A PROSPECTIVE RANDOMIZED STUDY ON THREE DIFFERENT PERITONEAL DIALYSIS CATHETERS

Proceedings of the First Asian Chapter Meeting ISPD December 13 15, 2002, Hong Kong Peritoneal Dialysis International, Vol. 23 (2003), Supplement 2 0896-8608/03 $3.00 +.00 Copyright 2003 International Society for Peritoneal Dialysis Printed in Canada. All rights reserved. A PROSPECTIVE RANDOMIZED STUDY ON THREE DIFFERENT PERITONEAL DIALYSIS CATHETERS Wai-Kei Lo, Sing-Leung Lui, Fu-Keung Li, Bo-Ying Choy, Man-Fai Lam, Kai-Chung Tse, Terence P.S. Yip, Flora S.K. Ng, Suk-Ching Lam, Wai-Ling Chu, and Suk-Wai Cheng Renal Unit, Department of Medicine, Tung Wah Hospital, The University of Hong Kong, Hong Kong SAR, China Objective: We compared outcomes for catheters with different configurations: conventional straight, swan-neck straight tip, and swan-neck curled tip. Design: The study was conducted as a prospective randomized controlled trial in the continuous ambulatory peritoneal dialysis (CAPD) unit of a university center. Patients and Methods: We randomized 93 new regular CAPD patients without prior peritoneal dialysis (PD) catheter insertion to receive a conventional straight, doublecuffed catheter (CS), a swan-neck straight catheter (SNC), or a swan-neck curled tip catheter (SNC) in 2:1:1 ratio. Results: The exit-site infection (ESI) rate was slightly lower with swan-neck catheters as compared with straight catheters, but the difference was not statistically significant. The peritonitis rate and overall catheter survival were similar. In Staphylococcus aureus nasal non carriers as compared with carriers, ESI-free catheter survival was significantly better with swan-neck catheters (p = 0.0302 and p = 0.82 respectively). As compared with SC catheters, SNC catheters had a significantly higher migration rate (p = 0.022). Conclusions: Swan-neck catheters were associated with a slightly better ESI rate, but SNC catheters are not routinely recommended because of a high migration rate. The SNS catheter is therefore recommended as the firstline catheter of choice, particularly in populations with a low rate of S. aureus nasal carriage. Perit Dial Int 2003; 23(S2):S127 S131 www.pdiconnect.com KEY WORDS: Catheter; exit-site infection. The peritoneal dialysis (PD) catheter is the lifeline of PD patients. Complications relating to PD catheters are common and may cause serious problems. Among all complications, exit-site infection (ESI) is the most common. An upward-directed exit predisposes to ESI because dirt cannot be drained out natu- Correspondence to: W.K. Lo, Department of Medicine, Tung Wah Hospital, 12 Po Yan Street, Hong Kong SAR, China. wkloc@hkucc.hku.hk rally. When the catheter is implanted with an arcuate tunnel and a downward exit, the bad effects of an upward tunnel are theoretically minimized, but the natural resilient force of the conventional straight Tenckhoff catheter may result in protrusion of the external cuff and therefore again predispose to ESI. The swan-neck catheter with a permanent bend at the intramural portion and a downward exit was developed to minimize the risk of ESI and protrusion of the external cuff. Improvement in the ESI rate was reported (1), but randomized controlled studies have not confirmed that improvement (2). One major disadvantage of the swan-neck catheter is that, if tip migration occurs, the catheter cannot be repositioned by guide-wire manipulation, and therefore surgical intervention is often required. Catheter tip migration was claimed to be less frequent with a curled tip configuration (3). Combining the curled tip configuration with the swan neck may minimize the disadvantage of the swan-neck configuration. In randomized studies, swan-neck curled catheters have also been reported to have less tip migration than straight curled-tip catheters (4) and less ESI than straight conventional catheters (5). However, despite the theoretical advantages and disadvantages and the observations, no single catheter has been shown to be clearly superior over the conventional Tenckhoff catheter (6). In the present prospective randomized study, we examined outcomes with three commonly used PD catheters to answer several questions: Can a downward exit really reduce ESI? Do curled tip catheters offer additional benefits in regard to catheter outcome? Which catheter is associated with the best outcome? PATIENTS AND METHODS From January 1998 to January 2001, we recruited into the study new regular PD patients without a S127

