Contents lists available at ScienceDirect Journal of Afective Disorders journal homepage: www.elsevier.com/locate/jad Research paper Clinical response and metabolic efects of lithium in 323 mood disorder patients Marco Pinna a,b , Mirko Manchia c,d, , Caterina Visioli a , Leonardo Tondo a,e a Mood Disorder Lucio Bini Center, Cagliari, Italy b Unit of Psychiatry, Department of Mental Health, Santissima Trinità Hospital, Cagliari, Italy c Unit of Psychiatry, Department of Medical Science and Public Health, University of Cagliari, Cagliari, Italy d Department of Pharmacology, Dalhousie University, Halifax, Nova Scotia, Canada e McLean Hospital-Harvard University, Boston, United States ARTICLEINFO Keywords: Bipolar disorder Major depressive disorder Longitudinal assessment Metabolism ABSTRACT Background: Lithium is the mainstay for the maintenance treatment of mood disorders (MD), but its efcacy needs to be weighed against its side efects profle. Here, we assessed retrospectively the clinical response to long-term lithium treatment, as well as the rate of associated metabolic side efects. Methods: Clinical data were collected from patients treated with lithium for at least 12 months at the Lucio Bini Center for Mood Disorders in Cagliari, Italy. Clinical response was determined as the diference in number of moodepisodesandpercentofillnesstimebeforeandduringlithiumtreatment.Symptomaticvaluesofmetabolic parameters (plasma levels of glucose, cholesterol, urea nitrogen [BUN], creatinine, TSH, white blood cells [WBC]), and Body Mass Index (BMI) were determined. Results: We studied 323 MD patients (60.2% women). The percent of illness time was signifcantly reduced for both depressive (F = 4.94, p<0.0001), and manic (F = 3.95, p < 0.0001) episodes, whereas the rates of episodes/year were signifcantly reduced for mania (F = 2.01, p = 0.02), but not for depression (F = 1.54, p = 0.06). Rates of MD patients with symptomatic values of metabolic parameters were 14.3% for WBC ≥10,000/µl, 26.9% for glucose levels ≥100 mg/dl, 54.2% for cholesterol levels ≥200 mg/dl, 7.5% for BUN ≥50 mg/dl, 21.9% for creatinine levels ≥1.2 mg/dl, and 20.9% for TSH levels ≥3.50 mU/l. Conclusions: Long-term lithium treatment was clinically efective, but the rates of metabolic efects were sub- stantial although some of them were also associated with older age. Lithium-treated patients should receive accurate clinical monitoring to decrease the impact of long-term side efects. 1. Introduction Lithium is an efective antimanic and prophylactic treatment for mood disorders (MD) clinically used since the 1950s (Cade 1949; GrandjeanandAubry2009; ManjiandLenox,1998; Price and Heninger 1994; Smith et al., 2007; Vieta et al., 2011). In addition to its mood- stabilizing properties, lithium reduces signifcantly the risk ofsuicidein MD patients probably more than anticonvulsants (Baldessarini et al., 2001; Geddes et al., 2004). Naturalistic studies report that approxi- mately one-third of lithium-treated patients achieve full remission (Alda, 2017; Kessing et al., 2011) and that an additional 40%, although still presenting recurrences, show substantial clinical improvement (Baldessarini and Tondo, 2000). Lithium treatment also is associated with safety concerns because of its low therapeutic index (ratio of median toxic to therapeutically efective doses, of approximately 3 requiring careful monitoring for adverse efects, mainly during long-term treatment) (Livingstone and Rampes, 2006; McKnight et al., 2012). Further, lithium treatment can have a substantial impact at metabolic level and it is frequently asso- ciated with hypothyroidism, hyperparathyroidism with consequent hypercalcemia, weight gain, and reduced urinary concentrating ability (McKnight et al., 2012). Concerning hypothyroidism, it is known that lithium might afect various aspects of thyroid gland function. The prevalence of hy- pothyroidisminpatientstakinglithiumvarieswidelybutcanbeashigh as 47%, with the female to male ratio being 5:1 (Esposito et al., 1997; Henry 2002) and may contribute to an increased depressive morbidity (Frye et al., 1999). Hypercalcemia associated with hyperparathyroidism is common in pa- tients treated with lithium, with a prevalence ranging from 6.3% to 50% https://doi.org/10.1016/j.jad.2020.03.066 Received 8 October 2019; Received in revised form 13 January 2020; Accepted 22 March 2020 Corresponding author. E-mail addresses: mirkomanchia@unica.it, mirko.manchia@dal.ca (M. Manchia). Journal of Affective Disorders 270 (2020) 9–14 Available online 01 April 2020 0165-0327/ © 2020 Elsevier B.V. All rights reserved. T