Immediate-start peritoneal dialysis without break-in-period: an 18-year retrospective cohort study on patient survival
Article information
Abstract
Background
Immediate-start peritoneal dialysis (ISPD) is an effective renal replacement therapy that can prevent central venous catheterization due to its immediate initiation of peritoneal dialysis (PD) after catheter insertion without a break-in period. This study aimed to investigate the effect of ISPD on long-term patient survival.
Methods
In this retrospective single-center cohort study, 178 consecutive patients who started PD from August 2005 to March 2023 were enrolled, from whom 144 patients with ISPD were analyzed. PD was initiated without a break-in period within 24 hours of catheter insertion using percutaneous needle-guidewire technique. The primary outcome was patient survival, estimated using the Kaplan-Meier method. A Cox proportional hazard regression model was used to identify factors independently associated with patient survival.
Results
The median follow-up period was 4.00 years (interquartile range, 1.23‒5.75 years). The mean age of patients was 61.6 ± 13.6 years; 58 patients (40.3%) were male and 93 patients (64.6%) were diabetic. Overall patient survival rates at 1, 3, 5, and 10 years were 98.5%, 93.5%, 92.1%, and 65.6%, respectively. The technique survival rates at 1, 3, 5, and 10 years were 88.1%, 74.9%, 63.2%, and 40.2%, respectively. The peritonitis-free survival rates at 1, 3, 5, and 10 years were 92.3%, 76.0%, 59.4%, and 28.0%, respectively. In the multivariate analysis, diabetes was the only factor associated with patient survival and technique survival.
Conclusion
Our study demonstrated that patient survival and technique survival rates were relatively high in ISPD patients who were catheterized using percutaneous needle-guidewire technique.
Introduction
With the exponential increase in the number of patients with chronic kidney disease worldwide, the number of patients with end-stage kidney disease (ESKD) requiring dialysis has also increased significantly [1]. This disease has become a major global public health problem. Particularly, unplanned dialysis is more problematic because patients undergoing unplanned dialysis after ‘late referral’ to a nephrologist account for more than 60% of patients with ESKD who newly started dialysis [2]. In many cases, patients undergoing unplanned dialysis must begin treatment with hemodialysis (HD) using a central venous catheter (CVC). Unplanned HD using a CVC may increase the risk of catheter-related complications including bacteremia, central venous stenosis, thrombosis, and mortality [3–5]. Therefore, peritoneal dialysis (PD) is sometimes used instead of HD using a CVC in patients who require unplanned dialysis. However, PD is usually started 2 weeks after peritoneal catheter insertion because of the risk of peri-catheter leaks or dysfunction, which is referred to as conventional-start PD (CSPD) [6,7]. Patients requiring emergency dialysis within a 2-week break-in period must undergo central venous catheterization for emergency HD. Early-start PD (ESPD) and urgent-start PD (USPD), which are initiated before a 2-week break-in period, help to prevent CVC insertion [8–11]. The clinical outcomes of ESPD, which initiates PD within 3–14 days after peritoneal catheter insertion, and USPD, which initiates PD within 72 hours after catheter insertion, are not significantly different from those of HD using CVC or CSPD [11–15]. These studies reported that the risks of mechanical complications, infectious complications, and technique failure of ESPD and USPD were similar to those of CSPD, even if PD was initiated earlier, without a 2-week break-in period. It has also been reported that USPD has a 24% lower hospitalization rate than urgent-start HD (USHD) [16]. Despite there being few studies on patient survival rates in ESPD and USPD, it has been reported that the long-term survival rate of ESPD patients who underwent unplanned dialysis was similar to that of CSPD patients [17,18]. Furthermore, observational data have implied that USPD has a similar or lower risk of mortality compared with USHD [3,19–21]. These findings suggest that ESPD and USPD are feasible treatment options for patients with ESKD requiring unplanned dialysis.
However, in patients with severe uremia requiring urgent dialysis, PD should be initiated immediately after peritoneal catheter insertion to prevent temporary HD using a CVC. Few studies have investigated the short-term clinical outcomes of immediate-start PD (ISPD), which is defined as PD initiation immediately after peritoneal catheter insertion without a break-in period. They reported that the risk of mechanical complications, infectious complications, or technical failure of ISPD was not higher than that of planned CSPD, and that ISPD has a similar or lower risk of the aforementioned outcomes than planned PD [10,22–24].
