IMPE Abstracts

Introduction: GH peak after GH stimulatory tests (GHST) diagnoses GH deficiency (GHD).

Aims: To identify GH peak cutoff best discriminating GHD and non-GHD short stature (SS) patients.

Methods: 901 children (0-18 yrs) were submitted to 1348 GHST between 2001 and 2016. We selected 458 patients suspected to have isolated SS (ISS= isolated GHD, familial SS, idiopathic SS) or GHD+ hypopituitarism (MPHD). In those with known adult height (AH), clinical and biochemical data were compared (Wilcoxon test). ROC curve was used to find discriminating GH peak value in the ISS group, considering true GHD those who underwent rhGH treatment, reached adult height (AH) ≥ -2 SDS and parental corrected adult height (AH – target height [TH]) ≥-1.6 SDS. Data are presented as median and interquartile range.

Results: Known AH was found in 195 out of 458 (67% male; 328 GHST). At admission: age= 10.6 yrs [7-13.4]; height (H)= -2.8 SDS (-3.4 to -2.3); H-TH= -1.6 SDS (-2.3 to -0.8), height velocity (HV)= -0.5 SDS (-1.4 to 0.9), plasma IGF1= -1.3 SDS (-1.9 to -0.5). Maximum GH peak was 5.5 ng/mL (1.4-10) and 116 patients (59%) performed two or three GHST. Forty-five patients (23%) presented MPHD and 150 (77%) were classified as ISS. At admission, MPHD patients had lower H (-3.4 vs -2.7 SDS; P=0.02), H-HT (-2.2 vs -1.4; P=0.03), IGF1 (-2.0 vs. -1.1 SDS; P<0.01) and GH peak (0.5 vs. 7.7 ng/mL; P<0.01). Ninety patients (46%) received rhGH until reaching AH (MPHD=25 and ISS=65). Therapy decision was based on clinical judgment, not only on GHST results. Treated patients had lower H (-2.9 vs -2.7 SDS; P=0.03), H-TH (-1.8 vs -1.3 SDS; P=0.02), HV (-0.7 vs -0.4 SDS; P=0.048), IGF1 (-1.5 vs -0.8 SDS; P<0.01), GH peak (3.5 vs 10 ng/mL; P<0.01) and achieved higher AH (-1.8 vs. -2.3 SDS; P<0.01). True GHD was identified in 37 ISS (25%), 50 patients couldn’t be classified as true or non-GHD. At admission, no clinical or plasma IGF1 differences were found between GHD and non-GHD ISS patients. Nevertheless, GH peak was lower in GHD ISS patients (4.2 [1-7.6] vs 9.5 [5.4-14.2] ng/mL; P=0.0004). ROC analysis showed that the best GH peak cutoff for ISS patients was 6.9 ng/mL (AUC=0.74; sensitivity=75%; specificity=72,4%; LR+=2.7).

Conclusions: In ISS patients, the most discriminative GH peak cutoff to identify GHD is 6.9 ng/mL. However, GHST's diagnostic performance is poor.