Our findings indicate that NC-PHPT is not associated with cardiometabolic alterations, which conversely have been displayed in our patients with HC-PHPT. Specifically, in our NC-PHPT patients, we found similar glucose, cholesterol, and systolic blood pressure levels, together with a comparable frequency of glucose alterations, lipid disorders, and hypertension in comparison to controls. On the contrary, as expected, HC-PHPT displayed a higher frequency of cardiometabolic disorders compared to controls. In addition, patients with NC-PHPT showed lower glucose, cholesterol, and systolic blood pressure levels as well as a reduced frequency of glucose and lipid alterations and cardiovascular events than HC-PHPT. Among indices of PHPT activity, calcium but not PTH levels displayed a significant correlation with some cardiometabolic parameters.

It is well known that HC-PHPT patients are characterized by a higher prevalence of cardiometabolic disorders, including glucose intolerance, diabetes, dyslipidemia, and hypertension, leading to an increased cardiovascular morbidity and mortality [6,7,8,9,10,11, 29]. This higher risk connotes HC-PHPT irrespective of its clinical presentation, including asymptomatic patients [16]. On the other hand, few studies investigated the association between cardiometabolic disorders and NC-PHPT, with unclear and controversial results. In line with our findings, Tassone et al. [17] and Cakir et al. [18] reported a similar insulin sensitivity and glucose tolerance between NC-HPHT patients and controls matched for sex, age, and BMI. Conversely, Beysel et al. [19] found higher insulin resistance, increased insulin levels, and HOMA-IR values in the former than in the latter, without any differences between NC-PHPT and HC-PHPT; in addition, in the study by Ozturk et al. [20], NC-PHPT had higher glucose levels than controls despite similar insulin, HOMA-IR, and HbA1c values. Concerning lipid metabolism, Beysel et al. [19] reported higher total cholesterol, LDL cholesterol, and triglycerides levels in NC-PHPT compared to controls, according to previous findings by Hagstrom et al. [21]. However, these results were not confirmed by other studies [18, 20]. In fact, in line with our data, Cakir et al. showed a similar lipid profile between NC-PHPT and controls matched for age, sex, and BMI. In the study by Tordjman et al. [22], NC-PHPT patients displayed similar rates of hypertension when compared to controls, but other previous studies [19, 20] reported higher prevalence of known hypertension and/or increased systolic and diastolic blood pressure levels in the former than in the latter. In our study, despite higher SBP values in HC-PHPT than NC-PHPT and controls, the history of hypertension did not differ between the three groups. Of note, when comparing our results with the few data available from literature, it should be underlined that most of the papers that highlighted a cardiometabolic derangement in NC-PHPT [19,20,21] were affected by a major bias. Specifically, in the studies by Beysel et al. [19] and Ozturk et al. [20], many patients diagnosed as NC-PHPT presented vitamin D deficiency, whose correction is mandatory to ascertain the disease. Therefore, their patients may not be real NC-PHPT and low vitamin D levels could contribute “per se” to cardiometabolic alterations. Moreover, in the studies by Hagstrom et al. [21], the lack of vitamin D measurement does not allow to exclude that their NC-PHPT patients, histologically proven as PHPT, could be “vitamin D—deficient masked” HC-PHPT, leading to a misclassification. Very little information is available about incidence and prevalence of cardiovascular events in patients with NC-PHPT. In line with our findings, Tordjman et al. reported a higher prevalence of cardio- or cerebrovascular disease in HC- than NC-PHPT patients, despite a similar cardiovascular risk profile between the two groups [22].