However, little is known about the long-term survival of patients with ISPD. Almost all patients with PD in our center are undergoing ISPD, and short-term clinical outcomes, such as mechanical and infectious complications, including catheter migration, have already been reported [10,24]. Therefore, this retrospective study aimed to evaluate the long-term clinical outcomes of ISPD, including patient survival, and identify the factors independently associated with long-term clinical outcomes.
Methods
Study population
This is a longitudinal retrospective cohort study of patients with ESKD who started PD from August 1, 2005 to March 31, 2023, at the Peritoneal Dialysis Clinic of Konkuk University Medical Center. Patient demographic data, including age, sex, body mass index, cause of ESKD, comorbidities, date of catheter insertion and removal, type of peritoneal catheter used, and PD modality were collected from the electronic medical record system. Laboratory parameters such as hemoglobin, albumin, low-density lipoprotein cholesterol, total cholesterol, intact parathyroid hormone, and high-sensitivity C-reactive protein levels were also measured. Additionally, clinical data such as the date and episode of peritonitis were collected. The inclusion criteria were patients aged >18 years and PD catheter implantation using the guidewire-assisted percutaneous insertion method at Konkuk University Medical Center during the study period. The exclusion criteria were as follows: break-in period of >24 hours, surgical insertion of a PD catheter, observation for <3 months, and PD/HD hybrid dialysis therapy.
The study protocol was approved by the Institutional Review Board of Konkuk University Medical Center (No. KUH2023-03-016). This study was completed in accordance with the Declaration of Helsinki. Informed consent was waived owing to the retrospective design of the study.
Immediate-start peritoneal dialysis protocol
As described previously, one experienced nephrologist inserted all the PD catheters using a guidewire-assisted percutaneous implantation method under local anesthesia [10]. All participants select PD as a kidney replacement therapy (KRT) modality through the ‘shared decision making (SDM)’ process. We used either a straight or a swan-neck double-cuffed straight-end Tenckhoff catheter. PD was initiated immediately after catheter insertion without a break-in period. It was performed with 500 mL of 1.5% dialysis solution for the first 3 days, 1,000 mL for the next 4 days, and 2,000 mL for the second week. Subsequently, continuous ambulatory PD (CAPD) or automated PD (APD) was performed depending on the patient’s condition. All patients were treated with a glucose-based PD solution, and icodextrin solution was used as needed.
Outcomes
The primary outcome measure was patient survival. The secondary outcomes included technique failure, peritonitis-free survival, and factors independently associated with patient survival. Technique failure was defined as a transfer from the PD for >30 days. In the analyses examining technique failure, patients were censored for any cause of technique failure. Peritonitis-free survival was defined as the time between the initiation of PD and the first episode of PD-related peritonitis. Peritonitis was defined according to the ISPD peritonitis guidelines (2022 updates on prevention and treatment) [25].
Statistical analysis
Normally distributed continuous variables are expressed as means and standard deviations. Skewed continuous variables are expressed as medians with interquartile ranges (IQRs). Categorical variables are presented as counts and percentages. The primary and secondary outcomes were analyzed using the Kaplan-Meier method and Cox proportional hazard regression model. The Kaplan-Meier method was used to analyze the survival rates of the primary and secondary outcomes. In the analysis of patient survival, patients were censored at the time of transfer to another hospital, loss to follow-up, kidney transplantation (KT), death from unrelated causes, or switching from PD to HD or PD/HD hybrid therapy. They were also censored at the end of the observation period (March 31, 2023) if they remained on the PD. The Cox proportional hazard regression model was used to analyze the factors independently associated with primary and secondary outcomes. The demographic data, clinical characteristics, and laboratory parameters were included as covariates. Multivariate Cox regression analysis was performed using univariate analysis variables with the p-values of <0.20 in the univariate analysis. Statistical significance was set at p < 0.05. Data were analyzed using the Stata MP/17 software (StataCorp LLC).
