OUTCOMES OF AMBULATORY PERITONEAL DIALYSIS CATHETER INSERTION IN THE NATIONAL KIDNEY AND TRANSPLANT INSTITUTE

The National Kidney and Transplant Institute (NKTI) in the Philippines has estimated that every hour, or about 120 Filipinos per million population a year, develops chronic kidney disease (CKD). For the past decade, the Philippines has seen a 400% increase in prevalence of CKD patients requiring dialysis. Currently, there are more than 70,000 Filipinos undergoing renal replacement therapy (RRT) – with hemodialysis (HD) being the most common form of RRT at 89%, followed by peritoneal dialysis which accounts for only 9%. Although HD and PD are equally effective modalities of RRT, PD offers several advantages such as patient autonomy, convenience and comfort of doing RRT at home, less exposure to medical facilities such as HD centers, better preservation of residual renal function, and higher survival advantage at early course of RRT. The PD First Policy refers to the policy where peritoneal dialysis, when feasible, is offered as the first dialysis modality to RRT patients. The NKTI is the first healthcare institution to adopt this policy and has implemented it in 2004. Since then, many patients get admitted in NKTI for this sole purpose and has been contributing to shortage of hospital beds and exhaustion of medical resources and personnels. PD involves a percutaneous insertion of a catheter to the peritoneum where the process of dialysis takes place. It may be done on an inpatient or outpatient basis. A program advocating for ambulatory peritoneal dialysis catheter insertion (PDCI) has been implemented in the NKTI last 2022 to encourage outpatient RRT initiation and avoid unnecessary hospital admissions. Data from different countries worldwide have shown no significant difference in PD outcomes between an inpatient or outpatient setting. Whether or not this holds true in the Philippine setting remains to be seen.

The general objective of this study is to determine the outcomes of patients who underwent ambulatory PDCI in the National Kidney and Transplant Institute (NKTI). The specific objectives includes 1) To determine the demographics of patients undergoing ambulatory PDCI in NKTI, 2) To determine outcomes of ambulatory PDCI in terms of infectious PD-related complications, non-infectious PD-related complications, technique survival, hospitalizations, and mortality; 3) To determine the onset of the different outcomes of ambulatory PDCI at several points from immediate post-PDCI (< 7 days), early (< 90 days), late (≥ 90 days), and long-term outcomes (6-12 months).

Research Design: This is a single-center, descriptive, retrospective study from August 1, 2022 to December 31, 2023.

Study population/Inclusion Criteria: All ESKD patients who underwent ambulatory PDCI in NKTI from August 1, 2022 to December 31, 2023 were included in the study.

Data Collection Procedure: Data were collected from electronic medical records of all who underwent the ambulatory PDCI from August 1, 2022 to December 31, 2023.

Data Presentation: Participants consist of the Ambulatory PDCI patients. The diagnostic classification of the participants is based on the cause of the end-stage renal disease obtained from the electronic database. Patient demographics, incidence of infectious complications (peritonitis and catheter-related infections), noninfectious complications (catheter-related mechanical complications, leaks, and other problems), hospitalizations, technique survival, and mortality were recorded. Participants’ post-PDCI outcomes were followed up at several points from immediate post-PDCI (< 7 days), early (< 90 days), late (≥ 90 days), and long-term outcomes (6-12 months). Technique failure was defined as transfer to hemodialysis therapy because of peritonitis, ultrafiltration failure, inadequate dialysis, exit-site or tunnel infection (or both), or mechanical problems.

Statistical Analysis: To summarize the general and clinical characteristics of the participants, descriptive statistics were employed. Nominal variables were evaluated using frequency and proportion, non-normally distributed interval/ratio variables were assessed using median and range, and normally distributed interval/ratio variables were evaluated using mean and standard deviation.  Missing variables were not replaced nor estimated. Data analysis was conducted using R 4.2.2.

Table 1 summarizes the demographic and clinical characteristics of the 28 patients who underwent peritoneal dialysis catheter insertion (PDCI) at the National Kidney and Transplant Institute. The median age of the cohort was 42 years (range: 22–69), with a slight predominance of females (57.1%). The median body mass index (BMI) was 23.1 kg/m² (range: 17.0–30.8), with most patients falling within the normal BMI range (64.3%). Residual renal function varied widely, with a median of 500 mL/kg/day (range: 100–1000). Hypertension was present in 21.4% of patients, while 35.7% had diabetes mellitus. Half of the patients reported having a caregiver, and the majority had achieved at least a high school education (53.6%) or held a college degree (42.9%). All catheter insertions were performed via an open surgical route.

Table 2 summarizes the infectious complications among the 28 patients, with 37 episodes reported. The most common complication was peritonitis, occurring in 11 patients (39.3%) with a total of 21 cases, accounting for 56.8% of all episodes. The median time to onset of peritonitis was 192 days (range: 32–472 days). Tunnel site infections (TSIs) were observed in 5 patients (17.9%) with 5 cases, and exit-site infections (ESIs) occurred in 6 patients (21.4%) with 11 cases. The median time to onset was shorter for TSIs (60 days; range: 16–386 days) compared to ESIs (181 days; range: 13–497 days). PDCI-related infections were less frequent and occurred earlier than PD-related infections for both TSIs and ESIs. The infection density rate (IDR) was highest for peritonitis (0.54 episodes per patient-year), followed by ESIs (0.28 episodes per patient-year).

