Coronary risk in growth hormone deficient hypopituitary adults: increased predicted risk is due largely to lipid...

Hypopituitarism in adults is associated with increased vascular mortality, which has been attributed to GH deficiency.

To compare the lipid profile and coronary risk predicted by the Framingham Heart Study equation in GH-deficient hypopituitary patients and healthy age and gender-matched controls.

A cross-sectional observational study.

We studied 50 adult-onset growth hormone deficient hypopituitary patients (23F, 27M), on appropriate conventional hormone replacement and 45 controls (22F, 23M) matched for age, gender and smoking habit. The subjects (age range 30-75 years) were free from diabetes, hypertension, ischaemic heart disease (IHD) and peripheral vascular disease. All hypogonadal male patients were on testosterone replacement therapy. A similar proportion of female patients (8/23) and controls (7/22) were on HRT. Body mass index (BMI), waist-hip ratio (WHR) and blood pressure were recorded. After an overnight fast blood glucose, total-cholesterol, triglycerides, HDL-cholesterol, apolipoproteins A-I, B and Lp (a) were measured. Coronary risk was calculated for each individual from age, gender, systolic blood pressure, total and HDL cholesterol, smoking habit and presence of diabetes and left ventricular hypertrophy using the Framingham equation.

BMI and WHR were significantly increased in GHD hypopituitary adults of both sexes, but to a greater extent in females. Triglycerides were elevated in both sexes. Total and LDL-cholesterol were increased in both sexes (significantly only in males), and HDL cholesterol and apo A-I were lower (significantly only in females). The reduction in HDL cholesterol was correlated negatively with adiposity (BMI), particularly when centrally distributed (WHR) in patients and controls. LDL cholesterol did not correlate to adiposity but higher levels were present in GH-deficient subjects. The total to HDL cholesterol ratio was significantly increased in patients of both genders (P = 0.002). There were no differences in the apolipoproteins B and Lp(a) between patients and controls. Absolute risk (mean +/- SEM) of a fatal or non-fatal coronary event during the next 5 years was significantly greater in GHD hypopituitary patients than control subjects (4.82 +/- 0.73% vs. 2.94 +/- 0.53, P = 0.04). Cardiovascular risk relative to the local population (RR) was significantly higher in GHD hypopituitary adults (RR = 1.43 CL 1.06-1.80, P = 0.011) but not in the control group (1.08 CL 0.59-1.6). When divided by gender, RR for male patients was not increased (1.14 CL 0.83-1.45, P = 0.096). However, female patients had significantly higher RR (1.7 CL 1.05-2.5, P = 0.048). The RR for male and female controls was not different from the local population.

Changes in lipid levels help to explain the results from risk factor modelling which show increased coronary risk in growth hormone deficient hypopituitary patients, particularly females. The abnormal lipid profile is characterized in both genders by an increase in the total to HDL ratio [corrected], an important parameter in the Framingham equation. The lipid abnormalities conferring increased risk is related to growth hormone deficiency either directly (LDL) or indirectly through increased central obesity (HDL) [corrected]. Adverse calculated coronary risk might provide a new objective indication for consideration of GH replacement therapy in adults.