Results
Patient characteristics
Overall, 178 consecutive patients with PD were enrolled in this study, from whom 144 patients were analyzed. Thirty-four patients were excluded for various reasons (Fig. 1; Supplementary Table 1, available online). The median observation period was 4.00 years (IQR, 1.23‒5.75 years). At the end of the follow-up period, 47 patients (32.6%) remained on PD, 17 (11.8%) died, 52 (36.1%) were transferred to HD, 14 (9.72%) underwent KT, and 14 (9.72%) were lost to follow up. The baseline demographics and characteristics are presented in Table 1. The mean age was 61.6 ± 13.6 years, 58 patients (40.3%) were male, and 93 (64.6%) were diabetic. Swan neck and non-swan neck catheters were used in 114 (79.2%) and 30 patients (20.8%), respectively. After starting ISPD, 89 patients (61.8%) underwent CAPD and 55 (38.2%) underwent APD. During the observation period, 70 patients (48.6%) developed one or more episodes of peritonitis while maintaining PD.
Flow diagram for enrollment and follow-up for ISPD patients.
HD, hemodialysis; ISPD, immediate-start peritoneal dialysis; IQR, interquartile range; KUMC, Konkuk University Medical Center; PD, peritoneal dialysis.
Baseline demographics and characteristics of participants
Clinical outcomes
The 1-, 3-, 5-, and 10-year patient survival rates were 98.5%, 93.5%, 92.1%, and 65.6%, respectively (Fig. 2A). The most common cause of death was infection (35.1%), followed by cardiovascular diseases (CVDs; 27.0%). In the comparison of patient survival between the diabetes mellitus (DM) and non-DM groups, non-DM patients showed a significantly higher survival rate (log-rank p < 0.001) (Fig. 2B). The 1-, 3-, 5-, and 10-year survival rates of patients without DM were 100.0%, 100.0%, 100.0%, and 95.2%, respectively. The 1-, 3-, 5-, and 10-year survival rates of patients with DM were 97.6%, 89.5%, 87.0%, and 35.1%, respectively. The 1-, 3-, 5-, and 10-year technique survival rates in the total population were 88.1%, 74.9%, 63.2%, and 40.2%, respectively (Fig. 2C). During the study observation period, 52 out of 144 patients (36.2%) switched from PD to HD. The mean time to transfer to HD was 38 months from the initiation of PD. The most common cause of technique failure was peritonitis (50.0%), followed by inadequate PD (19.6%). The mean time until the first episode of peritonitis was 27 months. The 1-, 3-, 5-, and 10-year peritonitis-free survival rates were 92.3%, 76.0%, 59.4%, and 28.0%, respectively (Fig. 2D). Although early-onset catheter migration, peri-catheter leakage, and infection occurred in 62.4%, 2%, and 4% of all patients, respectively, no patients switched to HD because of early complications such as catheter migration, peri-catheter leakage, and infection; however, the PD catheter was re-inserted in one patient due to outflow failure by PD catheter migration.
The Kaplan-Meier survival curve.
(A) Patient survival, (B) patient survival by diabetes mellitus (DM) and non-DM, (C) technique survival, and (D) peritonitis-free survival.
Factors associated with clinical outcomes
The results of univariate and multivariate Cox proportional hazards regression models associated with 10-year patient survival are presented in Table 2. DM and CVD were associated with long-term patient survival in the univariate Cox regression analysis. However, in the multivariate Cox regression analysis, DM was the only factor independently associated with long-term patient survival. The univariate and multivariate Cox proportional hazards regression models associated with 1-, 5-, and 10-year patient survival rates are presented in Supplementary Table 2 (available online). In the univariate Cox regression analysis, smoking was associated with short-term patient survival; however, this association disappeared in the multivariate Cox regression analysis. The results of the univariate and multivariate Cox proportional hazards regression models associated with technique survival are presented in Table 3. DM and CVD were associated with long-term technique survival in the univariate Cox regression analysis. However, DM was the only factor associated with 10-year technique survival in the multivariate Cox regression analysis. The univariate and multivariate Cox proportional hazard regression models associated with 1-, 5-, and 10-year technique survival are presented in Supplementary Table 3 (available online). No factors were associated with short-term univariate or multivariate Cox regression analyses. Furthermore, no factors were associated with peritonitis-free survival in the univariate and multivariate Cox regression analyses (Table 4; Supplementary Table 4, available online).