Table 3 presents the 21 non-infectious complication events. Peritoneal dialysis catheter (PDC) migration was the most frequent non-infectious complication, reported in 9 patients (32.1%) with 13 cases making 61.90% of all non-infectious complication cases. Most migrations occurred within the first 90 days, with a median time to onset of 46 days (range: 9–494 days). Other complications included omental wraps (4 unique cases in 4 patients), leaks (2 unique cases in 2 patients), and extruded cuffs (2 unique cases in 2 patients). Leaks occurred early, with a median time to onset of 40 days (range: 24–56 days), while extruded cuffs had a delayed median onset of 326.5 days (range: 272–381 days). Omental wraps showed variability in timing, with a median onset of 119.5 days (range: 27–312 days).

Table 4 summarizes treatment outcomes, including technique failure, hospitalization, and mortality among the 28 patients. Technique failure occurred in 7 patients (25.0%) at an incidence density rate (IDR) of 0.18 episodes per patient-year, with a median onset of 200 days (range: 39–442 days). Causes included PDC migration (4 cases), omental wraps (2 cases), and peritonitis (3 cases). Hospitalizations were reported in 14 patients (50.0%), with 31 events at an IDR of 0.8 episodes per patient-year. The median time to the first hospitalization was 179 days (range: 9–550 days), with 66.7% of hospitalizations occurring within 90 days. Mortality was observed in 4 patients (14.3%) at an IDR of 0.10 episodes per patient-year, with half of the deaths occurring beyond one year. Causes of death included acute ischemic stroke (two patients), sudden cardiac death and septic shock from refractory PD-related peritonitis (Acinetobacter baumanii). 

Table 5 summarizes the timeline of complications and outcomes, highlighting that most complications occurred within the first 90 days. Among infectious complications, 45.5% of ESIs and 60.0% of tunnel site infections were observed early, while only 19.1% of peritonitis cases occurred during this period. For non-infectious complications, 76.9% of PDC migrations and all leaks were reported within 90 days, while rarer events like extruded cuffs and some omental wraps emerged later. Outcomes such as technique failures, hospitalizations, and mortalities showed variable timing, with a significant portion occurring beyond six months.

Our study shows a median age of 42 years, with slight female predominance. It is consistent with global data that patients being PD-initiated are now younger. We have observed that diabetes is the most common comorbidity followed by hypertension which is also similar with the global epidemiology. Previous studies done in NKTI have these similar findings as well, but with male predominance.

Similar to data worldwide, PD peritonitis is the most common infectious complication among ESRD patients who underwent ambulatory PDCI, with peritonitis rate at 0.54 episodes per patient-year (ppy).  A retrospective study from 2015 to 2020 in NKTI involving PD patients who had inpatient PDCI also revealed a similar peritonitis rate of 0.55 episodes ppy. ISPD guidelines recommend improving PD-associated peritonitis incidence to <0.4 episodes ppy – which the NKTI has yet to achieve. Other countries such as USA, Japan, Australia, and New Zealand all have less than <0.4 episodes ppy of peritonitis. The Asia-Pacific region has the highest rate peritonitis of followed by Europe, Middle East, and Africa were associated with higher. Next commonly occurring infections in PD patients are ESI and TSI. In our study, ESI and TSI has an incidence rate of 0.28 and 0.13 episodes ppy respectively. Previous data in NKTI shows lower rates of  ESI at only 0.11 episodes ppy. TSI rates were not included in recent studies in NKTI.

Though PD catheter mechanical complications were not directly at the top of the hierarchy in Standardised Outcomes in Nephrology-Peritoneal (SONG-PD) outcome monitoring, it is highly linked to recurrent infection and technique failure. In fact, our study showed that the earlier the PDC mechanical complications occur, the more likely it is to be followed by repeated complications, both infectious and non-infectious. PDC migration is the most common non-infectious complication in our study at a rate 32.14%, which is higher than the previous study done in NKTI in 2018 which revealed only a 16.9% rate out of 159 individuals studied.

The shorter time-to-first-episode of complication was associated with repeat episodes of complications and technique failure, as observed in a previous study done in NKTI. In our study, we have observed 13 patients (46%) had an early-onset complications (both infectious and non-infectious), with 9 (69%) out of these patients had succeeding repeated complications. This indicates that a more frequent monitoring during these times may be warranted.

Technique failure occurred in 7 (25%) of our patients. Five (71.4%) of them all had an early complication of PDC migration at a median onset of 22.2 days (11 – 46 days). These findings are different from a previously done study in our institution which showed a high technique failure rate at 57.9% with PD peritonitis being the most common cause. The other 2 patients in our study who had a technique failure were due to an omental wrap and TB peritonitis.

Death-censored technique survival in our study was observed to be at 71%, with majority (92.8%) of patients followed up over a year. To date, the worldwide technique survival of PD patients varies widely from 29 to 91% which were attributed to differences in defining PD start date, PDCI date, and other factors. Previous studies in NKTI revealed a 1-, 3-, 5-year mean survival of 85%, 60%, and 40% respectively.

Mortality was observed in 4 (14%) of our patients, with only one case associated with PD. The patient was an 18-year-old male with clinically-diagnosed chronic glomerulonephritis without comorbidities, who deteriorated from septic shock secondary to refractory PD-related peritonitis (Acinetobacter baumanii). The patient had a total of 4 PD-associated infections with the earliest onset of 13 days post-PDCI. A previous study in NKTI from 2015 to 2020 revealed an almost similar mortality rate at 18% but with PD-associated infections being the most common cause of death. The other two patients in our study died of acute ischemic stroke, and the other one had a sudden cardiac death. Worldwide, the most common cause of death among PD patients is cardiovascular disease.

Overall, our study has shown that ambulatory PDCI has a similar outcome with in-patient PDCI in NKTI. A study done by Salonen in 2014 comparing outcomes of outpatient and inpatient PDCI yielded similar results as well.