The Nortriptyline Therapeutic Window: A systematic review and meta-analysis

Background: Some studies have found a curvilinear relationship between the nortriptyline plasma concentration and the clinical rating of depression in those treated. The concentration interval of optimum response has been called the “nortriptyline therapeutic window.” Objectives: To analyze all original data relevant to the therapeutic window. Data sources: English language articles found in PubMed using search terms “nortriptyline,” and “plasma concentration,” “serum concentration,” or “blood concentration”. The search was extended through the use of Google Scholar citation links, and reference mining of secondary articles and books. Study selection: Clinical trials or observational studies of adults that reported patient-level or aggregate results of depression response by concentration or concentration range were retained. Data extraction: Plasma concentration and depression ratings were extracted from text, tables and scanned figures. Data synthesis: Studies reporting only qualitative results were reported narratively. Quantitative patient-level and aggregate data were metaanalyzed to compare treatment inside the plasma concentration window with treatment outside the window. A meta-regression was performed to ad∗Biomedical Informatics Training Program and Department of Biomedical Data Science, Stanford University. Contact information: Stanford University School of Medicine, 1265 Welch Road, MSOB X-349, Stanford, CA 94305-5464, p: 650-725-8422, [email protected]. c ©2013, Steven C. Bagley dress variation in study designs and other studylevel variables. Results: In thirty-three clinical trials or observational studies, some evidence was found for a plateau or decline effect at higher plasma concentrations, but the consistency of the effect varied with the depression scale. There was marked variation in response within and outside the window, and no sharp cutoff in response at either the lower or upper window limits. A random-effects meta-analysis found a statistically significant risk difference for treatment within the window compared to treatment above the window of 0.17 (0.04– 0.30, 95% CI), corresponding to a number needed to treat of almost 6. The observational studies showed stronger response than the clinical trials; this depended on a single, anomalous outlier. The meta-regression showed a non-statistically significant trend for a stronger response in observational studies, and small and non-significant effects for studies with endogenous depression and those using fixed-dose regimens. Many studies were small and did not attempt to disconfirm the window’s existence by including patients with concentrations above the upper window limit. Conclusions: Weak, inconsistent evidence of highly varying quality for the nortriptyline therapeutic window was found. Its clinical utility over prudent prescribing practice is limited.