Failure of Dietary Protein and Phosphate Restriction to Retard the Rate of Progression of Chronic Renal Failure: A Prospective, Randomized, Controlled Trial

SUMMARY Ninety-five patients (63 male, 32 female), age 45±2 years(mean±SEM) with chronic renal failure of varied aetiologywere randomized to receive either a conventional low proteindiet (0.6 g/kg/day protein, 800 mg phosphate; n=33), a low phosphatediet (providing approximately 1000 mg phosphate plus an orallyadministered phosphate binder, minimum protein intake 0.8 g/kg/day;n=30) or to control (minimum protein intake 0.8 g/kg/day, nophosphate restriction; n=32). Patients were reviewed for a minimumof 6 months before randomization and were withdrawn from thestudy if plasma creatinine exceeded 900 µmol/1, plasmaphosphate was > 2.0 mmol/1 or at the onset of uraemic symptoms. Following randomization patients were studied for an averageof 19±3 months. Mean plasma creatinine rose from 398±33to 600±50 µmol/1. Dietary protein intake was estimatedat 0.69±0.02 g/kg/day in the low protein group, 1.02±0.05in the low phosphate and 1.14±0.05 in the controls, phosphateintake was 815±43, 1000± 47, and 1315±57mg/day, respectively. Urinary urea excretion and protein catabolicrates were significantly reduced (p<0.01) only in those onprotein restriction, at 213±9 mmol/24 hours and 0.71g/kg/day, respectively. Phosphate excretion was significantlylower (p<0.05) in both the low protein group (17.9±0.8mmol/24 hours) and the low phosphate group (18.6±1.0mmol/24 hours) compared to controls. Changes in body weight,muscle mass and serum transferrin, albumin and immunoglobulinswere comparable between the groups. Mean blood pressure followingrandomization was 150/89±3/1 (low protein), 148/87±3/1(low phosphate) and 146/87±3/1 (controls). Progression of renal failure was analysed by rate of fall ofcreatinine clearance (ml/min/ 1.73 m²/month), by rate of deteriorationderived from reciprocal plasma creatinine against time plots(1/mmol/year) and to assess individual patient's response totreatment by two phase linear regression (‘breakpoint’)analysis of reciprocal plasma creatinine/time plots. Progressionwas analysed only in patients seen for at least 3 months followingrandomization. The rate of fall of creatinine clearance was not significantlydifferent between the groups (ANOVA): 0.56±0.08 ml/min/1.73m2/month (low protein, n=28), 0.44±0.07 (low phosphate,n=23) and 0.69±0.11 (control, n=27). In 50 patients (18low protein, 16 low phosphate and 16 control) whose rate ofprogression could be calculated before and after randomization,there was a fall in rate of progression averaging 0.18 ml/min/1.73m2/month in those on low protein diet and those on low phosphatediet, but a rise of 0.08 in the controls. These differenceswere, however, not statistically significant. Similar resultswere obtained when the rates of deterioration were calculatedfrom plasma creatinine. Significant individual improvements(p<0.01) in rates of progression by ‘breakpoint’analysis occurred in 17 patients: six on low protein, sevenon low phosphate and in four controls. Sixty-one (72 per cent)of the patients examined by this method showed no significantchange in the rate of progression while seven patients had acceleratedprogression. There was no difference in the requirement formaintenance dialysis facilities between groups. No significant benefit of protein and phosphate restrictionwas therefore demonstrated.