Of Peritoneal Dialysis Access and Home Dialysis Accessibility

Central to the successful initiation and maintenance of peritoneal dialysis (PD) is a functional dialysis access. PD catheter insertion-related complications continue to be an important cause of early and modifiable transfers to hemodialysis, particularly in the first 3 months of therapy.¹

In this issue of JASN, Ku and colleagues use Medicare claims and United States Renal Data System data to explore the association between the specialty of the operator (categorized as vascular surgery, general surgery, interventional radiology, or interventional nephrology) and the risk of requiring a follow-up procedure within 90 days among adult PD-naïve, Medicare-insured individuals whose PD catheter was placed between 2010 and 2019.² Follow-up procedures were defined as PD catheter revision, replacement, or removal; the use of PD at 90 days was also explored. Percutaneous, laparoscopic, and open dissection methods were all included.

There were 46,973 PD catheters inserted by 5205 operators (71.1% general surgeons, 17.2% vascular surgeons, 9.7% interventional radiologists, 2% interventional nephrologists), representing 64% of all patients who received their first PD catheter placement during the study period. Overall, 20.9% were performed via open dissection. Patient characteristics varied by operator type in important ways. Patients undergoing nephrology-based insertion were less likely to be older than 65 years (53.5% versus 60.5%–63.8%) and more likely to be Hispanic (17.8% versus 7.4%–9.5%). Nephrology-based and radiology-based insertions were more likely to have received hemodialysis before PD (60.2%–62.7%) compared with surgically placed PD access (54.5%–57.9%). Among the cohort, 15.5% of patients required a follow-up procedure within 90 days (2.9% PD catheter reimplantation, 6.6% PD catheter removal, and 5.9% revision). A second PD catheter insertion and PD catheter removal were more common for catheters placed by interventional radiologists and nephrologists, while revisions were more common among interventional nephrologists. Catheters inserted by nephrologists and radiologists had higher odds of requiring a PD access follow-up procedure at 90 days compared with surgeons in the main analysis and in several sensitivity analyses.

This study is an important and timely contribution to the literature. The most notable findings were that PD catheters placed by interventional nephrology and radiology resulted in a higher proportion of follow-up procedures, compared with surgically placed catheters, and that only 78.4% of the cohort was receiving PD at 90 days. The very high attrition at 90 days is an important finding and speaks to the overall challenges in initiating and maintaining individuals on PD therapy. Unfortunately, PD access follow-up procedure risks were not separated out on the basis of the method of placement (laparoscopic versus open procedure) nor the use of advanced techniques during laparoscopic placement, which are important to consider. Information about surgical history and prior abdominal surgery was not available, which may have influenced the observed differences between groups. It is also important to note that PD catheter removal may have occurred for reasons unrelated to flow dysfunction, such as a decision to switch to another modality, which may not be related to the method of PD catheter insertion. However, the reasons for catheter removal were not captured in this study.

It is important to note that some individuals will have PD catheter insertion attempts, but never receive PD, so the traditional approach to measuring catheter-related complications from the start of PD likely underestimates the true incidence of early PD catheter complications. A multicenter study in Ontario, Canada, showed that 17% of individuals undergoing PD catheter insertion never received maintenance PD.³ In a preliminary report from the multicenter North American PD Catheter registry, 3% of the 500 individuals who underwent laparoscopic PD catheter insertion never started PD due to an insertion-related complication.⁴ Capturing this early loss is important, particularly from a quality improvement standpoint, as it likely negatively affects patient experience and is costly.

The most recent International Society for PD guidelines for creating and maintaining optimal PD access in the adult patient state that the PD catheter implantation approach should be “based on patient factors, facility resources, and operator expertise.”⁵ However, those very same factors have confounded the interpretation of the many studies that attempted to compare clinical outcomes with the various methods of PD catheter insertion.⁶ Available methods of insertion are surgical: either laparoscopic or by open dissection (typically performed by surgeons [general, transplant, urology, vascular]) and percutaneous: needle-guided with/without adjunct additional imaging, such as fluoroscopy and ultrasound guidance (predominantly performed by nephrologists or radiologists). The advantage of the surgical laparoscopic technique is the ability to perform additional, advanced procedures with the intent of minimizing PD catheter flow dysfunction and complications. This may be particularly advantageous in certain subsets of patients, such as those with adhesions from prior surgery. Advanced procedures include rectus sheath tunneling, omental diversion (omentopexy), lysis of adhesions, deep cuff purse-string suture placement, and concurrent hernia repair, although they are not consistently performed.⁴ The percutaneous option is ultimately a blind insertion method, without direct visualization of the peritoneal cavity. On the other hand, the percutaneous approaches obviate the need for general anesthesia and the resultant risk of postoperative complications, reduce intensive operative resources, and remain an attractive option for frail, comorbid patients, particularly those in need of urgent PD catheter placement.