PROCEEDINGS OF THE FIRST ASIAN CHAPTER MEETING ISPD DECEMBER 2003 VOL. 23, SUPPL 2 PDI history of prior PD catheter insertion. After obtaining informed consent, we randomized the patients to receive one of the three different catheters in a 2:1:1 ratio: a conventional straight double-cuffed Tenckhoff catheter (CS), a swan-neck straight catheter (SNS), or a swan-neck curled tip catheter (SNC). The randomization was designed to achieve a 1:1 CS to swan-neck (SN) ratio and also 1:1 SNS to SNC ratio. All catheter implantations were performed by the same group of four trained nephrologists using minilaparotomy on the lower paramedian abdomen under local anesthesia. The tunnel was directed obliquely upward, with a sideward exit for CS catheters and a downward exit for SN catheters. Position of catheter tip was checked by abdominal x-ray immediately after implantation, and in all cases the catheter tip was located within the true pelvic cavity according to the x-ray image. Cefazolin 1 g was given intravenously as a prophylactic antibiotic just before the operation. A nasal swab was taken before the implantation surgery and at time of continuous ambulatory peritoneal dialysis (CAPD) training, which usually occurred about 6 weeks after implantation. Mupirocin application to the nostrils was prescribed if methicillin-sensitive or methicillinresistant Staphylococcus aureus (MSSA or MRSA) were cultured. The nasal swab or mupirocin nasal application was not routinely repeated thereafter. Twice-weekly intermittent peritoneal dialysis (IPD) was started immediately after implantation in almost all cases. Training for CAPD was conducted at about 6 weeks after catheter implantation. Until maturation of the exit, renal nurses using normal saline performed exit-site care at the intervals recommended by Prowant and Twardowski (7). Thereafter, patients or their helpers used povidone iodine as the standard antiseptic solution for daily exit-site care (but other solutions, including aqueous chlorhexidine and saline, were also used). Prophylactic mupirocin was not applied to the exit site. Renal nurses or nephrologists regularly inspected the catheters (every 6 weeks). We defined ESI according to the classification by Twardowski and Prowant (8). For the purposes of the present study, acute and chronic ESI and cuff infections were all grouped under ESI. Equivocal exit sites were regarded as ESI when pathogenic organisms other than skin commensals were cultured from the exit-site swab. All ESIs were considered healed if signs and symptoms of ESI had disappeared for more than 6 weeks. Duration of an episode of ESI refers to the period from onset of ESI until the exit site had healed. The duration proportion of ESI refers to the total duration of ESI divided by the total duration of follow-up for that catheter. The study endpoint was the removal of the catheter or 31 January 2002 (1 year after the last patient recruitment). Based on power analysis to show a clinical significance of reducing ESI episodes by one third in the SN group, the original study was designed with a sample size of 60 patients. Because of a failure to show any significant difference in outcome by the time 60 patients had been recruited, the study was extended to recruit 50% more patients. The Student t-test and the Mann Whitney U-test were used for parametric and nonparametric data comparisons between groups. The chi-square test was used for nominal data comparison. Catheter survival and ESI-free catheter survival were compared using life-table analysis and the Kaplan Meyer test. RESULTS From August 1997 to January 2001, we recruited 93 patients, of whom 48 were randomized to CS catheters and 45 to SN catheters (23 SNS catheters and 22 SNC catheters). No statistical difference was observed between the demographics of the CS and SN groups (Table 1). No early leakage occurred postoperatively or during the IPD phase. Over a mean period of 27.0 ± 13.7 months of follow-up, ESI occurred in 34 patients (71%) in the CS group and in 28 patients (62%) in the SN group (p = NS). No tunnel infections or cuff protrusions occurred. The number of ESI episodes and the ESI duration proportion were slightly lower in the SN group, but the difference was not statistically significant. The demographic data and ESI outcomes in the SNS and SNC catheter groups were essentially identical (Table 2). Table 3 shows the causative organisms for ESI in the CS and SN catheters. The number of catheters removed did not differ between the groups; but, in the CS group, 5 catheters were removed for ESI, while in the SN group, no catheter was removed for that reason (p = 0.026, Table 4). Of the catheters removed because of ESI, 4 cases involved MRSA infection. (In 2 of those cases, culture of the patient s initial nasal swab had been positive for MSSA, and in 1 case, nasal MRSA had been treated with mupirocin). In the 5th case, the catheter was removed for Pseudomonas aeruginosa infection. Tip migration occurred more often in the SNC group [CS:SNS:SNC = 6:2:7 (12.5%:8.7%:31.2%), p = 0.069 overall, p = 0.053 for SNS vs SNC, and p = 0.022 for CS vs SNC], but only 3 catheters were removed for associated ultrafiltration problems, all in the SN group (Table 4). One CS catheter was removed for delayed leakage occurring 26 months after implantation. S128