Univariate and multivariate Cox regression analysis for 10-year patient survival
Univariable and multivariable Cox regression analysis for technique survival
Univariable and multivariable Cox regression analysis for peritonitis-free survival
Discussion
This cohort study, which followed up 144 patients with ISPD for 18 years, demonstrated that immediate initiation of PD after catheter insertion without a break-in period resulted in excellent long-term patient survival. The short- and long-term patient survival rates were not only higher than those of ISPD reported previously, but also noninferior compared to that of CSPD reported in previous studies. In addition, the long-term technique success and peritonitis-free survival rates were comparable to those previously reported for CSPD. The results of this study suggest that ISPD can be effectively used as renal replacement therapy (RRT) for ESKD patients with planned PD. In addition, if it is possible for a nephrologist to insert a PD catheter on an emergency basis, it suggests that it can be used as an RRT to prevent central venous catheterization in patients requiring emergency dialysis.
Generally, PD begins 2 weeks after peritoneal catheter insertion, and a 2-week break-in period is recommended in the ISPD guidelines [7]. However, the optimal break-in period remains unknown. Given the increase in the number of patients on unplanned dialysis who require dialysis urgently, early initiation of PD by shortening the break-in period to <2 weeks is being attempted [8–11,26]. Recently, Blake and Jain [27] defined ESPD as having a break-in period of 3 to 14 days after catheter insertion and USPD as having a break-in period of <72 hours after catheter insertion. However, this definition is limited since it does not include cases of PD initiation immediately after catheter insertion without a break-in period. In this study, we defined ISPD as PD started immediately without a break-in period, and then analyzed the effects of ISPD on long-term clinical outcomes, including patient survival.
In this study, all patients who underwent ISPD in a single center within 18 years (from 2005 to 2023) were followed up until conversion to HD, renal transplantation, transfer to another hospital, or death. The median patient observation period was 4 years, and the 1-, 3-, 5-, and 10-year survival rates of patients with ISPD were 98.5%, 93.5%, 92.1%, and 65.6%, respectively. The long-term survival patient rate from our study is higher than previously reported patient survival rates of CSPD, ESPD, and USPD [17,18,20]. A 2017 review of the global epidemiology of PD reported a 5-year survival rate of 48.4% to 64% [28], while in the 2022 report of the United States Renal Data System, the contemporary adjusted 5-year survival of PD patients in the 2016 cohort was 44.3% [29]. Therefore, our findings suggest that ISPD can be used as RRT in patients requiring urgent dialysis.
The survival rate of patients with CSPD with a break-in period of >14 days is not different from that of patients with HD. Since the mid-1990s, the risk of death among patients with PD has decreased significantly globally. Consequently, almost all recent studies have reported that PD and in-center HD have similar short-term (1 or 2 years) or long-term (up to 5 years) survival rates [15,29–33]. A study that analyzed 96,596 patients from the National Health Insurance Service database who started dialysis between 2004 and 2015 in Korea also reported the same trend [34]. The adjusted hazard ratio (HR) of PD versus HD for mortality was 1.31 (95% confidence interval [CI], 1.27–1.36; p < 0.001) for 2004 to 2007 and 1.21 (95% CI, 1.16–1.27; p < 0.001) for 2008 to 2011. However, the HR of PD to HD for mortality was not statistically significant from 2012 to 2015.
Recently, studies have compared the survival rates of patients undergoing ESPD and USPD with those of patients undergoing CSPD. However, the results varied from study to study: the 6-month, 1-year, and 5-year survival rates were 82% to 94%, 83% to 85.9%, and 36.1%, respectively. The 5-year survival rate was particularly lower than the previously reported 5-year survival rate of PD patients [17,30,35,36]. In a 30-month follow-up study involving 202 ESPD patients and 433 CSPD patients, Wang et al. [17] observed that the all-cause mortality of ESPD patients was higher than that of CSPD patients. Among patients with ESPD, the 6-month, 1-year, 3-year, and 5-year survival rates were 91.6%, 85.9%, 63.0%, and 36.1%, respectively. The 6-month, 1-year, 3-year, and 5-year survival rates of patients with CSPD were 94.3%, 91.4%, 74.0%, and 54.4%, respectively. There was a significant difference in patient survival rates between the two groups, especially in the long-term survival rate from the first year. However, in the subgroup analysis segregating the data into pre-2005 and post-2005, there was no difference in mortality between ESPD and CSPD in the post-2006 group. Bitencourt Dias et al. [36] prospectively observed 51 patients with USPD who started PD within 72 hours of catheter insertion for 6 months and reported a 6-month survival rate of 82.4%.
