Monitoring of treatment success in patients with acromegaly: The value of serum insulin-like growth factor binding protein-3 and serum leptin measurements in comparison to plasma insulin-like growth factor I determination
C. Paramo et al., Monitoring of treatment success in patients with acromegaly: The value of serum insulin-like growth factor binding protein-3 and serum leptin measurements in comparison to plasma insulin-like growth factor I determination, METABOLISM, 50(9), 2001, pp. 1117-1121
The aim of the present study was to determine the value of serum insulin-li
ke growth factor binding protein-3 (IGFBP-3) and serum leptin measurements
in comparison with plasma insulin-like growth factor I (IGF-I) measurements
as indicators of treatment success in patients with acromegaly. Thirty-fiv
e acromegaly patients, 25 female and 10 male, divided into groups of patien
ts with postadenomectomy "active" acromegaly (n = 20) and patients with pos
tadenomectomy "controlled" acromegaly (n = 15), and 44 healthy volunteers s
ex- and age-matched with the acromegaly patients were included in the prese
nt study. We comparatively analyzed plasma IGF-1, serum IGFBP-3, and serum
leptin levels in the aforementioned groups. Because serum leptin has sex di
morphism, the groups were divided into sexes when leptin was evaluated. As
expected, the patients with active acromegaly had significantly higher mean
values of plasma IGF-I and serum IGFBP-3 and lower mean values of serum le
ptin (only in women) than the control group. However, individual evaluation
showed that 1 of 20, 9 of 20, and many patients with postadenomectomy acti
ve acromegaly patients had values that overlapped values of control subject
s for plasma standard deviation score (SDS)-IGF-I, serum SDS-IGFBP-3, and s
ex-ajusted serum leptin, respectively. Application of the receiver operatin
g characteristic (ROC) curves method shows that plasma IGF-I measurement ha
s the best discriminatory power to differentiate patients with postsurgical
active acromegaly from healthy people. Its area under the curve (AUC) was
0.95, with a sensitivity and specificity of 86% and 94%, respectively. Its
positive and negative likelihood ratios were 14 and 0.15. Serum IGFBP-3 has
certain discriminatory power, its AUC being 0.89, with a sensitivity and s
pecificity of 83% and 77%. Its positive and negative likelihood ratios were
3.6 and 0.22. Serum leptin, both in women and in men, has a poor performan
ce with sensitivity and specificity of 53% and 50% for women and 55% and 56
% for men and positive and negative likelihood ratios of 1.06 and 0.94 for
women and 1.26 and 0.8 for men. Application of the ROC curves method and th
e determination of positive and negative likelihood ratios in comparative e
valuation of serum IGFBP-3 and serum leptin with plasma IGF-I as indicators
of treatment success in acromegalic patients showed that neither serum IGF
BP-3 nor serum leptin determinations have accuracy better than or similar t
o that of plasma IGF-1 for monitoring treatment success in acromegaly patie
nts. Serum IGFBP-3 is accurate but does not increase accuracy for age-ajust
ed plasma IGF-1, whereas determination of serum leptin level has no value i
n monitoring these patients. Copyright (C) 2001 by W.B. Saunders Company.