The relative underperformance of radiology-based and nephrology-placed PD catheters relative to those placed by surgeons in the study by Ku and colleagues is cause for concern and should be interpreted in context and with some caveats. First, this was not a randomized comparison and other observational studies have shown conflicting results. A recent multicenter study in 44 UK centers where percutaneous placement is more common compared with the United States compared the risk of catheter-related events in the first year (defined as a composite of catheter-related infection, hospitalization, and removal) in 325 patients with percutaneously inserted catheters versus 444 patients with surgically inserted PD catheters.⁷ The UK study found no significant difference between the two methods of PD catheter insertion, in an environment where percutaneous placement is more common than it is in the United States. It should be noted that there may have been a lower risk of infection in the percutaneous compared with the surgical group, but a higher risk of catheter malfunction, leading to a similar overall risk of the composite outcome. Second, there may be subgroups of patients who do best with either a surgical or percutaneous approach, but others that do well with either. In that scenario, the availability of percutaneous approaches might offload surgical programs and provide more timely and safe access to PD catheter insertion. This hypothesis needs to be further explored because there are insufficient data to make any definitive claims at this point. Third, operator factors may be important and may have influenced the observed results in the Ku study. Operator training, experience, and volume may all influence outcomes. Only 6.5% of the 46,973 PD catheter insertions in this study were inserted by either radiology or nephrology, and median procedure volumes were very low at three per year for interventional nephrologists and one per year for interventional radiologists. It is reasonable to wonder whether the outcomes may have been better in the hands of higher volume operators. It would also be important to determine whether there are operators who achieve excellent outcomes using percutaneous approaches. If there are, it would suggest that the operator may be more important than the choice of technique. It would be very important to explore these issues, as timely access to PD catheter placement is already limited in many places. Limiting options further could negatively affect the ability of programs to offer the therapy to patients. Finally, the benefits of percutaneous catheter insertion methods became apparent during the coronavirus disease 2019 pandemic. Limited availability of operating rooms and surgeons coupled with regulatory restrictions on what procedures were deemed urgent or elective saw many programs struggle with PD access, particularly those that only had access to surgical placement.⁸

This study has leveraged a large Medicare claims dataset not specifically designed to explore the effect of PD catheter insertion methods on the risk of PD catheter insertion-related complications and not validated for identifying important variables included in the analysis. The North American PD Catheter Registry was developed with the specific goal of improving PD catheter outcomes.⁴ To date, it has captured high-quality, granular, prospective data on more than 2500 catheter insertions at 23 centers in the United States and Canada. The hope is that data from this study will provide further insights into the center-related, provider-related, and patient-related variables that influence catheter function and outcomes. It will also guide the development and standardization of outcome measures to facilitate benchmarking and address some of the outstanding issues articulated above.

Ku and colleagues have made an important contribution to the literature and highlighted an issue that may be limiting growth in PD in the United States. Aligning with the Advancing American Kidney Health Initiative, which seeks to increase home dialysis use for individuals requiring kidney replacement therapy, will require initiatives that improve access to PD catheter insertion while reducing insertion-related complications. These may include further emphasis on training and mentoring for trainees and those already in practice on optimal methods for PD catheter insertion, regardless of method of insertion or operator type, and will require that current training resources to be expanded. The implementation of robust quality metrics for PD catheter placement with a culture of operator accountability for high quality placement will also be an important consideration, while other factors, such as increasing reimbursement for PD access placement, may influence interest and availability.

Disclosures

J. Perl reports Consultancy: AstraZeneca, Baxter Health Care Canada, Bayer, Davita Healthcare Partners, Fresenius Medical Care, LiberDi, Otsuka, and Outset Medical; Ownership Interest: I-Ren; Research Funding: AHRQ and Arbor Research Collaborative For Health; Honoraria: AstraZeneca, Baxter Healthcare USA/Canada, Bayer Canada, Davita Healthcare partners, DCI, Fresenius Medical Care, Innovative Renal Care, United StatesOtsuka, and US Renal Care; Speakers Bureau: Baxter Healthcare and Fresenius Medical Care; and Other Interests or Relationships: Salary Support: AHRQ and Arbor Research Collaborative For Health. I. Teitelbaum reports Consultancy: Triomed; Ownership Interest: i-RenMedical; Advisory or Leadership Role: Associate Editor of Blood Purification and Editorial Board of Peritoneal Dialysis International; and Other Interests or Relationships: Past President of the International Society for Peritoneal Dialysis. The remaining author has nothing to disclose.

Funding

None.

Author Contributions

Conceptualization: Jeffrey Perl, Robert R. Quinn, Isaac Teitelbaum.