However, there are few reports on the long-term survival rate of patients with ISPD who started PD without a break-in period, and these studies mostly investigated the 6-month patient survival rate. Parapiboon et al. [37] reported in a randomized controlled trial comparing patients who underwent ISPD (n = 104) and USHD (n = 103) that there was no difference in the mortality rate at 6 weeks between the two groups (4% vs. 5%; relative risk, 0.79; 95% CI, 0.22–0.87). Similarly, Koch et al. [38] reported the 6-month mortality rates in patients who underwent emergency unplanned PD (n = 66) and unplanned HD using CVC (n = 57). In their study, a peritoneal catheter was inserted using a laparoscope and APD was initiated immediately after catheter insertion; the 6-month mortality rates for PD and HD were 30.3% and 42.1%, respectively, with no significant intergroup differences between the two groups. However, the 6-month patient survival rate (69.7%) in their study was significantly lower than the previously reported 6-month patient survival rates for CSPD, ESPD, and USPD. The high survival rate of our study may have been influenced by several factors, including a standardized catheterization procedure by an experienced nephrologist, a guidewire-assisted percutaneous insertion method accompanied by minimal injury, skilled PD nurse dedicated to patient care, thorough patient education, and comprehensive follow-up care by a well-trained PD team. Among these factors, the possibility that the catheter insertion method had a significant effect on mortality cannot be excluded because it is an important difference between our study and previous reports. Previous studies used the percutaneous Seldinger insertion method, laparoscopic insertion, or open surgery. In contrast, a guidewire-assisted percutaneous implantation method was used in the present study. Unfortunately, we did not compare the guidewire-assisted percutaneous implantation method with other insertion methods in this study. Therefore, a controlled study is needed to prove this hypothesis.
DM was the only factor that independently affected patient survival in the present study. In the univariate and multivariate Cox regression model, the DM group had an HR of 4.78, indicating that the risk of mortality in the DM group was 4.8 times higher than that in the non-DM group. The survival rates of patients without DM were 100%, 100%, 100%, and 95.2% at 1, 3, 5, and 10 years, respectively, whereas those of patients with DM were 97.6%, 89.5%, 87.0%, and 35.1% at 1, 3, 5, and 10 years, respectively. The mortality of non-DM patients remained at 0% for up to 5 years, which may be attributed to the relatively low mean age of non-DM patients and their low prevalence of comorbidities such as CVD. The technique success rates of the patients with ISPD in our study were 88.1%, 74.9%, 63.2%, and 40.2% at 1, 3, 5, and 10 years, respectively. This result is similar to previously reported technique success rates of CSPD and ESPD [12,13,37]. Liu et al. [35] analyzed the effect of the break-in period duration on technique success in a study of 657 patients with PD. The 6-month technique success rates were 94%, 99%, and 98% in patients with break-in periods of <7 days, 8 to 14 days, and ≥14 days, respectively. However, while the technique success was lower in patients with a break-in period of <7 days, a shorter break-in period was not an independent predictor of technique failure in the Cox multivariate analysis. Kim et al. [39] reported a 6-month technique success rate of 78.6% in a study conducted on patients who started PD without a break-in period after the percutaneous insertion of a peritoneal catheter using a method similar to that used in this study. They also compared the technique survival rates of the ISPD group, which started dialysis immediately after percutaneous peritoneal catheter insertion, and the ESPD group, which started PD 7 days after surgical insertion; there was no significant difference between the two groups, but the survival rates were significantly lower than the rates from the present study using the same catheter insertion method. These findings further suggest that factors other than the catheter insertion method may affect the technique success rate. This hypothesis is also supported by a 10-year retrospective study by Ye et al. [40], which included 2,059 patients who started PD immediately after inserting a peritoneal catheter with open laparotomy without a break-in period. In this study, the 1-month, 1-year, 3-year, and 5-year technique success rates of ISPD were 99.5%, 97.0%, 90.8%, and 82.0%, respectively; the technique success rate was significantly higher than that in our study, but the reason for this has not yet been clearly identified. Compared with our study, the report showed a significant difference in the mean age of patients (61.6 ± 13.6 years vs. 49.7 ± 29 years), which might have contributed to the difference in technique success.
