Comparative effects of 18 antipsychotics on metabolic function in schizophrenia, predictors of metabolic dysregulation, and association with psychopathology: a systematic review and network meta-analysis

Abstract

Background: Antipsychotic-treatment is associated with metabolic disturbance. However, the relative degree to which metabolic alterations occur in treatment with different antipsychotics remains unclear. Furthermore, predictors of metabolic dysregulation are poorly understood, and association between metabolic-change and change in psychopathology is uncertain. As such, we aimed to compare and rank antipsychotics based on their metabolic side-effects, identify physiological and demographic predictors of antipsychotic-induced metabolic dysregulation, and determine the relationship between change in psychotic symptoms and change in metabolic parameters with antipsychotic treatment. Methods: We searched Medline, EMBASE and PsychINFO from inception until June 30, 2019. We included blinded randomised controlled trials (RCTs) comparing 18 antipsychotics and placebo in acute-treatment of schizophrenia. We performed frequentist random-effects network meta-analyses (NMAs) to investigate treatment-induced changes in body weight, BMI, total/LDL/HDL-cholesterol, triglycerides, and glucose. We performed meta-regressions to examine relationships between metabolic change and age/gender/ethnicity/baseline-weight/baseline-metabolic parameter level. We examined the association between metabolic change and psychopathology change by estimating the correlation between symptom severity change and metabolic parameter change. Outcomes: Of 6532 citations, 100 RCTs met inclusion criteria, including 25,952 patients. Median treatment-duration was 6-weeks. According to our NMAs, mean differences for weight-gain compared to placebo ranged from -0.23 (95% CI: -0.83, 0.36) for best (haloperidol) to +3.01kg (1.78, 4.24) for worst (clozapine); for BMI from -0.25 (-0.68, 0.17) for best (haloperidol) to +1.07kg/m2 (0.90, 1.25) for worst (olanzapine); for total-cholesterol from -0.09 (-0.24, 0.07) for best (cariprazine) to +0.56mmol/L (0.26, 0.86) for worst (clozapine); for LDL-cholesterol from -0.13 (-0.21, -0.05) for best (cariprazine) to +0.20mmol/L (0.14, 0.26) for worst (olanzapine); for HDL-cholesterol from +0.05 (0.00, 0.10) for best (brexpiprazole) to -0.10mmol/L (-0.33, 0.14) for worst (amisulpride); for triglycerides from -0.01 (-0.10, 0.08) for best (brexpiprazole) to +0.98mmol/L (0.48, 1.49) for worst (clozapine); for glucose from -0.29 (-0.55, -0.03) for best (lurasidone) to 1.05mmol/L (0.41, 1.70) for worst (clozapine). Greater increases in glucose were predicted by higher baseline-weight (p=0.0015) and male-gender (p=0.0082). Non-Caucasian ethnicity was associated with greater increases in total-cholesterol (p=0.040). Improvements in symptom severity were associated with increases in weight (ρ=0.36, p=0.0021), BMI (ρ=0.84, p<0.0001), total-cholesterol (ρ=0.31, p=0.047), and LDL-cholesterol (ρ=0.42, p=0.013), and decreases in HDL-cholesterol (ρ= -0.35, p=0.035). Interpretation: There are marked differences between antipsychotics in terms of metabolic side-effects, with olanzapine and clozapine exhibiting the worst profiles and aripiprazole, brexpiprazole, cariprazine, lurasidone, and ziprasidone the most benign profiles. Higher baseline weight, male gender, and non-Caucasian ethnicity are predictors of vulnerability to antipsychotic-induced metabolic change, and improvements in psychopathology are associated with metabolic disturbance. Treatment guidelines should be updated to reflect our findings. However, choice of antipsychotic should be made on an individual basis, considering the clinical circumstances and preferences of patients, carers, and clinicians.