Introduction
Despite the biggest and best efforts of the nephrology community to start patients on dialysis in a planned manner, nearly 60% of patients initiate dialysis in an unplanned way [1]. Traditionally, hemodialysis (HD) through a central venous catheter (CVC) has been the default treatment option for almost all patients, and peritoneal dialysis (PD) has not usually been considered a viable and safe option [2]. In the United States, in 2015, 92.8% of prevalent patients on dialysis were treated with HD [2]. Canadian registers showed that 80% of incident patients were treated with HD [3]. In Brazil, data from 2018 showed that 93.1% of patients in chronic dialysis are subjected to HD and only 6.9% to PD [3].
While PD was historically widely used in nephrology, it has been underutilized over the years due to unclear reasons, especially in patients incident in renal replacement therapy (RRT) [4]. Several reasons for the PD underutilization can be considered: the perception that it is inferior to HD because HD is associated with higher technology; the worries about infectious, mechanical, and metabolic complications associated with the PD method; the lack of experience and evidence concerning urgent-start PD, the relative ease of CVC catheter insertions and physicians’ familiarity and comfort with the initiation of HD in urgent-start dialysis; the greater financial reimbursement of HD and the difficulties with the implant of the peritoneal catheter by the nephrologist [2, 3, 5].
However, in the last decade, a number of reports describing the successful initiation of PD soon after the PD catheter insertion have been published making PD a viable option for urgent-start dialysis. This article examines the rationale behind this approach and reviews the current literature on the effectiveness, safety, and feasibility of urgent-start PD in developed and developing countries as compared to urgent-start HD and conventional PD.
Urgent-Start Dialysis: The Concept
In the last years, some authors have pointed out the impact of the type of vascular access utilized in the mortality of incident patients in HD [6-8]. These studies have identified that the use of CVCs is directly associated with lower survival, especially in the first 90 days of RRT. Besides that, there was major risk of bacteremia, sepsis, and hospitalizations when compared with the use of arteriovenous fistulas (AVF), arteriovenous graft, or PD catheter [6, 8].
Perl et al. [6] evaluated 40,526 patients between 2001 and 2008, 19% initiated PD and 81% in HD. Among the HD patients, 78.6% initiated RRT using CVC, which was an independent risk factor for death in comparison to patients which utilized the AVF method and those treated with PD, presenting mortality, after 90 days, of, respectively, 15.6, 6.1, and 7.4% (p < 0.05).
In this scenario, PD emerges as an option for unplanned start RRT in chronic kidney disease (CKD) patients stage 5 with no functioning vascular access, offering the advantage of lower use of temporary CVCs, besides the preservation of the vascular access and of the residual renal function [8-11].
By definition, urgent-start dialysis corresponds to the beginning of hemodialytic treatment without definitive vascular access (AVF), that is, through CVC or PD utilizing the catheter in <14 days after its implant [8-11]. Recently, the concept of urgent-start PD was revised by Blake and Jain [11], and it was considered a treatment strategy corresponding to patients with advanced but previously unknown CKD that start RRT in PD, using the Tenckhoff catheter within 72 h after the implant with no previous utilization of HD therapy [11].
Despite the great efforts of the nephrology community, the majority of CKD patients begin RRT in an unplanned manner [1]. Danish Register of Nephrology performed from 2008 until 2011 verified that 50% of incident patients on RRT started dialysis in an unplanned manner [8]. In Brazil, up 60% of incident patients had urgent-start RRT in 2018 [3].
Different protocols and prescriptions for urgent-start PD have been reported in the world literature. The utility of urgent-start PD is independent of the mode of PD catheter insertion. Successful urgent-start PD programs have been described with open surgical, laparoscopic, and percutaneous techniques of PD catheter insertions. Therefore, the method of PD catheter insertion is not a barrier to urgent-start PD provided the PD catheter can be inserted urgently at short notice. If clinically indicated, PD can be initiated immediately post PD catheter insertion [9, 10]. More stable patients can be discharged after the PD catheter insertion, and urgent-start PD can be started within a few days in an outpatient regime, either at an outpatient urgent-start PD facility or in the patient’s home under strict supervision [11-14].
