Aripiprazole Augmentation in the Treatment of Military-Related PTSD with Major Depression: a retrospective chart review

Participants consisted of a sample of 27 out of 123 consecutive veterans who consented to try aripiprazole augmentation between November 2009 and August 2010. All patients were receiving outpatient psychiatric care at a clinic specializing in the assessment and treatment of military-related psychiatric conditions. The clinic follows a standardized assessment and treatment protocol; the standardized assessments included the PTSD Checklist-Military Version (PCL-M), [40] Beck Depression Inventory (BDI), [41] and Medical Outcomes Study (MOS) 36-item Short-Form Health Survey (SF-36), [42] administered at intake and at each follow-up appointment over the course of treatment. In addition to providing psychoeducation, the standard psychiatric treatment at the clinic includes symptom management, treatment of comorbid disorders, and management of functional impairment [5]. Participants were prescribed aripiprazole after demonstrating minimal or partial response to their existing antidepressant and/or minimal or partial response to other antipsychotic augmentation strategies. All subjects received a comprehensive psychiatric evaluation and laboratory tests (complete blood count with white count differential, serum electrolytes, glucose, creatinine, blood urea nitrogen, liver function tests, and lipid profile). The initial dose of aripiprazole was 2 to 5 mg daily, with further dose titrations based on tolerability and clinical response, up to a maximum of 30 mg daily. Efficacy and adverse effects were assessed and recorded at each follow-up appointment (bi-weekly for the first month and then monthly). Antidepressants, anxiolytics, and mood stabilizers were allowed but had to be kept at a constant dose during the 12- weeks treatment phase. As we were specifically interested in examining the benefits of aripiprazole augmentation in veterans with military-related PTSD, patients not prescribed aripiprazole augmentation were excluded from this chart review. For patients who were already prescribed an antipsychotic, the antipsychotic was discontinued prior to the initiation of aripiprazole.

The sample was derived from a retrospective chart review with approval from the Office of Research Ethics of the University of Western Ontario. Consent from each patient was not obtained specifically for this chart review. However, as part of the initial orientation to the clinic, all patients are asked to provide written consent to participate in research. All patients met the DSM-IV criteria for PTSD and comorbid major depressive disorder. To maximize the generalizability of this evaluation to usual "clinical practice," all comorbid physical and psychiatric conditions were included in the chart review.

To diagnose PTSD, the Clinician-Administered PTSD Scale (CAPS) [43] was administered by a trained clinician and the diagnosis of major depressive disorder was determined using the Patient Health Questionnaire (PHQ-9)[44] and the psychiatric interview according to the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV) criteria [6]. To assess change in PTSD symptoms with treatment, the PCL-M [40] was used. Similarly, the BDI [45, 46] was used to assess change in depressive symptoms as a result of treatment. The SF-36 [47] was used to assess health related quality of life (HRQol). The SF-36 measures functional impairment in eight domains or subscales; the mental health sub scales can be collapsed into the Mental Component Summary (MCS) Score reflecting overall mental health [48]. The PCL-M, the BDI, and the SF-36 were all administered at each follow-up appointment including at intake (t1) and at 1-, 2-, and 3-month follow-ups (t2, t3, and t4) over the course of treatment. The total PCL-M and the total BDI were the primary efficacy variables; the total PCL-M re-experiencing, avoidance/numbing, and hyperarousal subscale scores served as secondary efficacy variables.

The LOCF data for each visit included the data recorded at that visit or carried forward from the last visit. For the primary outcome, the Wilcoxon signed rank test and effect sizes (Cohen's d) was used to determine the statistical significance of the change from baseline total PCL-M and Total BDI score at each follow-up time point (t2, t3, t4). For the three secondary outcome measures (the PCL-M reexperiencing, avoidance/numbing, and hyperarousal subscale scores), only the effects at t4 were examined. To correct for multiple comparisons, we used Hochberg step-up multiple comparisons procedure [49].

For each of the two main outcomes, we identified responders on our two outcome measures using criteria established in existing literature. More specifically, on the PCL, we identified those with at least a 20% reduction in their total PCL scores as responders; [33] this is in keeping with prior treatment efficacy and more specifically, prior Aripiprazole treatment efficacy studies for PTSD [33]. On the BDI, we identified those with at least a 50% reduction in their symptom scores as responders [50].

