Predictive value of baseline resistance in early response to antidepressants

Major Depressive Disorder (MDD) is the 3rd source for burden worldwide according to the World Health Organization (WHO). This comes partly from unsatisfactory response rates after usual treatment, highlighting the need for early indicators such as early improvement of depressive symptoms to adapt treatment strategies, especially for severe inpatients. Thus our objective was to assess the predictive value of baseline partial resistance in early antidepressant response (EAR), hypothesizing that previous treatment failures would decrease the probability of early response.

We included 122 consecutive inpatients with current unipolar MDE. The Mini-Neuropsychiatric Interview was used to ascertain DSM-IV diagnoses of MDD as well as psychiatric comorbidities, and to exclude patients with a history of bipolar disorder. A specifically designed questionnaire was used to collect data on previous treatment trials for the current episode so as to generate scores on the five existing methods for quantifying treatment resistance. We prospectively assessed EAR, defined as a 50% decrease in MADRS after 14 days of steady regimen of antidepressant.

In the per protocol sample (N=76), multivariate analyses showed that psychotic features at baseline remained an independent predictor of absence of EAR (p<.01), unlike baseline partial resistance, which may rather be associated with a lack of any improvement.

Lack of data about further response and non-randomized treatment allocation.

Available methods for quantifying treatment resistance are heterogeneous and do not predict short-term response among severely depressed inpatients, despite potential usefulness in predicting a lack of early improvement.