The majority of the reported prescriptions are designed to provide intermittent PD (IPD) with low fill volumes in strict supine posture, either manually or with the help of a cycler [5, 9-12, 15-18]. Typical therapy consists of 4–5 cycles over 8–10 h a day [9-11]. Depending on the patient’s condition, the treatment can be provided from 3 times a week to daily [5]. The fill volumes are gradually increased during the initial 2 weeks of therapy [5, 9, 10]. However, some patients needed high-volume PD (HVPD) for achieving metabolic and fluid control [1, 18].
Recently, the interest in urgent-start PD has emerged and increased in both developing and developed countries, with several publications comparing the urgent-start PD with conventional PD and urgent-start HD in CKD patients [5, 10, 12-21].
PD Urgent Start in Developed Countries
The first description of an urgent-start PD program was performed by Lobbedez et al. [19], in France, in 2008. The authors studied 60 patients that initiated unplanned dialysis (34 in PD and 26 in HD), and there were no differences between the 2 dialysis methods regarding the patients survival (78.8% in the HD group vs. 82.9% in the PD group (p = 0.26).
In 2012, Koch et al. [18] evaluated for 5 years (2005–2010), in German, 57 incident patients in HD and 66 in PD, both with unplanned start, and showed no difference in mortality between the 2 methods. In 2013, Danish data showed similar results, but when comparing urgent-start PD to the planned PD program, the unplanned manner presented more mechanical complications, mainly leakage, though it did not affect the technique and patient’s survival [16].
Alkatheeri et al. [15], in a Canadian university hospital, have established since July/2010 the unplanned PD program. In 3 years, 30 patients were included, and mechanical or infectious complications were evaluated (first 4 weeks after implant of the PD catheter). Dialysate leak was observed in 3 patients (10%), it was conducted in a conservative manner; and migration of the catheter tip happened in 6 patients (20%). Despite the premature use of the PD catheter, no infectious complications were registered [15].
Povlsen and Ivarsen [12] described the urgent-start PD program in patients over 65 years with successful results. In their publication, they reported that the insertion technique of the Tenckhoff catheter was a mini laparotomy performed by surgeon and with immediate catheter use; however, they utilized a fixed prescription for all patients and initial maximum dialysate volume of 1.2 L. Normalization of the potassium and bicarbonate levels was reached in the second session and adequate urea levels were reached after 7 days of therapy. Survival patients in 3 and 12 months were, respectively, 90 and 80% [12].
An Australian publication compared the early complications (developed in <4 weeks) and long-term survival of patients submitted to urgent-start PD vs. planned PD. There was difference in early dialysate leak and migration of the catheter tip, being higher in the urgent-start group (12 vs. 1%, p = 0.047). No differences were observed in infectious complications. Despite the higher frequency of mechanical complications in the urgent-start group, there was no difference in survival of the technique in 1, 2, or 3 years (68 vs. 80%; 48 vs. 54%, and 48 vs. 38%, respectively) [16].
Urgent-Start PD in Developing Countries
In Brazil, approximately 60% of patients start dialysis in an unplanned manner, using CVC [3]. In July/2014, the dialysis unit from Clinical Hospital from Botucatu School of Medicine had no vacancies for the chronic hemodialytic treatment. Thus, PD emerged as an alternative of unplanned chronic dialytic method, with the advantage of not using CVC.
Absolute or relative contraindications to PD were considered: presence of recent abdominal surgery (<30 days); multiple previous abdominal surgeries (superior to 2); presence of fibrosis or peritoneal adhesions; fungal peritonitis; severe respiratory insufficiency (FiO2 >70%); abdominal wall infections; severe hyperkalemia with electrocardiographic characteristic changes; acute pulmonary edema [17, 20-22]. All catheters have been implanted by the Nephrology team using Seldinger technique, and flexible catheters with double cuff were utilized.
After the implant of the PD catheter, patients were submitted to HVPD, using automatic and continuous method (Home Choice cycler, one session = 24 h). Prescribed Kt/V was 0.5 per PD session, infusion volume was 20–30 mL/kg (average of 2 L), and dwell time ranged from 30 to 50 min [20, 22]. Supine position was adopted during the sessions. After hospital discharge, patients were kept in IPD, in alternated days or daily, according to the clinical and laboratorial evaluation [17, 23, 24].
During the IPD, training of families and/or patients was performed and, when necessary, small home adaptations to receive the therapy were done. After that, the patient was included in the program of chronic PD, with home treatment and monthly nephrologist consult [17, 23, 24]. Figure 1 shows the flowchart of the urgent-start PD program in Dialysis Unit of Brazilian Clinical Hospital.