In the present study, the peritonitis-free survival rate over 1 year was 92.3%, which was higher than that reported in previous reports. In a study by Hu et al. [23] who performed ISPD with a peritoneal catheter inserted through open surgery in diabetic patients, the incidence of peritonitis was 3.2%, 5.2%, 9.0%, and 12.3% at 2 weeks, 1 month, 3 months, and 6 months, respectively. Wen et al. [22], who performed ISPD with a surgically inserted peritoneal catheter using the same procedure as Hu et al. [23], reported a peritonitis-free survival rate of 90.2% at 90 days and 81.4% at 6 months. Meanwhile, in our 2007 study where ISPD was conducted with a percutaneous peritoneal catheter inserted, the peritonitis-free survival rate was 96% at 30 days and 82% at 12 months [10]. The enhanced peritonitis survival rate observed in this current study, in comparison to the 2007 study, is attributed to the systematic approach of patient education and monitoring, facilitated by the establishment of a specialized PD team that includes a dedicated PD nurse.
Our study has several limitations. First, since it is a retrospective study without a control group, it is difficult to conclude whether the clinical outcomes of ISPD are superior to those of ESPD, USPD, and CSPD. It is possible that selection bias may have affected the clinical outcome. However, during the study period, a total of 167 patients started PD at our center, and 94.0% (n = 157) of them underwent ISPD. Thus, although this is a retrospective study, we expect the effect of ‘selection bias‘ on patients’ long-term outcomes to be minimal. Second, this study is not a randomized controlled trial. Because ISPD was not compared with CSPD as the control group, it is unclear whether the higher success rate of ISPD is significantly higher than that of CSPD. Unfortunately, the number of patients with CSPD was too small to compare the two groups, but as mentioned above, among the patients who decided to select PD as the KRT modality through SDM at our center, 94.0% of the patients underwent ISPD, with a few exceptions. Thus, the influence of selection bias on the clinical outcome can be excluded to some extent. Finally, ISPD was not always performed in patients requiring emergency dialysis, due to the requirement of having a nephrologist capable of performing PD catheter insertion readily available. However, ISPD which began immediately after PD catheterization had an excellent long-term prognosis with only minor side effects, such as peri-catheter leakage, in very few patients. Therefore, if it is feasible to consistently ensure nephrologist availability, ISPD may be used as an effective KRT even in emergency situations, thereby avoiding CVC insertion for emergency HD. Our study also has several strengths. First, the patients enrolled for 18 years were followed up for 10 years. Second, this study investigated the long-term survival rate of patients with ISPD, which has not been previously reported. Finally, our study identified the risk factors that independently affect long-term clinical outcomes, such as patient survival, technique survival, and peritonitis-free survival in patients with ISPD.
In conclusion, our study showed that the patient survival rates of ESKD patients with ISPD after guidewire-assisted percutaneous peritoneal catheter implantation were noninferior compared to those previously reported for CSPD, ESPD, and ISPD. In particular, the long-term patient survival rate of the non-DM patient group was remarkably high. These results suggest that ISPD can be used as an effective RRT for patients with ESKD requiring dialysis. Furthermore, if a nephrologist can insert a PD catheter percutaneously in an emergency, ISPD may be used as an effective RRT even in emergency situations, thus preventing the need for CVC insertion for emergent HD.
Supplementary Materials
Supplementary data are available at Kidney Research and Clinical Practice online (https://doi.org/10.23876/j.krcp.23.103).
Notes
Conflicts of interest
All authors have no conflicts of interest to declare.
Funding
This work was supported by Konkuk University Medical Center Research Grant 2022.
Data sharing statement
The data presented in this study are available on request from the corresponding author.
Authors’ contributions
Conceptualization: YIJ, JYL
Data curation, Investigation: JYL, HJC, YSJ, HJN, JHP
Formal analysis, Visualization: JYL
Funding acquisition, Resources, Methodology, Project administration, Supervision, Validation: YIJ
Writing–Original Draft: JYL
Writing–Review & Editing: All authors
All authors read and approved the final manuscript.