Primary aldosteronism (PA), a typical form of secondary hypertension, is characterized by autonomous aldosterone hypersecretion. The risk of cardio-renal-vascular events, such as stroke, myocardial infarction, cardiomegaly, heart failure, atrial fibrillation, and renal impairments, is higher in patients with PA than in patients with essential hypertension. In addition, several clinical studies have shown abnormal glucose metabolism in patients with PA. Catena et al. conducted a prospective study in patients with PA. They showed that patients with aldosterone-producing adenoma (APA) and idiopathic hyperplasia (IHA) had insulin resistance, which was rapidly and persistently restored to normal conditions by surgery or aldosterone antagonist treatment [1]. In the German Conn’s Registry, Reincke et al. reported that diabetes mellitus (DM) was more prevalent in patients with PA than in matched controls [2, 3]. In the Japan Primary Aldosteronism Study (JPAS), Akehi et al. also showed a higher DM prevalence in patients with PA than in the general population [4]. Furthermore, two meta-analyses assessing glucose metabolism in patients with PA showed a higher DM prevalence in patients with PA than in patients with essential hypertension (EHT) [5, 6].

Several hypotheses exist on the mechanisms that cause abnormal glucose metabolism in patients with PA (Fig. 1). In adipose tissue, excess aldosterone induces insulin resistance by producing oxidative stress [7] and increasing proinflammatory cytokines [8]. In the pancreas, aldosterone impairs pancreatic β-cells [9] and reduces insulin synthesis. In muscle, aldosterone decreases glucose uptake, glucose oxidation, and glycogen synthesis by inhibiting insulin signaling [8]. On the other hand, several clinical studies, including the German Conn’s Registry and JPAS, have reported that the plasma aldosterone concentration (PAC) itself has no significant effect on DM prevalence [2, 4]. Therefore, the direct effects of aldosterone on glucose tolerance are still under debate.

Fig. 1
figure 1

Schematic presentation of mechanisms that cause abnormal glucose metabolism in patients with primary aldosteronism

Full size image

The second mechanism that causes abnormal glucose metabolism in patients with PA is hypokalemia. Several studies have shown the relationship between thiazide-induced hypokalemia and impaired glucose metabolism [10, 11]. According to the German Conn’s Registry, even in patients with PA, potassium levels are significantly and negatively correlated with 2-h glucose levels in an OGTT [3]. Watanabe et al. also reported a negative correlation between the serum potassium concentration and the insulin sensitivity index in patients with PA [12]. However, several studies have reported that serum potassium was not associated with a high DM prevalence in patients with PA [2, 13]. Therefore, the effect of hypokalemia on glucose metabolism in patients with PA is also under discussion.

The third mechanism of abnormal glucose metabolism in patients with PA is obesity. Obesity induces the production of adipocytokines, such as C1q/TNF-related protein 1, leptin, and resistin, which have been reported to elevate aldosterone via a renin-independent pathway [13,14,15]. In the JPAS, obesity prevalence was higher in patients with IHA than in patients with APA and EHT [16].

The fourth mechanism of abnormal glucose metabolism in patients with PA is cortisol hypersecretion comorbidity. For adrenal incidentaloma with mild autonomous cortisol excess (MACS), many reports have shown that complicated MACS increases DM prevalence [17,18,19,20]. Similarly, for PA, the prevalence of cortisol cosecretion is high, at 12.8–33.3% [4, 21,22,23], and is associated with impaired glucose metabolism. Gerards et al. reported an association between autonomous cortisol secretion and DM in patients with PA [24]. In addition, Ohno et al. investigated the metabolic influences of a comorbid, apparently nonfunctioning adrenal tumor (AT) in patients with bilateral PA and showed that cortisol levels after a 1-mg dexamethasone suppression test and DM prevalence were higher in patients with AT than in those without AT [25].

Wu et al. previously reported that after a mean follow-up period of 5.2 years, patients with PA who underwent adrenalectomy had an attenuated new-onset DM (NODM) incidence, whereas those treated with mineralocorticoid receptor (MR) antagonists had an augmented NODM risk [26]. The present study evaluated the effect of autonomous cortisol secretion (ACS) on NODM in 387 patients with PA and clarified that PA patients with a concomitant cortisol level of ≧ 2.65 µg/dL after the 1-mg dexamethasone suppression test had a higher NODM incidence after a mean follow-up period of 4.31 years [27]. This result is consistent with those of previous studies and seems reasonable. However, the implication of accompanying MACS differed between APA and IHA. For APA, the presence of MACS generally suggests that an adenoma produces both aldosterone and cortisol. Therefore, when MACS accompanies APA, it can be generally cured by surgery after diagnosis. On the other hand, for IHA, MACS is generally thought not to be involved. However, if MACS accompanies IHA, MACS is considered to persist even after treatment with MR antagonists. Therefore, it is an important proposition whether IHA, which is probably caused by aldosterone-producing cell clusters, can be accompanied by MACS.

Overall, this study strengthened the findings of several previous reports showing that patients with PA have a higher DM prevalence and incidence. However, the molecular mechanisms and the difference between APA and IHA remain unclear and require further research.