Effects of long-term growth hormone replacement in adults with growth hormone deficiency following cure of acromegaly: a KIMS analysis.

CONTEXT: GH deficiency (GHD) may occur in adults with cured acromegaly (acroGHD).

OBJECTIVE

Our objective was to examine the effectiveness and safety of GH replacement in acroGHD.

DESIGN

This study was a retrospective analysis of data from KIMS (Pfizer International Metabolic Database).

SETTING

Data were extracted from a pharmaco-epidemiological survey of >16 000 GHD adults from 31 countries.

PATIENTS: The effectiveness population included 115 adults with acroGHD and 142 age-, gender-, and body mass index-matched GHD adults with nonfunctioning pituitary adenoma (NFPA) followed up to 5 years on GH. The safety population included 164 adults with acroGHD and 2469 with NFPA, all GH-replaced. Both acroGHD and NFPA were compared with several cohorts from the general population (including the World Health Organization Global Burden of Disease).

OUTCOME MEASURES: Outcome measures included quality of life (QoL-AGHDA), lipids, serious adverse events, and additional safety endpoints.

RESULTS

Median GH dose was 0.3 mg/d in acroGHD and NFPA at 5 years. There were comparable improvements in QoL-AGHDA and total and low-density lipoprotein cholesterol in acroGHD and NFPA. High-density lipoprotein cholesterol increased only in acroGHD. Cardiovascular mortality was increased in acroGHD vs NFPA (standardized mortality ratio = 3.03, P = .02). All-cause mortality was similar in acroGHD (ratio between observed/expected cases [95% confidence interval] = 1.32 [0.70-2.25]) and lower in NFPA [observed/expected = 0.58 [0.48-0.70]) in comparison with the general population. There was no difference in incidence of all cancers, benign or malignant brain tumors, or diabetes mellitus between acroGHD and NFPA.

CONCLUSIONS

GH replacement has comparable effects on quality of life and lipids in acroGHD and NFPA. Further investigation is needed to examine whether the increased cardiovascular mortality may be attributed to the history of previous GH excess in acroGHD.