Changing Patterns of Peritoneal Dialysis Utilization in the United States

Abstract

The proportion of incident U.S. patients with end-stage renal disease starting chronic peritoneal dialysis (CPD) has historically been low. The low take-on for CPD in the United States is likely multifactorial, but limited physician training and inadequate pre-dialysis patient education appear to be particularly important. Furthermore, two key changes have occurred in the United States: a steep decline in CPD take-on and a progressive increase in the use of automated peritoneal dialysis (APD). The decline in CPD take-on has affected virtually every subgroup examined and has occurred, paradoxically, when the CPD outcomes in the country have improved. Understanding the reasons for historically low CPD take-on and recent steep declines in utilization may allow for plans to reverse these trends to be developed.

Keywords

End-stage renal disease chronic kidney disease hemodialysis initiation of dialysis

The first demonstration that peritoneal dialysis (PD) could be used as a long-term therapy in the management of end-stage renal disease (ESRD) occurred in the United States with the description of continuous ambulatory peritoneal dialysis (CAPD) by Moncrief and Popovich (1). Today, the United States has the second-largest population of PD patients in the world—25,765 in 2004 (2). However, some significant changes have occurred in the pattern of utilization of PD in our country, and herein an attempt to understand the causes and consequences of this changing pattern will be made. Three changes will be discussed: the low background rate of PD utilization; the rapidly declining utilization, particularly during the last decade; and the increasing utilization of automated PD (APD).

Low Background Rates of PD Utilization

The proportional utilization of PD by incident dialysis patients peaked in 1985, but still only 16% of new patients were treated with the therapy. Similarly, the largest proportion of prevalent dialysis patients ever to undergo PD in the United States was seen in 1993, and even at that peak, only 15% of dialysis patients utilized the therapy (3).

It is important to understand the reasons for this low background utilization.

In a survey of directors of training programs in the United States and Canada, a few key findings emerged. As compared with the situation in Canada, the number of patients treated with PD in training institutions in the United States is rather small, and programs assign very small amounts of time for trainees to learn how to provide care to patients undergoing PD (4). As a result, most practicing nephrologists in the United States have received only limited formal training in providing care to PD patients.

The deficit in physician education is compounded by the fact that two thirds of the U.S. patients starting renal replacement therapy are not offered the choice of PD (5). Inability to offer PD is not related to the presence of any medical condition that could be considered a contraindication to the therapy. Even though the probability of offering PD increases with increasing duration of pre-dialysis care by nephrologists, only 46% of patients followed for more than a year before the need for first dialysis reported that PD was discussed as a method of treatment (5). Furthermore, on multivariate analyses, patient education variables (PD being offered and time spent discussing treatment options) were the only significant determinants of starting treatment with PD (5). Thus, the association of demographic variables such as age, race, and employment status with PD take-on rates may be a result of these patients being more likely to be offered PD.

A bi-directional relationship probably exists between deficits in patient and physician education, and together, these deficits create a vicious loop that has led PD to be a niche and probably underutilized therapy in the United States.

Steep Decline in PD Utilization over Last Decade

Since the peak in 1993, an inexorable decline in PD utilization has occurred such that only 6.8% of dialysis patients were on PD in 2004 (3). Even though the rate of decline has slowed, the proportion of dialysis patients undergoing PD in the United States is continuing to decrease (2). Many theories have been forwarded to explain this continuing decrease.

Analyses of the U.S. Renal Data System data from 1996 – 2003 indicate that the decrease in PD take-on rate has occurred in virtually every subgroup examined (6). The decrease has been particularly pronounced among men, white patients (relative to black and Asian patients), retired or unemployed patients (as compared with employed patients), and patients with diabetes, pre-existing cardiovascular comorbidities, or low body mass index, suggesting that PD take-on rates decreased in the sicker subgroup of patients (6). Yet, on multivariate analyses, the decrease in PD utilization over time has occurred largely independently of any change in age, body size, or co-existing disease in the incident ESRD population (6). This finding indicates something far more pervasive in the delivery of care. Future studies need to determine the nature of this elusive factor if the trend of declining utilization is to be reversed.

Examination of crude mortality rates indicate that mortality among PD patients has improved dramatically over the last decade (6). This improvement occurred at a time when the cohort of PD patients in the United States became healthier. Multivariate analyses indicate that most of the improvement is a result of the above-noted change in the profile of patients undergoing PD. However, the probability of technique failure (death or transfer to hemodialysis) within the first year of therapy has decreased independently of the demographic variables (6). Thus, the steepest declines in PD take-on rates have occurred at a time when outcomes have gradually improved.

Increasing Use of Automated Peritoneal Dialysis

The third important pattern that has changed in the United States is the utilization of APD, which gradually increased such that in 2004, 58% of PD patients were undergoing APD (2). Yet very little information exists to show how APD differs from CAPD, if at all.

Two concerns have been raised about the use of APD: first, some studies have shown that APD patients may lose residual renal function more rapidly than CAPD patients do. Second, short, rapid cycling at night may lead to removal of water without sodium. That circumstance has the potential to be associated with greater volume overload and worsened hypertension among APD patients. However, that concern remains unproved.

In our studies, we evaluated whether APD may have implications for the nutrition status of patients. Even though day dwells accounted for 27% of daily dialysate volume, they accounted for 40% of daily protein and amino-acid losses in APD patients (2). Three significant predictors of daily protein loss were frequency of nighttime cycling, duration of dwell, and peritoneal transport rate. Dialysate protein and amino acid losses constituted 24% and 3% of dialysate nitrogen and 14% and 2% of dietary nitrogen intake respectively (2).

The foregoing findings have two important implications. First, rapid nighttime cycling exaggerates peritoneal protein losses and should therefore be avoided. Second, dialysate protein losses may account for up to one third of the dietary protein requirement for CPD patients (as compared with healthy adults).

Conclusions

Significant chaanges have occurred in the patterns of PD utilization in the United States. Understanding the causes and consequences of these changing patterns is important in modifying the patterns and improving outcomes.

References

1. Popovich

Moncrief

Nolph

Ghods

Twardowski

Pyle

. Continuous ambulatory peritoneal dialysis. Ann Intern Med 1978; 88:449–56.

2.United States Department of Health and Human Services, Public Health Service, National Institutes of Health, National Institute of Diabetes and Digestive and Kidney Diseases. USRDS 2006 annual data report. Bethesda: United States Renal Data System; 2006.

3. Mehrotra

Peritoneal dialysis penetration in the United States: march toward the fringes?

Perit Dial Int 2006; 26: 419–22.

4. Mehrotra

Blake

Berman

Nolph

. An analysis of dialysis training in the United States and Canada. Am J Kidney Dis 2002; 40:152–60.

5. Mehrotra

Marsh

Vonesh

Peters

Nissenson

Patient education and access of ESRD patients to renal replacement therapies beyond in-center hemodialysis. Kidney Int 2005; 68:378–90.

6. Mehrotra

Kermah

Fried

Kalantar–Zadeh

Norris

Nissenson

Declining utilization and improving outcomes of chronic peritoneal dialysis (CPD) in U.S. [Abstract]. J Am Soc Nephrol 2006; 17:101A.