In our study we analysed the prevalence of primary (18.2 %) and secondary hyperparathyroidism (47.7 %) in patients with PA and we compared it with its prevalence in EH patients. We found that all the cases of primary hyperparathyroidism in patients with PA were normocalcemic forms. We did nor found differences in the cardiometabolic profile of patients with primary hyperparathyroidism and without in patients with PA.
The prevalence of primary hyperparathyroidism in our series was of 18.2%. This is higher than the reported in previous studies, that described prevalences that varied between 1.2% and 2.6% [11][12]. However, it is difficult to know the real prevalence as most of the series were limited to the report of single cases, and the largest series reported included 503 cases of PA, but only 141 patients prospectively included [11]. Nevertheless, the association of PA and primary hyperparathyroidism does not seem to be explained only by chance, as several studies [4, 8, 13, 20] have defended the existence of a bi-directional functional link between the adrenocortical zona glomerulosa and the parathyroid gland. In this way, Rossi et al. [22] have demonstrated the expression of the mineralocorticoid receptor in parathyroid adenoma and in normal parathyroid gland tissue, suggesting a role of aldosterone in the regulation of iPTH synthesis and secretion.
We found a prevalence of secondary hyperparathyroidism of 47.7% in patients with PA. This prevalence was similar that the observed in the EH cohort, and no differences in 25OH-vitamin D were found between groups. However, several previous studies have observed higher prevalence of secondary hyperparathyroidism in patients with PA than EH patients [11, 23–26]. In this line, Petramala et al. [23] found higher plasma iPTH values in PA patients than in essential hypertensive (EH) and healthy people (𝑃<0.001 and 𝑃<0.001, respectively), but no differences in serum calcium levels; and Maniero et al. [24] described plasma iPTH levels 31% greater in PA than in EH patients despite comparable urinary calcium excretion and 25OH-vitamin D insufficiency/deficiency. The finding of higher iPTH levels in PA than in EH independently of serum calcium levels was also confirmed by a recent meta-analysis [25]. We cannot exclude that the differences between the findings of our studies and previous reports were related with a selection bias as we only have included patients with evaluation of complete phospho-calcium metabolism that could explained the high prevalence of hyperparathyroidism in both groups and may also reduce the expected differences between groups. Nevertheless, we did not find any correlation between iPTH and PAC levels neither in PAC between hyperparathyroidism patients, supporting that no differences are expected in PTH levels between PA and EH patients. Other authors have suggested that aldosterone has a stimulatory effect in iPTH [27].
After adrenalectomy vitamin D improved in patients with PA and iPTH levels tended to decrease. In this line, previous studies described the correction of hyperaldosteronism after adrenalectomy in patients with aldosterone producing adenoma (APA), and it was associated with an increased in ionized calcium (Ca2 +) and decreased serum iPTH [13][22]. In this way, although the real mechanisms underlying the increase in serum iPTH in PA patients remain to be elucidated, previous studies found that the administration of aldosterone and 1% NaCl treatment for 4-6 weeks to normotensive rats increased iPTH, probably by causing hypercalciuria and thus lowering serum levels of Ca2 + [28]. This outcome can also occur in human PA because of the hyperfiltration-induced calciuretic effect of hyperaldosteronism, which may impact on bone health [28, 29]. Therefore, the elevated iPTH secretion would represent a compensatory mechanism to the hypercalciuria associated with PA [28][30], that can be reversed after adrenalectomy. Moreover, the interactions between adrenal and parathyroid glands showed differences between patients with upregulated aldosterone production, as in PA, and those with EH and adrenalectomized patients with APA cured of the hyperaldosteronism. In this context, in patients with PA, the acute inhibition of angiotensin II with captopril had no effect on the overtly elevated iPTH levels [27]. In contrast, captopril lowered serum iPTH in patients without hyperaldosteronism, including patients with EH and patients with APA biochemically cured by adrenalectomy [27]. These results suggest that the parathyroid gland probably loses its ability to respond to acute angiotensin II inhibition when hyperaldosteronism coexists with increased iPTH secretion [24][22]. These findings are not surprising as angiotensin II is low or suppressed in PA patients due to the indetectable levels of renin, so it is expected that angiotensin II inhibition has no effect in patients with PA. However, we could not found an explanation for the observed vitamin D increased after adrenalectomy in our patients, it could be explained by chance as the limited sample size of the adrenalectomy group, so we consider that further studies are needed to elucidate this finding.
We did not observe differences in the cardiometabolic profile of patients with PA regardless of the presence of associated hyperparathyroidism. The increased cardiovascular morbidity and mortality in PA is well established [2]. On the other hand, increased PTH concentration has been shown to represent an independent risk factor for cardiovascular events and cardiovascular mortality with a significant reduction in the associated risk one year following parathyroidectomy [31]. However, there are very few data to evaluate whether the coexistence of both PA and hyperparathyroidism as opposed to PA alone confers an increased cardiovascular morbidity and mortality risk. To our knowledge, only the data from the German Conn´s Registry has already addressed this association. Asbach et al. found that there was a non-significant trend towards higher cardiovascular morbidity in patients with PA and secondary hyperparathyroidism. There was a 31.2% (n = 77) cumulative cardiovascular morbidity in patients with PA and secondary hyperparathyroidism vs 21.3% (n = 61) in patients without hyperparathyroidism. The increased cardiovascular risk was mainly driven by the increased prevalence in atrial fibrillation (7.8 vs 4.9%), acute coronary syndrome (3.9 vs 0%) and congestive cardiac failure (11.7 vs 6.6%) [11]. This is no surprise, as increased iPTH secretion has been demonstrated in patients with PA possibly due to aldosterone-mediated tubular calcium and magnesium losses [29], with higher iPTH levels in those with the most severe phenotypes, i.e those with APA [11].
The synergistic effect of aldosterone and iPTH increasing the cardiovascular risk was investigated by Tomaschitz et al in a cross-sectional analysis in over 3000 patients referred to coronary angiography in a tertiary center in Germany. Using multivariate Cox proportional hazard analysis, they found that both plasma aldosterone concentration and iPTH were independently associated with cardiovascular mortality, with a synergistic interaction (P= 0.028). Indeed, plasma aldosterone concentration and iPTH were associated with cardiovascular mortality only in patients in whom both hormones were above the median [32].
There are several limitations in the present study. This is a single-centre, retrospective study, so the information was not always available for comparison. Moreover, the lack of age and renal function matching may play a role as biologically relevant confounding factors. The information about urine calcium excretion was missing in large number of patients. This information could have been interesting as previous studies found increased levels in PA patients compared with EH patients [33][23]. We did not find differences in the prevalence of cardiovascular disease in patients with PA and hyperparathyroidism compared to patients with only PA, although our relatively small sample size might have precluded to find statistically significant differences representing a type 2 statistical error. The same interpretation would apply to the associated risk for chronic kidney disease. Because the catheterization of the adrenal veins was not successful in a majority proportion of patients during the adrenal venous sampling, we could not test whether there were differences in the associated risk with patients with APA and those with idiopathic adrenal hyperplasia. Another important aspect that should be investigated is the co-secretion of cortisol in patients with aldosteronomas, since the detrimental effects of excess aldosterone on bone and the alterations of phospo-calcium metabolism might be associated, or at least worsened by the co-secretion of cortisol [34][35]. Furthermore, it will be useful to analyse the impact of this biochemical alterations on bone health.