Fig. 1.
The flowchart show the urgent-start PD program in dialysis unit of Brazilian clinical hospital. CKD, chronic kidney disease; PD, peritoneal dialysis; CVC, central venous catheter; HD, hemodialysis; HVPD, high-volume PD; RRT, renal replacement therapy.
Dias et al. [23], in Brazil, evaluated 51 patients in urgent-start PD. Mean age was 57.7 ± 19.2 years, diabetes was the main cause of CKD (40.6%), and uremia was the most common indication of RRT start (54.3%). The metabolic control was reached after 5 sessions of HVPD, and 11.5 sessions of IPD were performed at dialysis unit [23]. Mechanical complications with need for intervention occurred in 17.1% and peritonitis in 14.2%. Patients and technique survival in the first 180 days were higher than 80%, and the impact on the growing of chronic PD program was higher than 80% in only 360 days [24].
A Chinese study evaluated 178 patients (96 in unplanned PD and 82 in unplanned HD) and showed higher early mechanical complications (first 30 days) in the HD group (24.4 vs. 5.2%; p < 0.001). There was no difference in infectious complications (2.1% in PD group and 11% in HD group; p > 0.05). At multivariate analysis, urgent-start HD was an independent predictor of dialysis-associated mechanical complications in the short-term (OR 5.02 [95% CI 1.76–14.34]; p = 0.003) [25].
Another Chinese study that included 922 patients treated with urgent-start PD through manual exchanges evaluated mechanical complications in the long term. Only 4.8% of patients developed mechanical complications related to PD after follow-up time of 5.2 months: 18 inguinal hernias, 6 umbilical hernias, 11 patients with hydrothorax, and 6 with hydrocele, 2 episodes of dialysate leak to the subcutaneous and 1 case of pericatheter leak. Twenty one of the 44 patients needed to change dialysis method. Male sex and previous abdominal were independent risk factors for mechanical complications surgery (hazard ratio [HR] 5.41 [95% CI 2.15–13.59]; p < 0.001; HR 2.34 [95% CI 1.04–5.26]; p = 0.04). Evaluating 189 patients included after 2010, 9.5% presented mechanical catheter dysfunction: 13 patients had fibrin obstruction, 4 patients had catheter tip migration, and 1 patient had omentum obstruction. Three of the 4 translocated catheters required surgical repositioning, and the patient with omentum obstruction was definitely transferred to HD. Higher age was a protective factor for catheter dysfunction (HR 0.95 [95% CI 0.91–0.98], p = 0.005) [26].
Javaid et al. [5] highlighted the importance of infrastructure and protocols in the success of the urgent-start PD program, besides the involvement of a multiprofessional team committed to the approach. In Singapore, the strategy of urgent-start PD has started in July/2015, including 17 patients. These patients were compared to 33 patients that initiated planned PD at the same period. Despite the urgent-start group initiating therapy with higher creatinine and urea levels, there was no difference between the 2 groups in hospitalization (7.3 vs. 7.29 episodes/patients/month, p = 0.99), mechanical complications (14 vs. 15%, p = 1), and patients survival after 180 days (88 vs. 94%, p = 0.59).
Table 1 summarizes the main publications of the last decade about urgent-start PD.
Conclusion
PD is a safe, efficient, and cost-effective alternative to HD for patients with urgent indication of chronic dialysis, and urgent-start PD has comparable short-term outcomes to urgent-start HD and planned PD. In particular, as compared to urgent-start HD, urgent-start PD has significantly fewer incidences of catheter-related bacteremia, mechanical complications, and catheter re-insertions during the initial phase of treatment. The approach of urgent-start chronic PD should be widespread in the nephrological community and offered to all patients that require RRT in urgent situations, with no contraindication of the method.
Acknowledgments
We thank nurses Laudilene Marinho and Edwa Bucuvic and doctors Pasqual Barretti and Vanessa B Banin for their contribuition.
Statement of Ethics
We confirm that all methods were carried out are in accordance with relevant guidelines and regulations from Botucatu Medical School.
Disclosure Statement
The authors declare that there are no competing financial and nonfinancial interests.
Funding Sources
There were no funding sources.
Author Contributions
D.B.D. and M.L.M. revised the papers. D.B.D. and D.P. wrote the main manuscript text. D.P. and J.T.C. reviewed the final version of the paper.
PubMed ID