PDI DECEMBER 2003 VOL. 23, SUPPL 2 PROCEEDINGS OF THE FIRST ASIAN CHAPTER MEETING ISPD TABLE 1 Demographic Data and Exit-Site Infection (ESI) Results for the Conventional Straight and Swan-Neck Catheter Patient Groups Conventional catheter Swan-neck catheter p Value Patients (n) 48 45 Patients with diabetes (n) 27 18 0.117 Sex (male:female) 24:24 21:24 0.748 Age (years) 60.8±13.6 62.6±42.6 0.513 S. aureus nasal carriage (n) 22 25 0.349 Catheter months 25.2±12.7 28.9±14.6 0.184 Patients with ESI [n (%)] 34 (71%) 28 (62%) 0.379 ESI episodes/catheter 2.15±2.04 1.64±1.85 0.184 ESI episodes/catheter year 1.07±0.96 0.77±0.81 0.138 ESI duration proportion (%) 14.7±18.8 9.3±11.8 0.217 Patients with peritonitis (n) 25 24 0.904 Peritonitis episodes/catheter 1.12 1.11 0.896 Peritonitis episodes/catheter year 0.62 0.88 0.997 TABLE 2 Demographic Data and Exit-Site Infection (ESI) Results for Swan-Neck Straight and Swan-Neck Curled Catheters Swan-neck straight Swan-neck curled p Value Patients (n) 23 22 Patients with diabetes (n) 7 11 0.181 Sex (male:female) 10:13 11:11 0.661 S. aureus nasal carriage (n) 13 12 0.894 Catheter months 29.8±13.1 28.1±16.4 0.709 Patients with ESI (n) 14 14 0.848 ESI episodes/catheter 1.44±1.80 1.86±1.91 0.459 ESI episodes/catheter year 0.69±0.68 0.86±0.94 0.710 ESI duration proportion (%) 8.9±11.7 9.8±12.1 0.816 Patients with peritonitis (n) 12 12 0.873 Peritonitis episodes/catheter 1.17±1.59 1.05±1.25 0.990 Peritonitis episodes/catheter year 0.55 1.22 0.737 Catheter survival and ESI-free survival were not statistically significantly different between the three groups (Figures 1 and 2), even when CS catheters were compared with SN catheters (Figure 3). The SNS catheters had slightly better catheter survival as compared with CS catheters, but that difference was not statistically significant (p = 0.0915). Nasal S. aureus carriers had a slightly shorter ESI-free catheter survival (p = 0.14), but the ESI-free survival benefit was seen mainly in non carriers (p = 0.0302, Figure 4) rather than in carriers (p = 0.82). Nasal carriage of S. aureus did not affect the overall catheter survival. DISCUSSION Climate may affect the incidence of ESI and, possibly, outcome comparisons of various types of PD catheters. A catheter having good outcome in one locale will not necessarily do as well in other locales with different climates. Studying catheter outcomes in various localities is essential. Our study, conducted in the hot and humid summers and dry and cool winters of Hong Kong, showed that the three different types of PD catheters commonly used do not have markedly different outcomes. However, we noticed a trend toward lower risk of ESI in SN catheters, and significantly fewer SN catheters were removed because of ESI. With a larger sample size, the difference in infection rate may reach statistical significance. The main benefit of SN catheters was found in nasal non carriers of S. aureus. If that finding applies in other localities, the inconsistently superior result of SN catheters may be found to be partly attributable to variations in nasal S. aureus carriage rate. The loss of benefit from SN catheters among nasal S. aureus carriers suggests that other means of preventing ESI in S. aureus nasal carriers are needed. S129

PROCEEDINGS OF THE FIRST ASIAN CHAPTER MEETING ISPD DECEMBER 2003 VOL. 23, SUPPL 2 PDI TABLE 3 Causative Organisms in the Episodes of Exit-Site Infection (ESI) for the Two Major Catheter Types Conventional straight catheters Swan-neck catheters MSSA 32 26 MRSA 28 9 CNS 6 3 S. viridans 1 0 Diphtheroids 2 3 C. jeikeium 10 7 E. coli 1 4 P. aeruginosa 17 14 X. maltophilia 1 0 Serratia species 5 1 A. baumannii 1 0 Fungi 1 3 MSSA = methicillin-sensitive Staphylococcus aureus; MRSA = methicillin resistant S. aureus; CNS = coagulasenegative Staphylococcus. The ESI rates for the two types of swan-neck catheter were essentially identical in our study. Compared with another randomized study on SNC and CS catheters from Singapore by Lye et al. (5), our ESI rate was higher and our patients did not achieve as marked a benefit from the swan-neck configuration. Those