Peritoneal Dialysis as an Option of Unplanned Initiation of Chronic Dialysis

1. Background: Chronic Kidney Disease (CKD) is a matter of global public health. Data from United States of America (USA) has shown a steady increase in the number of patients on renal replacement therapy (RRT). The census of the Brazilian Society of Nephrology (SBN) showed 65,121 patients on dialysis program in 2005, while in 2011 there were 91,314 and in 2013, there were more than 100,000 patients on dialysis in Brazil. Among the methods of dialysis, nephrologists have two options: the peritoneal dialysis (PD) and hemodialysis (HD), which is the most commonly used method in the world. Although, historically, PD was widely used in nephrology, for not clear reasons it has been underutilized in last years, especially in incident patients on RRT. In Brazil, data from 2013 showed that 90.6% of chronic patients underwent HD and only 9.4% were treated by PD. Several studies have compared the differences between the two types of dialysis - PD vs. HD - in incident patients on RRT. There is no evidence of the superiority of one method over the other in regards to general mortality within the first two years of therapy. Recently, some authors have pointed out the impact of the vascular access used in the mortality of incident patients in HD. These studies found that central venous catheters (CVC) use is associated with reduced survival, especially in the first 90 days of RRT. Furthermore, there is a greater risk of bacteremia, sepsis and hospitalizations in patients using CVC when compared to patients using arteriovenous fistulas (AVF) or graft or PD. In this scenario, the PD appears as an option of unplanned initiation of chronic dialysis. Starting dialysis in an unplanned manner is a frequent situation in dialysis centre even for patients without or with a regular nephrology follow-up. Unplanned dialysis initiation can be defined as an end-stage renal disease patient beginning HD with no functional vascular access, using CVC or as the starting PD in less than 7 days after peritoneal catheter implantation. It has been shown that unplanned dialysis and using temporary HD catheters are independently associated with mortality in incident dialysis patients. For no clear reasons, PD is underused in unplanned dialysis patients. In Brazil, approximately 60% of incident patients on RRT have no definitive access and need to be treated through CVC. However, there are few studies that describe PD method as an immediate treatment option in patients without functioning vascular access and only two small studies that compared unplanned start of HD vs. PD. Lobbedez et al followed by the 2-year period, 60 patients who started unplanned dialysis - 34 patients treated by PD and 26 by HD. Among the patients who started on PD, no significant difference in mechanical or infectious complications when compared to patients who had "rest time" post-catheter insertion. There was no significant difference between the two unplanned dialysis methods (HD x DP) in patients survival (78.8% in the HD group vs. 82.9% in PD (p = 0.26). After adjusting for comorbidities index, survival between the two groups remained similar. Koch et al. evaluated 57 incident patients in unplanned HD and 66 in unplanned PD. HD patients had more bacteremia than PD patients in the first 6 months of dialysis (21.1% vs 3%, p <0.01), which was associated with the use of CVC as initial access . However, there was no significant difference in the mortality rates between the two methods. So it is clearly needed to show that PD is a suitable method for unplanned dialysis patients and that acute automated PD may help nephrologists to deal with patients without any permanent vascular access at dialysis initiation. The investigators believe that PD can offer the advantage of not using CVC, which can reduce the morbidity and mortality of CKD patients. In addition, in October/2014, there was not vacancy for incident patients on chronic HD in Dialysis Unit of University Hospital of Botucatu Medical School, worrying fact for a high complexity service that has received an average of 8 incident patients on RRT, monthly.
2. Objectives: The main objective of this study is to compare technique and patients survival on unplanned PD vs. unplanned HD in the first 180 days of therapy. This study is also carried out:

• To compare the rates of hospitalization and infectious related with PD vs HD.
• To determine the impact on the PD and HD programs (growing or reduction)

1. Methodology 3.1. Patients: This study is a quasi experimental study that will evaluate incident unplanned PD and HD patients at Dialysis Unit of Botucatu Medical School. In an attempt to increase the use of PD for unplanned dialysis patients, a programme devoted to this group of patients has been set up since 1 November 2014.

