Insulin-Like Growth Factor I Correlates with Protein Intake Estimated from the Normalized Protein Catabolic Rate in Hemodialysis Patients

Abstract

Background/Aim: Malnutrition and catabolism are predominant problems in patients undergoing hemodialysis. The aim of this study was to clarify the relationship between insulin-like growth factor I (IGF-I), the serum levels of which are influenced by nutrition and which by itself promotes amino acid uptake, and insulin-like growth factor binding protein 1 (IGFBP-1), known to regulate serum (s) IGF-I and protein intake, in end-stage renal disease patients. Methods: Thirty hemodialysis patients were studied, and s-IGF-I and s-IGFBP-1 levels were measured by radioimmunoassay. The s-IGF-I method used was validated according to a reference method. The s-IGF-I standard deviation (SD) score was calculated, giving the individual deviation from the mean of a reference population. The protein intake was estimated both directly by 3-day food recall by a dietician and indirectly by normalized protein catabolic rate (PCRn). Results: The mean serum IGF-I level was 166 ± 10 μg/l, corresponding to a normal s-IGF-I SD score (0.5 ± 0.3). S-IGFBP-1 was elevated threefold to 101 ± 11 μg/l as compared with normal subjects. The s-albumin was 39.9 ± 0.5 g/l and the s-bicarbonate 24 ± 0.4 mmol/l. There were significant correlations between s-IGF-I SD score or s-IGF-I (log-transformed) and PCRn (r = 0.37, p < 0.004, and r = 0.41, p < 0.001, respectively). The s-IGF-I/s-IGFBP-1 ratio was also positively correlated with PCRn (r_s = 0.36, p < 0.007, by Spearman’s rank correlation). The s-albumin was inversely correlated with log s-IGFBP-1 (r = –0.38, p < 0.01) and positively with the s-IGF-I/s-IGFBP-1 ratio (r = 0.36, p < 0.007) but not with s-IGF-I (p < 0.13). Serum total cholesterol, triglycerides, and total body fat as percentage of body weight correlated with s-IGF-I (r = 0.47, p < 0.004, r = 0.45, p < 0.01, and r = 0.42, p < 0.004, respectively) as well as with the s-IGF-I SD score. No correlations were seen between s-IGF-I and protein or caloric intake by direct estimates from dietary food recalls. Conclusions: The s-IGF-I and the s-IGF-I/s-IGFBP-1 ratio were correlated with estimates of protein intake of the patients calculated from urea kinetics (PCRn) but not with direct estimates by the dietitian. The s-IGF-I SD score and the ratio s-IGF-I/s-IGFBP-1 might be a tool to monitor anabolic status and to select hemodialysis patients for therapeutic intervention with recombinant human IGF-I and/or recombinant human growth hormone to counteract catabolism.