What is the meaning of treatment resistant/refractory major depression (TRD)? A systematic review of current randomized trials

Abstract

Objective

To summarize and discuss the conceptual and operational definitions of treatment resistant/refractory depression (TRD) by systematically reviewing randomized controlled trials (RCTs) on its somatic treatment.

Data sources

We searched the MEDLINE, Cochrane Library, PsycINFO, and EMBASE for potentially relevant RCTs on the somatic treatment of TRD published from January 1996 to June 2006.

Study selection

Studies were included if they: (a) enrolled patients at least 18 years old with a primary diagnosis of unipolar major depression considered resistant/refractory to treatment at baseline, (b) had a randomized design, (c) were published in peer-reviewed journals, and (d) were written in English. Trials that enrolled patients with secondary depression were excluded from our review. Finally, the bibliographies of relevant articles were hand-searched for additional references. In total, 233 full electronic references were retrieved, from which 47 meet the inclusion criteria.

Data extraction

Through a standardized form, we collected data describing the diagnostic procedure, the terminology and definition of TRD employed, the methodology for assessing the adequacy of previous treatments (in terms of type of ascertainment, dose, and duration), and the minimum required depressive symptoms at baseline.

Data synthesis

Overall, RCTs diverged regarding the majority of the conceptual and methodological issues involved in the ascertainment of TRD. For example, eleven terms were used to describe resistance/refractoriness in depression, and six different criteria were employed to define the categorical presence of TRD (ranging form non-response to one antidepressant to non-response to two or more antidepressants from different pharmacological classes). Regarding the evaluation of previous treatments, the majority of RCTs did not use systematic methods to gather data, and diverge substantially on the minimum acceptable medication doses and trial durations.

Conclusions

There is a clear need for an internationally shared framework of concepts and methods for the investigation of TRD that could reduce current idiosyncrasies and provide a reference system. Such a foundation is essential for the interpretation of research findings and for their translation to clinical practice.

Introduction

Although the therapeutic strategies available for the management of depressive disorders have expanded substantially over the years, resistant/refractory major depression (TRD), in its broadest sense, still characterizes a significant number of patients in therapy (O'Reardon and Amsterdam, 2001). Indeed, about one third of patients treated for major depression do not respond satisfactorily to the first antidepressant (AD) (Souery et al., 1999). Furthermore, a considerable proportion of cases have a poor prognosis in follow-up observations, with as much as 20% still being ill 2 years after the onset of illness (Fava, 2003). Finally, there remains a large proportion of individuals (up to 15%), for whom multiple interventions will be unhelpful and who will be significantly depressed in spite of aggressive pharmacological and psychotherapeutic approaches (Burrows and Norman, 1999).

Not surprisingly, in the last decade, a growing number of intervention studies on TRD (including randomized controlled trials (RCTs) have been published. However, a major problem has been the inconsistency in TRD definitions and characteristics (Fava and Davidson, 1996, Berman et al., 1997, Ananth, 1998). There are no consensual operational criteria for TRD, and one can find more than 10 disparate definitions while searching the specialized literature (Sackeim, 2001, Fagiolini and Kupfer, 2003). These definitions range, for example, from a failure to respond to an adequate trial of a single antidepressant for a minimum of 4 weeks (Roose et al., 1986), to a failure of at least one trial of electroconvulsive therapy (ECT) (Fink, 1991). However, none of these definitions have been thoroughly systematically examined for reliability or validated for prospective predictive utility (Souery et al., 1999). Interestingly, a recent study by Petersen and colleagues (2005) empirically tested (using a retrospective design) two of the existing staging models for TRD, namely the Massachusetts General Hospital method (MGH; see Fava, 2003 for details) and the Thase–Rush method (TR; see Thase and Rush, 1997 for details). For this purpose, they evaluated both approaches as predictors of remission status in a sample of 115 outpatients with current major depression. Briefly, their results suggested that the two models are highly correlated, but with the use of multivariate analysis, the MGH model demonstrated significantly greater ability to predict non-remission status than the TR approach. However, this study had several limitations, including its relatively small sample size, the use of chart review methodology, and the non-generalizability of its findings to other populations.

Therefore, as summarized by Fagiolini and Kupfer (2003), available studies “(…) have used different criteria for the number and type of previous failed trials needed to establish a diagnosis of TRD, the definition of treatment adequacy (dose, titration, and duration), the definition of treatment response, and the assessment of primary and comorbid diagnoses”.

Considering the remarkable degree of variation in TRD definition, the growing number of published studies on this subject, and the clinical relevance of resistance, there is a clear need for systematic reviews, which may help professionals working in this field to establish common criteria by providing a representative frame of reference. While previous studies on the conceptual definitions of TRD were primarily expert reviews that proposed different sets of criteria, the present paper aims to summarize and discuss conceptual and operational definitions of TRD by systematically reviewing available RCTs on its somatic treatment.