Left ventricular hypertrophy (LVH) is often seen in patients with primary aldosteronism (PA), and the prevalence of LVH is reportedly higher among patients with PA than patients with essential hypertension. However, the correlation between aldosterone levels and LVH is undefined, and how aldosterone affects LVH in patients with PA remains unclear. We, therefore, retrospectively assessed a large PA database established by the multicenter JPAS (Japan Primary Aldosteronism Study) to reveal the factors associated with LVH in patients with PA without suspected autonomous cortisol secretion. In the 1186 patients with PA studied, the basal plasma aldosterone concentration, plasma renin activity, and the aldosterone-to-renin ratio did not significantly correlate with left ventricular LV mass index (LVMI) in single or multiple regression analyses. However, the plasma aldosterone concentration after the captopril challenge test or saline-infusion test, which are associated with autonomous aldosterone secretion, correlated significantly with LVMI, even after adjusting for patients' backgrounds, including age and blood pressure. In addition, hypokalemia and the unilateral subtype also correlated with LVMI. Longitudinal subanalysis of medically or surgically treated patients with PA showed significant reductions in LVMI in both the surgery (63.0±18.1 to 55.3±19.5 g/m2.7, P<0.001) and drug treatment (56.8±14.1 to 52.1±13.5 g/m2.7, P<0.001) groups. Our results suggest the autonomous aldosterone secretion level, not the basal aldosterone level itself, is relevant to LVH in patients with PA. In addition, the elevated LVMI seen in patients with PA is at least partially reversible with surgical or medical treatment.