HDCN Article Review/Hyperlink

Duranti E, Imperiali P, Sasdelli M

Is hypertension a risk factor in dialysis?

Kidney Int (Suppl. 55) (Jun) 49:S55:S-173 1996

Cardiovascular disease is the most frequent cause of death among dialysis patients. Hypertension is an important cause and modifier of atherosclerotic vascular disease and is also very prevalent in the dialyzing population.

In this retrospective survival analysis of 370 dialysis patients dichotomized as having hypertension, or not, hypertensive patients are identified as either having mean BP values >150/90 or as taking anti-hypertensive medications. We aren't told the proportion of hypertensive patients in each category. No information is provided regarding the population from which study patients were selected. In comparison with North American patients, there is a disproportionately high proportion with interstitial nephritis (28.4%) and relatively few with diabetic nephropathy (6.8%) or nephrosclerosis (12.2%).

Median survival on dialysis for all 370 patients is about 10 years. After adjusting for age, underlying disease, and type of dialysis, hypertension is not a significant predictor of mortality in dialysis patients.

Comment: This paper tries to address what is probably considered a rhetorical question by most nephrologists. It seems obvious that hypertension should accelerate the vascular disease so prevalent in our patients. The findings of Duranti, et al suggest that this, in fact, may not be the case. How can this be? The effect of hypertension may have been "adjusted out" in the analysis. Older diabetic or nephrosclerotic patients are more likely to be hypertensive so adjusting for underlying disease and age may "hide" the effect of hypertension. We might suspect selection bias: we don't know how patients were included or excluded from analysis. Most important, however, is the authors' definition of hypertension. Normotensive patients on antihypertensive meds are considered hypertensive. This classification would tend to bias the analysis against showing a deleterious effect of high blood pressure. In the end, hypertension probably is not the most important cause of mortality in dialysis patients, but this study would have been unlikely to show any effect of high blood pressure from its inception. (The Nephrology Clinical Research Training Group, University of Washington, Seattle, WA)