Furthermore, we investigated the association of serum calcium and PTH levels with cardiometabolic parameters in HC-PHPT, without obtaining statistically significant results. We ascribed this negative finding to both the lack of a wide distribution of serum calcium and PTH values and the small sample size when considering the groups separately. Thus, to widen the range of serum calcium and PTH levels and to increase the sample size, we chose to perform the same analyses in all PHPT, considering that the statistical validity of the association between the above-mentioned variables is maintained. By including the whole sample of PHPT in our analysis, we found a positive correlation of serum calcium with fasting blood glucose and systolic blood pressure, with PTH levels not significantly related with cardio-metabolic parameters. The predominant role of calcium in the pathogenesis of glucose, lipid, and blood pressure disorders in PHPT is still debated. Previous studies in HC-PHPT have not yet clarified which is the main responsible for the metabolic imbalance and the increase of cardiovascular morbidity and mortality between hypercalcemia and the rise of PTH serum levels. It is well-known that hypercalcemia can compromise glucose tissue uptake by inducing peripheric insulin resistance, then leading to glucose intolerance. Tassone et al., investigating the frequency of glucose intolerance in 122 patients with PHPT in comparison to 61 healthy subjects, found a reduced insulin sensitivity in the former and showed a negative independent association of serum Ca with insulin sensitivity [12]. In addition, hypercalcemia may also have a role in favoring blood pressure disorders in PHPT. Accordingly, Hanson et al. [13] showed that the increase of intracellular calcium in vessel wall cells in PHPT leads to vasoconstriction, with a rise of peripheric resistance and blood pressure. Moreover, in line with our data, Letizia et al. [30] showed an independent positive correlation between ionized calcium and blood pressure in 53 patients with primary hyperparathyroidism. Finally, serum calcium levels have also been correlated to an increase of cardiovascular mortality in PHPT [31], in agreement with our findings showing an increased frequency of cardiovascular events only in HC-PHPT. On the other hand, previous studies suggested that also PTH could be involved in glucose metabolism homeostasis, and blood pressure regulation. Animal studies suggest that PTH underregulate the insulin intracellular signaling mediated by IRS-1, with a consequent increase in peripheric resistance to insulin [32], while Chiu et al. [14] found an inverse correlation of PTH with the insulin sensitivity index (ISI) in a group of 52 healthy subjects. Moreover, it is known that prolonged infusion of parathyroid hormone increases blood pressure probably due to an alteration of blood vessels reactive properties [33] and an increase of blood volume through activation of renin-angiotensin-aldosterone system [34] and a positive correlation of PTH levels with systolic blood pressure has been shown [15].

Based on these findings, one may hypothesize that both high PTH levels and hypercalcemia could contribute to the development of cardiometabolic disorders in PHPT. In this regard, NC-PHPT is an interesting pathological model, that allows us to analyze the contribution of the two factors separately. Our data support a predominant role of hypercalcemia, rather than hyperparathyroidism, in favoring cardiometabolic derangement in PHPT. Nevertheless, it has to be noted that our NC-PHPT patients showed less high PTH levels compared to HC-PHPT. Therefore, it may be that the supposed effect of hyperparathyroidism is weaker in the former than in the latter, according to a low cardiometabolic derangement in NC-PHPT.

Finally, one possible explanation for the lack of association between serum ionized calcium and cardiometabolic parameters is that, despite the theoretically greater accuracy of ionized than total calcium in defining the true biologically active calcium, its precision in clinical practice may be often impaired by many analytical and post-analytical interfering factors.

Our study has some limitations. First, its cross-sectional design did not allow the identification of cause-effect relationships and temporal trends. Second, though considering three study groups with patients well-matched for age, sex, and BMI, the strength of the conclusions is limited by the small sample size, that did not allow us to perform a regression analysis to estimate the effect of exposure adjusted for multiple confounders. Conversely, the major strength of our work is represented by the soundness of the diagnosis of NC-PHPT, according to international guidelines and after the exclusion of all potential causes of secondary hyperparathyroidism.

In conclusion, our data show that cardiovascular alterations do not involve NC-PHPT. The positive correlation of serum calcium levels with fasting blood glucose and systolic blood pressure suggests a predominant pathogenetic role of hypercalcemia rather than hyperparathyroidism in the development of cardiometabolic disorders and could account for the absence of such alterations in NC-PHPT.