findings may be related to a difference in S. aureus nasal carriage rate (data not reported by Lye et al.) and possibly also to a difference in the definition of ESI, because the Singapore study was performed before the ESI classification by Twardowski and Prowant was published (8). In our study, we routinely took exit-site swabs of equivocal exits and regarded the exits as infected if pathogenic organisms were cultured from the swabs. The cases so classified might not have been regarded as ESI in other studies. Figure 1 Catheter survival for the three catheter types. Solid line = conventional straight catheters (SC); dotted line = swan-neck straight catheters (SNS); dashed line = swan-neck curled-tip catheters (SNC). p = 0.273 overall; p = 0.0915 between CS and SNS catheters; p = 0.27 between SNS and SNC catheters. In contrast to results found in other reports, we experienced a very high tip migration rate with SNC catheters. According to Gadallah et al. (4), the migration rate of SNC catheters is as low <1%. In our study, the rate was 30% (only 11% of straight catheters migrated). The reason for our result is unclear. The migrations were unlikely to be attributable to a free-floating catheter tip; rather, they probably occurred as a result of omental activity. The fixed bend of the SN configuration makes SNC catheters impossible to reposition by fluoroscopic guide-wire manipulation. Catheter replacement or laparoscopic repositioning would be required in cases in which urine output cannot compensate for the reduction in ultrafiltration attributable to migration. TABLE 4 Final Outcome of Catheters Conventional Swan-neck straight Swan-neck curled p Value Catheters (n) 48 23 22 Intact at end of study (n) 35 20 16 Death 7 5 4 Transplant 4 2 1 Removed 13 3 6 0.386 For peritonitis 7 2 4 0.647 For ESI 5 0 0 0.084 a For migration 0 1 2 0.128 b For leakage 1 0 0 0.330 a p = 0.026 for conventional straight versus swan-neck catheters. b p = 0.069 for conventional straight versus swan-neck catheters. S130

PDI DECEMBER 2003 VOL. 23, SUPPL 2 PROCEEDINGS OF THE FIRST ASIAN CHAPTER MEETING ISPD Figure 2 Exit-site infection (ESI) free survival for the three catheter types. Solid line = conventional straight catheters; dotted line = swan-neck straight catheters; dashed line = swan-neck curled-tip catheters. p = 0.306 overall. Figure 4 Exit-site infection (ESI) free survival in nasal S. aureus non carriers. Swan-neck catheters (dotted line) had significantly better results as compared with conventional straight catheters (solid line), p = 0.032. Although catheter survival was similar for the three catheter types, the SNS catheter should be the catheter of choice in our locality, because swan-neck catheters have a slightly lower risk of ESI, particularly in nasal S. aureus non carriers. The expensive SNC catheter offers no additional outcome advantage and has a higher migration rate. It should therefore not be considered the first-line choice. REFERENCES Figure 3 Exit-site infection (ESI) free survival for swanneck catheters (dotted line) and conventional straight catheters (solid line). Survival was slightly better for swanneck catheters, but the difference did not reach statistical significance (p = 0.124). The prices of the three catheters showed great discrepancy. In Hong Kong, the catheter cost was about US$50 for CS, US$100 for SNS, and US$150 for SNC. CONCLUSION 1. Twardowski ZJ, Prowant BF, Nichols WK, Nolph KD, Khanna R. Six-year experience with swan neck catheters. Perit Dial Int 1992; 12:384 9. 2. Eklund BH, Honkanen EO, Kala AR, Kyllonen LE. Peritoneal dialysis access: prospective randomized comparison of the swan neck and Tenckhoff catheters. Perit Dial Int 1995; 15:353 6. 3. Nielsen PK, Hemmingsen C, Friis SU, Ladefoged J, Olgaard K. Comparison of straight and curled Tenckhoff peritoneal dialysis catheters implanted by percutaneous technique: a prospective randomized study. Perit Dial Int 1995; 15:18 21. 4. Gadallah MF, Mignone J, Torres C, Ramdeen G, Pervez A. The role of peritoneal dialysis catheter configuration in preventing catheter tip migration. Adv Perit Dial 2000; 16:47 50. 5. Lye WC, Kour NW, van der Straaten JV, Leong SO, Lee EJ. A prospective, randomized comparison of the swan neck, coiled, and straight Tenckhoff catheters in patients on CAPD. Perit Dial Int 1996; 16(Suppl 1): S333 5. 6. Piraino B. Which catheter is the best buy? Perit Dial Int 1995; 15:353 6. 7. Prowant BF, Twardowski ZJ. Recommendation for exit care. Perit Dial Int 1996; 16(Suppl 3):S94 9. 8. Twardowski ZJ, Prowant BF. Classification of normal and diseased exit sites. Perit Dial Int 1996; 16(Suppl 3):S32 50. S131