Comparison of the effects of primary somatostatin analogue therapy and pituitary adenomectomy on survival in patients with acromegaly: a retrospective cohort study

Abstract

Objective: Acromegalic patients have an increased risk of mortality. The objective of this study was to compare the effect of different therapies for acromegaly on mortality. Design and methods: The mortality rate of 438 consecutive acromegalic patients was compared with that of the general population using the standardized mortality ratio (SMR); the effect of different therapies on survival was evaluated using Cox regression analysis. Results: Twenty patients (4.5%) died between 1999 and 2009. Age-and sex-adjusted SMR was 0.70 (95% CI 0.43-1.08). The Cox regression analysis revealed that, in the whole population, both general risk factors (age and physical status) and specific factors for acromegaly (macroadenoma, hypopituitarism and uncontrolled disease) were associated with death. The most compromised patients at diagnosis had a higher mortality rate (PZ0.001), which also occurred in patients with controlled acromegaly. Death occurred in 2.4% (adenomectomy), 2.6% (adenomectomy followed by somatostatin analogue (SSA) therapy) and 11.4% (SSA therapy as the primary therapy) of the patients. The risk of death was higher in patients receiving SSA therapy as the primary therapy (hazard ratio (HR) 5.52, 95% CI 1.06-28.77, PZ0.043) than in all patients submitted to adenomectomy; however, a higher risk of death occurred only in diabetic patients treated with SSAs alone (HR 21.94, 95% CI 1.56-309.04, PZ0.022). Radiotherapy was associated with an increased risk of mortality, which occurred in patients with the more locally advanced disease. Conclusions: Therapies for acromegaly and comorbidities have lowered the risk of mortality to the level of the general population; the effect of SSA therapy alone or that following pituitary adenomectomy was comparable to that of curative neurosurgery on survival in non-diabetic patients; on the contrary, SSA therapy as the primary therapy may be less effective than adenomectomy in reducing mortality rate in diabetic patients.

Could different treatment approaches in acromegaly influence life expectancy? A comparative study between Bulgaria and Campania (Italy)

“…A recent meta-analysis by Holdaway et al (7) has demonstrated that studies with O30% utilization of SSAs or O70% of remission rates reported normal life expectancy in contrast to higher mortality rates in those with lower percentages of SSA administration and remission rates. In line with these findings, the Campania cohort had a normal life expectancy -an SMR of 0.66 (95% CI 0.27-1.36; PZ0.269) in agreement with two recent studies concerning mortality among the Italian population (33,34,35). Owing to funding restrictions in Bulgaria, SSAs and GHRAs were not available in clinical practice until 2008 and none of the patients were under such Table 5 Characteristics of patients in the three treatment groups.…”

“…Some research groups reported even lower GH targets: 2 mg/l by Ayuk et al and a further improvement in mortality when GH was reduced to !1 mg/l (23,26). Indeed, the last GH value (or post-glucose GH (30)) has been ascertained as a strong predictor of mortality by uniand multivariate analyses in various surveys (23,26,27,33,35). In agreement, the last GH value was an independent predictor of mortality in our study as well (Table 4).…”

“…Mortality rates were restored to levels comparable to the general population when IGF1 values were also below the ULN (25,26,30,39). Although the majority of patients with normal IGF1 values had safe GH levels, the independent role of the growth factor for survival was not demonstrated in such a definite manner (23,26,27,33,35). A probable explanation for such controversial results could be the relatively smaller number of patients tested for IGF1, methodological difficulties in IGF1 measurements, as well as the lack of own age-adjusted reference values.…”

“…Cardiovascular and glucose metabolism alterations are known to be more frequent in patients with acromegaly and dependent on disease control (6). This interrelation could be a possible explanation for the observation that comorbidities have lost their significant influence in multivariate analysis where biochemical status has preserved its dominant role (2,33,35,38). In agreement, in our study, hypertension, diabetes mellitus, and dyslipidemia were not independent predictors of mortality in the multivariate analysis.…”

“…Thus far, four studies have presented SMR in irradiated vs nonirradiated patients with acromegaly and all of them have demonstrated significantly increased death rates with respect to radiotherapy, mainly due to higher cerebrovascular death rates (23,24,27,34). Radiotherapy has also been proved to be an independent predictor of mortality (23,24,31,33,35). In our study, both irradiated and nonirradiated patients had a higher all-cause mortality compared with the general population, with SMRs of 2.59 (95% CI 1.53-3.65) and 1.77 (95% CI 1.27-2.28) respectively.…”