation suggests that it is almost impossible to use a catheter of a single fixed length to fit all patients. In our procedure, we further shortened the catheter by 1 cm based on the measured distance. This practice proved to be feasible, and no catheter bending was observed in the MO-S group.

Finally, we created a long, straight, upward subcutaneous tunnel, which is helpful for preventing catheter migration, because the torque induced by the arcuate tunnel can be dissipated by the long, straight, upward subcutaneous tunnel.

Because the imperfect traditional open surgery is widely used for catheter placement, it is important to modify the current surgical procedure to further improve initial catheter function. As a proof of concept, the present study has shown (as have others) a promising decrease in catheter malfunction with modification of the open surgery procedure (19,29,30,33). Although the use of a low incision site has previously been reported, the introduction in this study of an individualized length of catheter, coupled with a long, straight, upward tunnel, may bring more potential benefits to catheter placement, as has already been discussed. Moreover, the long, straight, upward tunnel effectively avoids an extremely low exit site, commonly seen in other low-incision techniques, which may be inconvenient for incision care (29,33). An over-sharp angle between the two limbs of the straight catheter may increase the torque and should be avoided in the traditional technique. However, an angle is feasible in the modified technique, because the torque can largely be dissipated by the long, straight, upward tunnel.

Although several modifications are introduced in our modified technique, none are overcomplicated, and all can be completed in several minutes, without prolonging the overall operation time significantly. Our study also proved that the modified technique was feasible in clinical use, having no observable side effects during a 2-year follow-up period. The rates for complications such as prominent bleeding, inflow or outflow pain, leakage, and hernia in the MO-S group were comparable with those in the TO-S, TO-C, and TO-SN groups (p > 0.05). In the present study, the incidence of exit-site infection in the MO-S group was comparable to that in the TO-SN group, suggesting that a long uncuffed catheter does not increase the chance of infection. No bladder injury or intestinal perforation was observed with the modified catheter placement technique. And no patient in the MO-S group has complained of any discomfort because of the short intra-abdominal segment. No skin erosion was observed in the MO-S group even for slender patients. Drainage of effluent was always smooth and occurred within less than 10 minutes.

We acknowledge that the present study has some limitations because of its retrospective nature. First, it is well recognized that the experience of the operator placing the PD catheter is crucial for minimizing catheter complications (11,22,48-50). Although the same group of operators was involved in all surgeries in the present study, improved experience of those operators over time could more or less affect the result (51). Also, lower rates of exit-site infection were observed in the TO-SN and MO-S groups than in the TO-C and TO-S groups. Whether that improvement is directly a result of the catheter type or the implantation technique is still unclear, because mupirocin has routinely been prescribed to the patients in the TO-SN and MO-S groups to prevent exit-site infection. Despite the fact that none of our patients left the PD program because of dialysate leakage, we observed a high rate of leakage in the study. That observation may mainly be a result of the short interval between surgery and dialysis for our patients. However, a small incision in the peritoneum and tight suturing are absolutely necessary to avoid leakage, because more favorable data have been reported in patients who began dialysis without a break-in period (31). Furthermore, Stegmayr reported that the three-purse-string suture technique can permit early initiation of PD with a low rate of leakage (28). Another caveat of the present study is the relatively small number of samples. Although we observed an obvious trend of improved catheter outcomes in the MO-S group compared with the TO-SN group, that improvement was not statistically significant. This particular finding might be improved in further studies with larger sample numbers.

Conclusions

We demonstrated a modified open surgery catheter placement technique with a high initial success rate. The multiple modifications introduced in this new technique are directed at eliminating known risk factors for catheter malfunction and might help to reduce the occurrence of catheter malfunction complications.

Disclosures

The authors have no financial conflicts of interest to declare.