Peritoneal catheter survival: The impact of unroofing

Methods: From our database, we prospectively evaluated the outcome of 139 peritoneal catheters placed in 121 patients (1.14 catheters per patient, 73 female, 48 male) from 08-03-1993 to 12-312016. Twenty-three infected catheters needed surgical unroofi ng of the tunnel tract. We calculated the cumulative catheter survival rates (Kaplan-Meier) of 1) overall catheters and 2) not unroofed catheters, considering catheter removal as an endpoint. We also calculated 3) the unroofed catheter survival, considering the date of unroofi ng or catheter removal as the endpoint and, 4) continuity of the unroofed catheters post-unroofi ng, considering the unroofi ng date as if it were a new catheter and catheter removal as the endpoint. Likewise, we compared the survival of: a) unroofed catheters vs. continuity of the unroofed catheters and, b) no unroofed catheters vs. continuity of the unroofed catheters (Log-rank test) (signifi cance value P< 0.05).


Introduction
Infection of the peritoneal catheter during peritoneal dialysis can facilitate entry of microorganisms into the peritoneal cavity, leading to peritonitis. Failure to cure these infections can lead to the removal of the peritoneal catheter and, thus, technique failure or other more severe scenarios. Likewise, frequent infections of the exit site, often chronic, generate the continuous use of antibiotics, facilitating the appearance of Gram-negative exit site infections and fungi peritonitis [1]. In the latest recommendations for International Society for Peritoneal Dialysis (ISPD) infectious complications for the catheter exit site, there is no mention of how to avoid the removal of peritoneal catheters in cases of chronic infections [2][3][4]. This procedure, known as unroofi ng of the peritoneal catheter, is nowadays rarely recommended.
Some articles have described the technical characteristics of the procedure and their outcomes in adult and paediatric patients, as well as other subcutaneous surgical techniques to avoid catheter removal, but no articles have investigated the long-term durability of unroofed catheters [5][6][7][8][9]. In this work, we share our experience of many years of follow-up in the evolution of our peritoneal catheters and the impact of the unroofi ng on catheter survival.

Materials and Methods
From our database, we prospectively evaluated the outcomes of 139 peritoneal catheters (34 Tenckhöff straight, 104 Swan Neck and 1 Missouri) placed in 121 adult patients (1.14 catheters per patient); 73 female, 48 male; mean age 52.97 ± 15.32-yearsold; time at risk 6672 ± 48 patient months; 19% diabetics) during the period 08-03-1993 up to 12-31-2016. Twenty-three persistently infected catheters needed surgical unroofi ng of the tunnel tract. The unroofi ng procedure essencially consist DOI: http://dx.doi.org/10.17352/acn.000027 by incising the skin from above the superfi cial cuff and the exteriorization of the subcutaneous external cuff removing the infected tissue many times helped by an electrosurgical knife. After sanitized the wound, absorbent monofi lament threads are used to repair the subcutaneous tissue, fi nally, suture the skin [5]. In our protocol we do not shaving the exteriorized cuff.
We calculated the cumulative catheter survival using the Kaplan-Meier method of: 1) overall catheters, 2) not unroofed catheters, considering in (1) and (2) the catheter removal as an endpoint. Also, we measured 3) the unroofed catheter survival (considering the date of unroofi ng or catheter removal as an endpoint) and 4) continuity of the unroofed catheters postunroofi ng, considering the unroofi ng date as if it were a new catheter and catheter removal as the endpoint. Death of the patients, renal transplant or transfer to another dialysis centre were censored. We compared the survival of: a) unroofed catheters vs. continuity of the unroofed catheters and b) no unroofed catheters vs. continuity of the unroofed catheters. To compare the survival curves, we used the log-rank test (signifi cance value P< 0.05). We also used the Chi-Square test to analyse the proportion of unroofi ng between Tenckhöff and Swan Neck catheters (signifi cance value P < 0.05). Data were analysed using the software IBM-SPSS statistic, version 24.
Twelve out of 34 straight Tenckhöff and 11 of 104 Swan neck catheters required unroofi ng. Therefore, there was a greater need for this procedure in the former than the later catheters (χ² P <0.05).

Discussion
Similar to arteriovenous fi stula survival for haemodialysis, catheter lifespan is an important indicator of the quality of the peritoneal dialysis programs. In patients undergoing chronic haemodialysis, the creation of a new arteriovenous fi stula following vascular access failure likely has a different body sensation and uncertainty compared to the placement of a new peritoneal catheter for patients on peritoneal dialysis, especially when the indication of peritoneal dialysis was made because of the impossibility of continuing in haemodialysis.
Since the beginning of the chronic peritoneal dialysis, the characteristic of the peritoneal catheters (length, shape, gauge, material, cuffs number, etc.), as well as their placement and survival, have been studied [9][10][11]. On the other hand,   it is interesting to note that the survival rates of peritoneal catheters are equivalent to the survival of arteriovenous fi stulas for haemodialysis [12][13][14][15][16].
During chronic treatments in which access to therapy depends on an artifi cial device, complications related to the material and methodology are inevitable. During peritoneal dialysis, patient compliance to the guidelines of care taught during the training of the technique is important.
Exit site infection is an important threat to catheter survival and, therefore, of the continuity of treatment in peritoneal dialysis. The therapy of this infection has been extensively studied in numerous articles and guidelines of the ISPD [17,18].
During the evolution of our peritoneal dialysis program, we performed a thorough follow-up of catheter survival and the impact of the unroofi ng on the extension of the useful life of the catheters. We found that long-term overall catheter survival is satisfactory [9,26,27]. The survival of unroofed catheters (considering the date of the procedure as if day-zero of a new catheter) was similar to that of not unroofed catheters.
Thus, there was a positive contribution of this procedure to the overall survival of catheters.
It is recognised that there are no differences in the outcomes when comparing different types of peritoneal catheters [28,29]. Shape memory of the peritoneal catheters has been implicated in the catheter migration process [30]. In a prospective randomised study, Lye and co-workers observed that the Swan neck confi guration resulted in a signifi cant reduction in the rate of exit-site infection [31]. In our study, it is important to highlight that the Tenckhöff catheters had the highest proportion of infected catheters that required unroofi ng; it seems likely that catheter migration related to the shape memory of the straight catheters will have facilitated the exit site erosion and secondary infection.

Conclusion
The overall long-term catheter survival rate was satisfactory. Unroofi ng contributed signifi cantly to the lifespan of the catheters. This simple procedure should be considered in persistently infected catheters before deciding to remove the catheter.