By Peter Laird, MD
The American dialysis experience especially in the last three decades is at great variance to the outcomes of all of the other industrialized nations with large dialysis programs. Many inside the US dialysis industry squarely place the blame on the patients themselves citing higher levels of diabetes, HTN and elderly patients than other nations. However, extensive studies over the last two decades by DOPPS and other organizations came to a different conclusion. While there are differences in patient demographics, these alone do not entirely explain the differences in outcomes and higher mortality rates. DOPPS concluded that practice variations of vascular access, treatment times and ultrafiltration rates better explained many of these differences in outcomes.
Survival among HD patients in Japan, Europe, and the United States: Data from DOPPS I (1997 through 2001). *Cox proportional hazards model adjusted for age, male gender, black race, coronary artery disease, congestive heart failure, other cardiac disease, left ventricular hypertrophy, cardiomegaly by x-ray, hypertension, cardiovascular disease, peripheral vascular disease, diabetes, lung disease, dyspnea, smoking, cancer, HIV/AIDS, gastrointestinal bleed, peptic ulcer disease, hepatitis B, hepatitis C, neurologic disorder, psychiatric disease, recurrent cellulitis or gangrene, and vision problems. **P < 0.001.
Saran et al found a direct correlation between increased treatment time and a 7% reduction in mortality with every thirty minutes of addition dialysis per session:
Longer treatment time (TT) and slower ultrafiltration rate (UFR) are considered advantageous for hemodialysis (HD) patients. The study included 22,000 HD patients from seven countries in the Dialysis Outcomes and Practice Patterns Study (DOPPS). Logistic regression was used to study predictors of TT > 240 min and UFR > 10 ml/h/kg bodyweight. Cox regression was used for survival analyses. Statistical adjustments were made for patient demographics, comorbidities, dose of dialysis (Kt/V), and body size. Europe and Japan had significantly longer (P < 0.0001) average TT than the US (232 and 244 min vs 211 in DOPPS I; 235 and 240 min vs 221 in DOPPS II). Kt/V increased concomitantly with TT in all three regions with the largest absolute difference observed in Japan. TT > 240 min was independently associated with significantly lower relative risk (RR) of mortality (RR = 0.81; P = 0.0005). Every 30 min longer on HD was associated with a 7% lower RR of mortality (RR = 0.93; P < 0.0001). The RR reduction with longer TT was greatest in Japan. A synergistic interaction occurred between Kt/V and TT (P = 0.007) toward mortality reduction. UFR > 10 ml/h/kg was associated with higher odds of intradialytic hypotension (odds ratio = 1.30; P = 0.045) and a higher risk of mortality (RR = 1.09; P = 0.02). Longer TT and higher Kt/V were independently as well as synergistically associated with lower mortality. Rapid UFR during HD was also associated with higher mortality risk. These results warrant a randomized clinical trial of longer dialysis sessions in thrice-weekly HD.
For over forty years, Belding Scribner and many of the pioneers of dialysis in America advocated changing our practice patterns as the central focus of change in the US dialysis industry. While I applaud the efforts of Gary Peterson, Rich Berkowitz and Denise Eilers in their lobbying efforts to educate congress on the dysfunctional US renal system, focusing on employment as the most important measure of dialysis outcomes may have unintended consequences that could paradoxically damage the ESRD program and its funding.
On the other hand, the ESRD demographics are important indicators of the total rehabilitation potential in this population. In the 1960's and early 1970's, with limited resources, diabetic patients and elderly patients were excluded from treatment. In addition, those that were granted access to dialysis treatment were expected to learn and preform self care dialysis at home much more so than in-center treatments once again due to limited resources. This highly selected ESRD population had a very high rehabilitation potential and most were able to continue working. Many such as Nancy Spaeth raised a family and worked as a nurse and continue to remain very active in dialysis education and advocacy today after being diagnosed with CKD in 1959.
Today, approximately 44% of the incident and prevalent US dialysis patients suffer from diabetes.
