Editor - Do we think that we can do better for our dialysis patient's? Dr. Theodore Steinman provides his thoughts....
Dr. Steinman is Clinical Professor of Medicine at Harvard Medical School and Senior Physician in the Renal Division at the Beth Israel Deaconess Medical Center (BIDMC) and the Brigham and Women's Hospital, Boston, Massachusetts. Dr. Steinman has held several senior positions at the BIDMC, including director of the dialysis unit, medical coordinator of the kidney transplantation service, and co-director of the metabolic stone unit.
During his career, Dr. Steinman has had varied research interests, including polycystic kidney disease and kidney progression. He has served in senior positions in several organizations, including vice-chairman of the Polycystic Kidney Research Foundation, member of the Life Options Rehabilitation Advisory Council, and member of the oversight committee of the Joint ASN-NKF-RPA Project on Dialysis Outcome Quality Initiative. He is past president of the National Kidney Foundation (Region I), and past president of the Renal Physicians Association. He has received numerous awards and prizes, including in 2010, the National Kidney Foundation Outstanding Physician award for his dedication and commitment to his patients, the kidney community, and the NKF Serving New England.
COMMENTARY: We Can Do Better - Dr. Theodore Steinman
Mortality for the ESRD population maintained on hemodialysis (HD) remains in the 20% range, with minimal improvement in the past two decades. For the incident dialysis population the all-cause mortality can be as high as 40% in the first 120 days of dialysis. Examination of factors contributing to our dismal outcomes, as compared to other industrialized nations, leads to the conclusion that our problem are related to a multiplicity of issues including: 1) access problem; 2) inadequate treatment time, often not in the best setting; 3) nutrition issues; 4) inadequate attention paid to the extracellular fluid (ECF) volume; 5) coordination of care concerns, especially around the time of initiation of dialysis and inadequate transition following hospitalization back to the outpatient dialysis unit; 6) complications from ultrafiltration; 7) too early start of chronic HD, based on GFR alone; 8) consequences of less than optimal CKD care.
Central venous catheters (CVCs) for dialysis access is the major issue contributing to excessive mortality. 82% of the incident dialysis population initiate HD with a CVC (both temporary and permanent). Care by a nephrologist for 12 months prior to the start of dialysis is still associated with abysmal results of 75% starting HD with a CVC. CVCs are associated with a 3-4 fold greater risk for hospitalization as compared to grafts and fistulas. High mortality rates wih CVCs are related to an increased rate of thrombosis and attendant complications, septicemia, and the need for repeated surgical revisions.
Fistulas need to be created at a much earlier stage of CKD and it is suggested that this be done at a GFR of approximately 20 cc/min. This will allow for adequate maturation of the access. There is a lack of appreciation of starting with an A-V graft placement and later converting it to an upper arm fistula after vessel dilatation.
In addition, the nephrology community has underused peritoneal dialysis (PD) as a viable option. We must do a better job with pre-ESRD education. Our nephrology training programs are failing our fellows with regards to adequate training in every aspect of PD.
Treatment time on dialysis is inadequate in most places around the country. For every 15 minutes less than 4 hours of HD, the mortality rate increases by about 11%. Therefore, the frequently accepted 3.5 hr. dialysis time prescription is assigning the patient to a greater than 20% mortality risk as compared to a 4 hr. treatment. That is simply an unacceptable level of care.
Over the years we have incorrectly focused on urea clearance (Kt/V) as the primary outcome measure and ignored treatment time as a factor in mortality. Middle and large molecule clearance is more influenced by time when compared to that which occurs with the use of high flux membranes.
With time as a major issue, the needed push to change our dialysis delivery to include frequent dialysis approaches is raised. Both in-center and at home, we need to better utilize short-daily and slow nocturnal dialysis. The positive outcomes associated with the Frequent Hemodialysis Network trials and other observational studies are sufficient to promote greater use of these modalities.
Starting dialysis based on GFR alone (at 10-15 cc/min) is noted in several studies to be associated with higher mortality as compared to initiation when uremic symptoms are the deciding factor.
Coordination of care, especially during the first 4 months of HD, is associated with better outcomes. For example, making sure there is timely placement of a fistula or graft (a secondary choice), arranging necessary consultations and transportation, nutrition counseling, dealing with patient depression and providing family emotional support are but a few illustrations of what can be accomplished early in the course of dialysis initiation.
In the area of nutrition we have concentrated too much on a multitude of dietary restrictions. Hence, we have created a population on dialysis who are frequently malnourished. A change is needed - the four new rules of dietary management should be: eat, eat, eat, no salt. Adjust the dialysis prescription on an individual basis to account for the increased nutrient intake. If the emphasis is placed on sodium restriction, the issues related to volume overload and its consequences may be ameliorated.
Left venticular hypertrophy (LVH) is documented to be present in 80-90% of the ESRD population at the time they start dialysis. We have not done an adequate job during the chronic kidney disease (CKD) transition period to address this problem. The consequences of LVH in the ESRD group are highlighted by the fact that sudden death and diastolic dysfunction, resulting in severe congestive heart failure, are the major causes for cardiovascular (CV) mortality (accounting for >60% of the CV mortality). In contrast, coronary artery disease (CAD) is less of a mortality in the ESRD cohort as compared to the general population. There is a need to focus on ECF volume control, both during the CKD and ESRD periods.
When we attempt to achieve ultrafiltration (UF) in a shortened treatment time we run the risk of cardic "stunning". UF rates of >10 ml/kg/hr are also associated with a greater number of adverse events (i.e., hypotension, cramps, etc).
Finally, the decision to start dialysis should be based on the hypothesis that the patient will derive overall benefit from renal replacement therapy. Sadly, this is not the case. Initiating dialysis is not the sine quo non of continuing treatment when the patient is not aware of their environment nor participating in activities of daily living. Time limited trials on dialysis should become the standard approach if the physician is not sure that renal replacement therapy is providing benefit. The nurses are the ones who will tell you the individuals who should be transitioned to conservative/palliative care.
