The issue of whether hypopituitarism is a risk factor for increased mortality from all causes, and particularly from vascular disease, has been evaluated in four retrospective analyses. In all studies, all-cause mortality was significantly increased by 1.3-2.2-fold compared to age- and sex-matched cohorts. The increase in vascular mortality was significant in the two studies from Sweden (1.8-1.9-fold) but not in the two from the UK (0.8-1.4-fold). The total number of deaths relating to all-cause and vascular mortality was relatively small, and the 95% confidence intervals on the observed to expected (O/E) mortality ratios were wide, in some instances crossing the unity line. As retrospective epidemiological studies, these trials suffer from several drawbacks. They included small numbers of patients and events (i.e. deaths), and determination of the number of deaths and their causes may be incomplete, with small additions or subtractions impacting on O/E mortality ratios. A precise diagnosis of pituitary hormone deficiency in patients recruited in the earlier years of the studies (pre-1965) often relied on historical details and not on firm biochemical values. None of the studies provided any measurement of growth hormone status for the cohorts, and in only one study was there any relationship between a particular axis deficiency and all-cause mortality. This finding suggested that hypogonadal subjects were relatively protected compared with eugonadal counterparts. However, this must be regarded as a preliminary conclusion. These retrospective studies indicate that there are no data to support a relationship between vascular or all-cause mortality in hypopituitarism and growth hormone deficiency. Nevertheless, they do provide preliminary evidence to suggest that it would be worthwhile to conduct a large, multicentre, prospective study of well-characterized and documented patients with a sufficient number of events, in order to establish definitively the relationship between various axis deficiencies and mortality.
Read full abstract