Besides its effects on water balance, arginine vasopressin (AVP) increases peripheral vascular resistance and decreases cardiac output, mainly by decreasing heart rate. The current study was designed to evaluate cardiac performance in patients with central diabetes insipidus (CDI), focusing on the acute effects of desmopressin replacement withdrawal and its subsequent reinstatement in patients with CDI. Twelve patients with CDI and 12 sex- and age-matched healthy subjects entered the study. All patients were receiving treatment with intranasal desmopressin at standard doses. All patients and controls were assessed for water balance, by measuring plasma osmolality and total body water, anterior pituitary function, heart rate, systolic and diastolic blood pressure. Left ventricular (LV), end-diastolic and end-systolic diameters (LVEDD, LVESD) and volumes (LVEDV, LVESD), end-diastolic and end-systolic interventricular septum thickness (EDIVST, ESIVST) and posterior wall thickness (LVEDPWT, LVESPWT), and mass (LVM) were measured by echocardiography. Moreover, LV systolic function was assessed by measuring the ejection fraction (EF), the fractional shortening (FS), the Suga index, the stroke volume and the cardiac output, while LV diastolic function was assessed by measuring early (M1) and late (M2) maximal transmitral blood flow velocities, the ratio between M1 and M2, the mitral deceleration time (MDT) and the isovolumetric relaxation time. All parameters were assessed in the patient group 24 h after discontinuing treatment with nasal desmopressin (baseline study) and 1 week after re-starting replacement treatment, while in the control group before (baseline study) and after 1-week of a nasally administered placebo. At baseline, compared to controls, patients with CDI had increased plasma osmolality (P < 0.01), plasma ACTH (P < 0.01), serum (P < 0.01) and urinary cortisol (P < 0.01) levels, and heart rate (P < 0.05), and decreased total body water (P < 0.05). Systolic and diastolic blood pressure and the other anterior pituitary hormones were similar in patients and controls. At echo-cardiography, EDIVST (P < 0.05), ESIVST (P < 0.01), LVEDPWT (P < 0.05) and LVESPWT (P < 0.01), EF (P < 0.01), Suga index (P < 0.05), FS (P < 0.05), M2 (P < 0.01) and IRT (P < 0.05) were significantly higher while LVESD (P < 0.01), LVESV (P < 0.01), LVEDD (P < 0.05), LVEDV (P < 0.05), M1 (P < 0.05), and M1/M2 (P < 0.01) were significantly lower in patients than in controls. LVM, stroke volume and cardiac output, were similar in patients and controls. In the patient group, after 1 week of replacement treatment with desmopressin, all echocardiographic parameters were normalized, except IVT, LVPWT and the diastolic parameters that were still abnormal compared to controls. No difference was found in any of the parameters in the control group one week after placebo administration. Patients with central diabetes insipidus have increased heart rate and left ventricular contractility, and impaired diastolic function. The altered heart rate and left ventricular contractility, reversible after desmopressin replacement, is likely to be due to stimulation of sympathetic nervous activity, induced by the hypovolemia associated with arginine vasopressin deficiency. Conversely, the impairment of diastolic function, persistent after desmopressin replacement, probably relates to a stable impairment of the left ventricular compliance.