Inclusion criteria are CKD patients stage 5 (creatinine clearance < 15 ml/min) requiring dialysis treatment immediately. Unplanned HD was defined as initiation without definitive vascular access and using CVC and unplanned PD as initiation in less than 48 hours after implantation of the peritoneal catheter.

Absolute contraindication to the PD method are: presence of recent abdominal surgery (less than 30 days); multiple previous abdominal surgery (more than two); severe respiratory insufficiency (FiO2> 70%); abdominal infections; severe hyperkalemia with changes characteristic in EKG; and acute pulmonary edema. These patients will be treated with HD.

The decision concerning the dialysis modality has been made according to the medical's indications and patient's choice. The implant of Tenckhoff catheter will be held through the percutaneous Seldinger technique by nephrology team. High volume PD (HVPD) will be used during the first 7 days of PD in order to achieve metabolic and fluid control. The procedure for acute PD has been published recently. When HD is the modality of choice, a temporary catheter is inserted rapidly by nephrologist, at the bedside if necessary, and after one week it will be changed to tunneled line.

Patients will be followed by the same research team from the time of the peritoneal catheter implantation until the outcome (stay in the method for a minimum period of 90 days, death, transplantation, recovery of renal function or switch to HD due to technical failure or infectious and mechanical complications without success with the treatment).

3.2. Research Protocol: The study protocol will consist of clinical and laboratory data, dialysis and monitoring of patients on PD and HD unplanned.

3.2.1. General data:

• Identification: name, gender, age, color
• Etiology and stage of CKD
• Presence of co-morbidities
• Start of unplanned dialysis (date)
• Date of the peritoneal catheter or CVC implantation
• Complications during or after implantation of the catheter
• Start date of family training (PD method)
• Date of the end of the family training (PD method)
• Number of training days (PD method)

3.2.2. Dialysis:

• Dialysis indications

PD: Catheter implantation using the Seldinger technique

• Prescription: HVPD to achieve adequate metabolic and fluid control. At this point, patient will be discharged from hospital.
• Intermitent PD on alternate days or daily: patients will be treated by PD at the dialysis unit of the University Hospital every other day or every day, according to their lab exams and clinic exam. Laboratorial exams will be performed weekly, until their family be trained and home be prepared for the implementation of technique.

HD: on alternate days at Dialysis Unit

3.2.3. Follow up of unplanned dialysis patients:

• Evaluation of mechanical and infectious complications related to PD or HD
• Technique and patients survival

3.3. Statistical Analysis: The sample size calculation was based on the assumption the difference of 15% between patients undergoing unplanned PD and HD in outcomes as hospitalization, infection and mortality. With a first-order error of 5% and a power of 80% a sample size of 94 patients is needed in each treatment group.

From the study protocol, the data will be entered in a spreadsheet and verified typographical errors and their analysis will be performed using the statistical program SAS for Windows (version 9.2: SAS Institute,Cary, NC, USA, 2012). Initially descriptive analysis will be done for all patients treated in the period, calculated measures of central tendency and dispersion for continuous variables and frequencies for categorical variables. Patients will be separated into two groups according to initiation modality (unplanned PD group and unplanned HD group) in order to evaluate the impact of unplanned PD on technique and patient survival and on survival free of peritonitis. Intent-to-treat analysis will be used to analyse patient survival. T-test will be used to compare parametric continue variables between the two groups and Kruskal-Wallis test for nonparametric continue variables. Chi Square will be used to compare categorical variable between the two groups. For the analysis of repeated measures, asymmetric distribution (gamma) under the GENMOD procedure will be used.

At the end of the study will be presented survival curves using Kaplan Meyer of patients and the method during the monitoring period. Cox proportional hazards model was used to adjust survival on baseline characteristics. The statistical difference was considered to be significant for p < 0.05.