Demographic and clinical characteristics of the sample are presented in Table 1. Of the 27 participants, almost all were men (n = 26, 96.3%), with an average age of 39.36 years (SD = 5.98) and an average of 10.26 years (SD = 7.38) with PTSD symptoms. At intake, the majority (n = 20, 74.1%) of the sample had been released from the military. Years of military service averaged 15.04 years (SD = 7.41). Almost all (n = 26, 96.3%) had exposure to combat or to a war zone during their deployment and the most common deployments reported were Afghanistan (n = 9, 33.3%), the former Yugoslavia (n = 8, 29.6%), and Africa (Somalia, Rwanda, Eatrea, and Sierra Leone) (n = 5, 18.5%). All patients were taking an antidepressant prior to initiating aripiprazole, most commonly a norepinephrine dopamine reuptake inhibitor (NDRI) (n = 16, 59.3%), followed by noradrenergic specific serotonergic antidepressants (NaSSA) (n = 12, 44.4%); serotonin norepinephrine reuptake inhibitor (SNRI) (n = 11, 41.7%), and selective serotonin reuptake inhibitor (SSRI) (n = 10, 37%). Most patients (n = 14, 51.9%) were taking two antidepressant and an additional 14.8% (n = 4) were taking three antidepressant. In addition to an antidepressant, 8 patients (30.77%) were prescribed an anticonvulsant, 5 patients (19.23%) were prescribed a stimulant, and 4 patients (15.38%) were prescribed a hypnotic.

Twenty-seven patients had at least one post-baseline efficacy evaluation thus were included in the efficacy analysis. At intake, all participants met full criteria for PTSD based on the CAPS interview. The average duration of PTSD symptoms was 10.26 years (SD = 7.38), suggesting a chronic course for this sample. Intake scores on the PCL-M and the BDI averaged 56.11(SD = 12.66) and 30.44 years (SD = 7.86), respectively. On the PCL-M, a score of 50 is the conventional cut-off score for a positive screen for PTSD in veteran populations; [51] on the BDI, scores above 29 are considered indicative of severe depression [52]. SF-36 MCS and PCS were 20.22 (SD = 10.11) and 32.48 (SD = 15.06), respectively, indicating significant impairment in scales measuring both mental and physical functioning.

The final average dose of aripiprazole was 12.40 (SD = 4.35) mg daily. The mean value of weight decrease from baseline (mean = 99.25 kg, SD = 13.35) to the final visit is 1.05 kg (SD = 4.95). Only two patients discontinued the aripiprazole; one patient due to non-response and one due to intolerable restlessness. The remaining 25 patients tolerated the aripiprazole. The most common side effects reported were insomnia (N = 5/27, 18.5%); agitation/irritability (N = 4/27, 14.8%); restlessness (N = 3/27, 11.1%); and fatigue (N = 2/27, 7.4%). The good tolerability was likely related to lower starting doses (2 mg daily) and slow titration (increasing the dose by 2-5 mg every two weeks).

Results from Wilcoxon signed rank tests showed significant decreases between baseline and each of the three follow-ups for the BDI and the PCL, as well as the last visit for the PCL Reexperiencing, Avoidance, and Hyperarousal subscales (Table 2), even after multiple testing correction. The total PCL score decreased from 56.11 (SD = 12.66) at baseline to 46.85 (SD = 13.53) at three months and the total BDI score 30.44 (SD = 7.85) at baseline to 20.67 (SD = 10.05) at three months, figure 1. Effect sizes and changes in the clinical outcome variables from intake to the final visit are reported in Table 3.

The number and percentage of responders at each of the follow-ups are reported in Table 4. Thirty seven percent (10/27) were considered responders, as defined by a decrease in total PCL scores of at least 20 percent and for depression, 19% (5/27) were considered responders as defined by a decrease in total BDI score of at least 50%. Overall, a higher percentage of participants met criteria for being a responder on the PCL than the BDI.

Additional analyses to examine the efficacy of aripiprazole among those with severe depression (BDI > 29) at intake (n = 14) found significant reductions in PCL symptom scores for this subsample as well, with the PCL scores averaging 61.79 (SD = 9.60), 51.86 (SD = 13.54), 52.14 (SD = 15.53), and 48.71 (SD = 13.02) at intake and each of the follow-ups, respectively. The Wilcoxon signed rank test results showed significant reductions from intake to each of the follow-ups (p = 0.0051, p = 0.0405, and 0.0010 for each of the follow-ups). The percentages of responders on PCL at each of the follow-ups are 35.71%, 42.86%, and 57.14%, respectively. It is worth noting that at t4, having severe depression was found to be associated with higher percentage of responders on PCL (P = 0.0461 from Fisher's exact test).