[Difficulty in assessing renal function in elderly hypertensive patients. Results of a survey in the geriatric environment].

OBJECTIVES

In elderly patients, modifications observed in renal function often make difficult the differentiation between the pathological cause and the physiological consequences of aging. The presence of arterial hypertension can enhance these hemodynamic modifications. The state of renal function is often assessed only on creatinemia, which is not a marker adapted to the elderly, since creatinine levels often remain normal even in cases of altered renal function. The aim of this study was to determine the prevalence of chronic renal failure (CRF) in elderly hypertensive patients, based on the calculation of creatinine clearance using Cockcroft and Gault's method, and comparing it with the prevalence of CRF when renal function is assessed by methods currently used by practitioners.

METHODS

In 841 treated hypertensive patients aged over 65, hospitalized in geriatric wards (median or long term hospitalization), the state of renal function was assessed in two ways: on the one hand by the physicians themselves, according to their method of choice and, on the other, by calculating creatinine clearance according to Cockcroft and Gault's method. These assessments were then compared and the main conflicting factors were revealed by logistics regression. Furthermore, the importance given to the state of renal function in the therapeutic attitude of the physician was determined using a graduated scale from 0 to 5. Logistics regression analysis identified the principle determinants of this attitude.

RESULTS

In this population with a mean age of 81 +/- 8 years, predominantly composed of women (65%), the prevalence of renal failure (clearance calculated according to Cockcroft and Gault's formula < 80 ml/min) was of 89%, and of 42% at the stage of severe or pre-terminal renal failure (clearance < 30 ml/min). Conversely, when assessed by the physicians according to their method of choice, it was only of 64% and 16% respectively. The agreement between calculated and assessed renal function was of 35%. Renal damage is often underestimated by the physicians compared with calculated values (for 61% of patients). Logistics regression revealed the following determinating discrepancies: gender (p = 0.0001), age (p = 0.0025) and body weight (p = 0.0031), which are the principle variables of Cockcroft and Gault's formula. The principle elements taken into account in the assessment of renal function in the therapeutic attitude of the physicians towards these elderly patients are: the existence of an alteration in renal function (p = 0.0001), the level of systolic arterial pressure (p = 0.001), the number of cardiovascular risk factors (p = 0.002), notably diabetes (p = 0.003) and the female gender (p = 0.024).

CONCLUSIONS

This study emphasizes the great prevalence of renal failure in elderly hypertensive patients, with only 11% of hypertensive patients exhibiting normal renal function after the age of 80. Assessment of renal function by the physicians according to their method of choice suggests an underestimation of renal damage compared with the calculated values of creatinine clearance. Availability of simple calculation tools would facilitate the systematic application of Cockcroft and Gault's formula and provide enhanced appreciation of renal function in the elderly.