Diabetes affects the cardiovascular system from the small to large blood vessels and this is the primary mechanism of disabling conditions leading to strokes, heart attacks, blindness, amputations, chronic skin ulcers and infections as well as numbness and tingling from peripheral neuropathies. It is a systemic disease where early measures of damage to the kidney correlate with a higher risk of stroke and heart attack. Many consider microalbuminuria a direct marker of cardiovascular disease risk in diabetic patients. For diabetic patients with ESRD, many if not most of these patients have serious comorbid conditions that unfortunately are irreversible and difficult to treat. (here)
In addition, the median age of the incident ESRD population is 64.2 and the median age of the prevalent ESRD population is 59.4 . For those considering emplyment as the main focus of lobbying congress to change dialysis outcomes, the fact that we have at least half of the ESRD patients in the US above the working ages limits its application except to a fraction of the current population. The demographics at the time of the origins of the ESRD program in 1973 no longer exist. I believe we have rightly extended coverage to all patients who wish to be treated and should continue to do so. There are certainly renal replacement candidates that will not benefit from dialysis due to excessive burdens of disease but that is an individual question for each patient and their families to decide. I do believe we have a societal duty to prevent death and suffering when we have that ability.
Currently, there are only ~18% of the working age dialysis patients that continue to engage in active employment. (here) One aspect that has not yet been completely elicidated is what are the modifiable factors that could improve employment rates among dialysis patients. Patient characteristics such as educational status and occupation prior to iniating dialysis are fixed. Evaluating employment based on facility characteristics is an area in need of further study. Two variables found by Kutner et al in 2008 were access to evening dialysis sessions as well as facilities that have a home dialysis program. Both factors significantly correlate to higher levels of employment. There are likely many other factors that should be explored.
Taking into consideration the burden of comorbid diseases, we must ask what goal of employment is reasonable? We are currently employing 35-40% of working age dialysis patients. If the goal is set too high in such a system of evaluating dialysis units according to employment rates of eligible patients, those truly that are disabled with irreversible conditions could be forced into a situation where their safety net is removed. Even if we increased employment rates by a factor of 100% (70-80% of working age patients) from current levels which I doubt anyone believes is achievable, the total unemployment rate among dialysis patients would still be near 65%. Focussing on employment alone does not achieve the desired results since the majority of dialysis patients by age or severe disability are not candidates to continue or return to work.
I do believe that we can and should improve all quality of life indicators but do so the same way that Dr. Scribner advocated over forty years ago, by applying physiologic renal replacement strategies best deliniated by Dr. Carl Kjellstrand in his "Unphysiology Hypothesis" in 1975. More frequent and longer duration dialysis improves mortality, reduces hospitalizations, significantly reduces the costs of treatment and conversely also raises the number of patients who are able to return to work. (here) Employment in the end analysis is one indicator of optimal dialysis. Lower mortality, reduced depression, improved sexual function and many other indicators likewise optimize when patients obtain longer duration and more frequent dialysis.
I would hope that we continue in Dr. Scribner's tradition of calling for incentives for home hemodialysis, avoidance of high ultrafiltration rates and longer and more frequent dialysis sessions as all other nations around the world have already practiced for decades. Setting employment quotas gains nothing at all if you don't first fix the underlying problem, our unique American dialysis practices. That has and should continue to remain our focus as dialysis advocates seeking reform. Under employment is the symptom, not the disease.
Designing an entire medical program around one symptom alone is not prudent in my opinion. Designing the system as Dr. Scribner advocated for decades remains our one and true course we should all join together to support. Indeed, there are many indicators that Dr. Scribners life long advocacy was not and is not in vain. His message matters today more than ever as we begin a dialogue among all involved in the care of dialysis patients. Simply breaking down the wall between part A and Part B would drive the financial incentives to avoidance of hospitalizations and complications. In the end, the best medicine once again turns out to be the cheapest medicine.
Utilizing our current resources with improved financial incentives offers much more hope of true dialysis reform than any other method. Interestingly, that is the message that the large dialysis organizations are also lobbying for with CMS/Congress at the present time. I believe that a convergence of the LDO's, CMS/Congress and patient organizations is not only possible, but it will happen. Putting that system together should be the dominant goal of our outreach and advocacy.
LDO's and for-profit medicine are a reality of American medicine whether people are satisfied with that answer or not. Unleashing the properly incentivised goals of dialysis care will bring about the corporate response we wish to see with better outcomes for all patients, not just a select few who are capable of employment. Recent CMS demonstration projects show that this is not only feasible, it is desirable. (here) Interestingly, FMC already demonstrated these same types of outcomes with their joint Kaiser/FMC dialysis partnership in Southern California. I can readily assert that I have personally witnessed the results at my own dialysis unit as both a physician and a patient where my nephrologist was able to reduce annual mortality from over 20% down to only 7% in 2009.
We already have the tools available today to more efficiently utilize our current resources in a better fashion for all patients. Applying these simple concepts to dialysis care was the basis for Dr. Scribner's visionary recommendations. It remains my central core strategy and goal as well.