Starting Antipsychotics for Schizophrenia Tied to Blood Sugar Spikes

 When patients with schizophrenia begin second-generation antipsychotic therapy, clinicians typically monitor the scale for early signs of metabolic harm.

However, findings from a recent electronic health record study of adults with severe mental illness (SMI) suggested that blood glucose can surge rapidly within the first 90 days of treatment through both weight-dependent and weight-independent pathways.

Published in the Lancet Psychiatry, the study from Juan Carlos Bazo-Alvarez, PhD, of University College London, and colleagues tracked within-person trajectories of HbA1c and body weight before and after initiating antipsychotics. "The first 3 months after starting an antipsychotic appear to be a particularly important metabolic window," Bazo-Alvarez told MedPage Today.

"Clinicians should not wait for substantial weight gain before checking glycemic measures, because HbA1c and weight do not always change in parallel," he said. "Early monitoring of both outcomes may help identify risk sooner and support timely intervention."

90-Day Metabolic Acceleration

Using England's Clinical Practice Research Datalink Aurum database (2000-2023), the researchers evaluated 96,849 adults ages 18 to 59 (mean age 37.2) with SMI -- including schizophrenia, schizoaffective disorder, and bipolar disorder -- who initiated oral second-generation antipsychotics.

Median baseline weight across the cohort was 167.6 lb and HbA1c was 35 mmol/mol (5.4%). Prior to treatment, participants' HbA1c rose by 0.01 mmol/mol weekly, while weight rose by 0.04 lb weekly.

In the 90 days post-initiation, the weekly rate of HbA1c increase jumped roughly 14 times to 0.12 mmol/mol, producing an excess increase of 1.42 mmol/mol compared with what would be expected had the pre-treatment trend continued. Beyond day 90, the rate of increase returned to baseline levels (0.01 mmol/mol per week), though the early increase was not reversed.

Weight followed a similar surge. Patients gained an average of 0.46 lb weekly during the first 90 days (an excess gain of 5.22 lb) before slowing to 0.07 lb weekly from day 91 onwards.

Notably, testing was skewed. While 76.6% of patients had a recorded weight during follow-up, only 24.9% had an HbA1c measurement.

Uncoupling of Weight and Glucose

Early weight and glycemic changes were not the same across individual medications at day 90:

• Olanzapine: 7.26 lb excess weight gain; 2.33 mmol/mol excess HbA1c increase (steepest combined increases)
• Aripiprazole: 3.84 lb excess weight gain; 2.27 mmol/mol excess HbA1c increase
• Quetiapine: 4.98 lb excess weight gain; 0.15 mmol/mol excess HbA1c increase
• Risperidone: 4.92 lb excess weight gain; 0.54 mmol/mol excess HbA1c increase

"The pronounced increases in both weight and HbA1c with olanzapine were consistent with its well-established metabolic liability," Bazo-Alvarez noted. However, the team was surprised by how quickly the HbA1c trajectory climbed.

"We were also struck by the fact that weight and HbA1c did not always move together; for example, quetiapine was associated with substantial weight gain but relatively little HbA1c change," he said. "This reinforces the message that weight alone does not fully capture early glycemic risk."

Disparities were also evident across subgroups. Women experienced steeper initial surges than men (excess 90-day HbA1c of 1.68 vs 1.18 mmol/mol; excess weight of 5.73 lb vs 5.18 lb), peaking in women starting olanzapine (excess 9.17 lb; excess HbA1c of 3.27 mmol/mol).

Additionally, baseline prediabetes substantially amplified long-term risk, as 13.7% of these patients were diagnosed with type 2 diabetes during follow-up (estimated 2-year cumulative incidence 19.0%) compared with 3.0% of normoglycemic patients.

Commentators Call for Earlier Prophylaxis

In an accompanying commentary, Caroline Hynes-Ryan, MSc, of St. John of God University Hospital in Dublin, and colleagues noted that rapid HbA1c shifts often go unnoticed compared with visible weight gain.

"As weight gain is rapid, visible by the individual and their clinician, and associated with distress, it often receives the most attention as a modifiable or preventable risk factor in this population," they wrote. "However, other cardiometabolic risk factors, such as dysglycemia and dyslipidemia, can be present but asymptomatic and are only identified on routine screening as recommended by international guidelines."

Because current guidelines generally reserve preventive metformin for higher-risk agents -- such as olanzapine or clozapine -- or for patients who have already experienced substantial weight gain, Hynes-Ryan's group suggested the findings support considering metformin (unless contraindicated) for all patients with SMI starting a second-generation antipsychotic.

Clinical Takeaways

The findings support assessing both weight and HbA1c at baseline and at 12 weeks after initiation, rather than relying on weight change as a proxy for metabolic risk, Bazo-Alvarez stressed.

"Patients starting antipsychotics with greater metabolic liability, particularly olanzapine, as well as women and those with prediabetes or other cardiometabolic risk factors, may benefit from closer follow-up and earlier preventive support," he said. "Monitoring should also continue beyond the first 3 months in line with clinical guidance."

However, he cautioned against using the observational findings to avoid specific therapies altogether. "These findings should not be interpreted as meaning that one antipsychotic is always preferable to another," Bazo-Alvarez stated.

"Treatment choice has to balance psychiatric effectiveness, previous response, patient preference, and adverse-effect profiles," he pointed out.

"Our study was observational, so it was not designed to make formal causal comparisons between individual drugs. The practical implication is therefore less about prescribing a specific antipsychotic and more about recognizing that metabolic effects can emerge early and vary between patients," he said.

Disclosures

Bazo-Alvarez reported consultancy fees from the Wellcome Trust, Juli Health, and Swiss Re.

Hynes-Ryan and co-authors reported no disclosures.

https://www.medpagetoday.com/spotlight/schizophrenia/123672