Diabetes or Prediabetes in Newly Diagnosed Patients With Nonaffective Psychosis? A Historical and Contemporary View

Abstract

Conflicting evidence on the risk of type 2 diabetes mellitus (T2DM) in people with schizophrenia prior to antipsychotic treatment has been published in recent years. In the meta-analysis presented in the current issue, Mitchell and colleagues1 have compiled all available data about the prevalence of the metabolic syndrome and its components in first episode psychosis with or without antipsychotic medication, concluding that T2DM prevalence was similar between drug-naive patients and the general population. However, glucose intolerance exists on a continuum, and abnormal glucose tolerance less severe than diabetes may only be identifiable after a glucose challenge. These less severe degrees of glucose intolerance were reported in almost all the studies of glucose tolerance that antedated modern antipsychotics. The work of Lorenz2 can be used as an example, but more than ten other manuscripts were published before 1954. Pooled together, these old reports showed abnormalities in the post-glucose challenge blood sugar curve in 16%–66% of cases, while basal metabolic rates were mostly normal. These studies were largely ignored during the new wave of studies of diabetes in schizophrenia. Not surprisingly, recent studies have had similar results. Abnormalities in glucose tolerance were only seen after a challenge (such as oral glucose tolerance test or oGTT) in almost all such recent studies.3–5 To our knowledge, only one study has failed to find a statistically significant difference,6 probably due to a small number of schizophrenia patients. Baseline fasting glucose, the same measure used in this meta-analysis, did not differ between the two groups. It was also seen in our study,7 which had the largest sample so far of unmedicated first episode psychosis patients undergoing oGTT. Results mimic the pre-neuroleptic era studies, with approximately 20% of patients exhibiting abnormal glucose tolerance prior to antipsychotic treatment, above what is expected in the general population. We and others have proposed that schizophrenia is associated with accelerated aging.8 The replicated abnormality in the oGTT, found in patients groups that are typically in their late twenties, provides support for this concept. This evidence does not contradict the strong evidence that health habits, access to care, and antipsychotic drugs increase the risk of T2DM in people with schizophrenia and related disorders. Moreover, most published studies have not considered all of the important confounding variables.9 The impact of diabetes on the lives of people with schizophrenia makes this area an important topic for future research.