Treating concurrent chronic low back pain and depression with low-dose venlafaxine: an initial identification of "easy-to-use" clinical predictors of early response

Abstract

Objective: Depression and chronic low back pain (CLBP) are both frequent and commonly comorbid in older adults seeking primary care. Serotonin-norepinephrine reuptake inhibitors (SNRIs) such as venlafaxine may be effective in treating comorbid depression and CLBP. For patients with comorbid depression and CLBP, our goal was to identify "easy-to-use" early clinical variables associated with response to 6 weeks of low-dose venlafaxine pharmacotherapy that could be used to construct a clinically useful predictive model in future studies.

Methods: We report data from the first 140 patients completing phase 1 of the Addressing Depression and Pain Together clinical trial. Patients aged ≥60 with concurrent depression and CLBP received 6 weeks of open-label venlafaxine 150 mg/day and supportive management. Using univariate and multivariate methods, we examined a variety of clinical predictors and their association with response to both depression and CLBP; change in depression; and change in pain scores at 6 weeks.

Results: About 26.4% of patients responded for both depression and pain with venlafaxine. Early improvement in pain at 2 weeks predicted improved response rates (P = 0.027). Similarly, positive changes in depression and pain at 2 weeks independently predicted continued improvement at 6 weeks in depression and pain, respectively (P < 0.001).

Conclusions: An important minority of patients benefitted from 6 weeks of venlafaxine 150 mg/day. Early improvement in depression and pain at 2 weeks may predict continued improvement at week 6. Future studies must examine whether patients who have a poor initial response may benefit from increasing the SNRI dose, switching, or augmenting with other treatments after 2 weeks of pharmacotherapy.

Trial registration: ClinicalTrials.gov NCT01124188.

Keywords: Back Pain; Clinical Trial; Depression; Geriatrics; Predictors of Response; Primary Care; Venlafaxine.

Figure 1

Change in PHQ-9 at 2 weeks as a Predictor of Change in PHQ-9 at 6 weeks (n=140) Univariate Pearson Correlation: r=0.51, p<0.001 This figure illustrates the correlation between Change in PHQ-9 at 2 weeks and Change in PHQ-9 at 6 weeks.

Figure 2

Change in NRS at 2 weeks as a Predictor of Change in NRS at 6 weeks (n=140) Univariate Spearman Correlation: rho=0.41, p<0.001 This figure illustrates the non-parametric (spearman) correlation between Change in NRS at 2 weeks and Change in NRS at 6 weeks. Rank-Order Change in NRS at 2 weeks was computed by ordering values of “Change in NRS at 2 weeks” in ascending order (1–140). The largest “changes in NRS at 2 week” values had the highest numbers for rank-order change (e.g. 140).