- Email: [email protected]
Predicting relapse in major depression after successful initial pharmacological treatment
Journal of Affective Disorders 250 (2019) 108–113
Contents lists available at ScienceDirect
Journal of Affective Disorders journal homepage: www.elsevier.com/locate/jad
Research paper
Predicting relapse in major depression after successful initial pharmacological treatment
T
⁎
Tatsuo Akechia, , Akio Mantanib, Ken'ichi Kuratac, Susumu Hirotad, Shinji Shimoderae, Mitsuhiko Yamadaf, Masatoshi Inagakig, Norio Watanabeh, Tadashi Katoi, Toshi A. Furukawah, for the SUND Investigators a
Department of Psychiatry and Cognitive-Behavioral Medicine, Nagoya City University Graduate School of Medical Sciences, Nagoya, Japan Mantani Mental Clinic, Hiroshima, Japan c Kabe Mental Health Clinic, Hiroshima, Japan d Hirota Clinic, Hiroshima, Japan e Department of Neuropsychiatry, Kochi Medical School, Kochi, Japan f Department of Neuropsychopharmacology, National Institute of Mental Health, National Center of Neurology and Psychiatry, Tokyo, Japan g Department of Neuropsychiatry, Okayama University Hospital, Okayama, Japan h Kyoto University Graduate School of Medicine, School of Public Health Department of Health Promotion of Human Behavior, Kyoto, Japan i Aratama Kokorono Clinic, Nagoya, Japan b
A R T I C LE I N FO
A B S T R A C T
Keywords: Major depression Antidepressant Relapse Predictor
Background: Identifying the predictors of relapse could help to develop more individualized treatment strategies for major depression. The study aim was to explore predictors of depression relapse after remission using data from our previous multicenter randomized practical trial of patients with major depression. Methods: Our cohort comprised subjects with Patient Health Questionnaire (PHQ-9) scores less than 5 after antidepressant treatment for 9 weeks. Relapse was defined as a PHQ-9 score of 5 or more at week 25. We examined patient demographic and clinical characteristics at baseline (age, sex education, job status, marital status, onset age at first depressive episode, number of previous episodes, length of current episode, scores on the nine PHQ-9 criteria at week 0) and Frequency, Intensity, and Burden of Side Effects Rating Scale and PHQ-9 total scores at week 9 (residual symptoms) as potential predictors of depression relapse at week 25. Results: Of 494 patients remitted at week 9, 71 (14.4%) experienced relapse at week 25. Logistic regression analysis showed that lower PHQ-9 depressive mood score at week 0, higher suicidal ideation score at week 0, and total PHQ-9 score at week 9, and greater severity of side effects at week 9 were significant predictors. On the other hand, when relapse was defined as a PHQ-9 score of 10 or more at week 25, there were no significant predictors. Limitations: There may be other important predictors that this study failed to identify and the findings obtained may be sensitive to the specific definition of relapse. Conclusions: Approximately one-seventh of subjects who remitted after 2 months of acute-phase treatment experienced depression relapse within 4 months of remission. Lower depressive mood and higher suicidal ideation upon development of the current depression episode, the presence of residual symptoms, and greater severity of side effects at remission may predict subsequent depression relapse.
1. Introduction Major depression is a leading cause of disability and its burden is expected to increase during the next 10 years (WHO, 2008). As major depression is one of the most prevalent psychiatric disorders in most
developed countries (Kawakami et al., 2004; Kessler et al., 2003), the establishment of effective health strategies to manage and prevent depression is important. Although numerous treatments are available for major depression and meta-analysis indicates that cognitive behavioral therapy has an
⁎ Correspondence to: Department of Psychiatry and Cognitive-Behavioral Medicine, Nagoya City University Graduate School of Medical Sciences, 1 Kawasumi, Mizuho-cho, Mizuho-ku, Nagoya, Aichi 467-8601, Japan. E-mail address: [email protected] (T. Akechi).
https://doi.org/10.1016/j.jad.2019.03.004 Received 9 August 2018; Received in revised form 12 February 2019; Accepted 3 March 2019 Available online 05 March 2019 0165-0327/ © 2019 Elsevier B.V. All rights reserved.
Journal of Affective Disorders 250 (2019) 108–113
T. Akechi, et al.
used the minimization method, adjusting for site, whether 50% or greater reduction on PHQ-9 score had been achieved, and whether patients reported moderate or greater impairment owing to side effects. In Step 2, in the continue-sertraline arm, sertraline was administered at 50 or 100 mg/day according to the initial randomization (The maximum sertraline dose was 100 mg daily, half of the recommend maximum dose in the U.S because the maximum of the licensed dosage in Japan is 100 mg/day). In the augmentation arm, sertraline was continued and mirtazapine was added at between 7.5 and 45 mg/day at the discretion of the study psychiatrist. In the mirtazapine switch arm, mirtazapine at between 7.5 mg and 45 mg/day was administered; sertraline was tapered and discontinued by week 7. Co-administration of benzodiazepines, but not of other antidepressants, antipsychotics, or mood stabilizers, was allowed up to week 9. After week 9, treatment was at the physician's discretion. All treatments were open-label. Institutional review boards at each participating site approved the study. The lead site, Kyoto University, hosted the regulatory and data management cores. An independent data monitoring committee oversaw the trial. All participants provided written informed consent. This study used the data of 494 subjects who remitted at week 9 and could be followed-up at week 25.
enduring effect following termination of the acute treatment (Cuijpers et al., 2013), pharmacotherapy is one of the most widely used standard therapies in clinical practice and is mentioned in most clinical practice guidelines. Most guidelines recommend at least 6–9 months of continuous treatment for maintenance therapy (preventing relapse and recurrence) after remission(APA, 2010; Bauer et al., 2013, 2015; Cleare et al., 2015; Geddes et al., 2003; Kennedy et al., 2016; NICE, 2009, last updated April 2016), because many patients (typically 20%–40%) experience depression relapse within 1 year even after depression remission (Geddes et al., 2003; Kanai et al., 2003; Maj et al., 1992; Melartin et al., 2004; Mueller et al., 1999; Ramana et al., 1995). Metaanalysis demonstrates that prophylactic antidepressant drug therapy is useful for preventing future relapses (Williams et al., 2009). Several guidelines and studies suggest that clinical factors (particularly number of previous depression episodes and subclinical residual symptoms) rather than patient demographic factors (e.g., age and gender) can predict depression relapse(Bauer et al., 2015; Cho et al., 2008; Hardeveld et al., 2010; Harkness et al., 2012; Jang et al., 2013; Kennedy et al., 2016; NICE, 2009, last updated April 2016). However, most of these studies have tended to use small samples, typically 60–80 subjects,(Kanai et al., 2003; Maj et al., 1992; Ramana et al., 1995) except for two studies conducted by Melartin et al. (2004) (n = 198) and Mueller et al. (1999) (n = 380). As the identification of possible predictors of relapse could help to develop more individualized management strategies for major depression, we investigated the proportion and predictors of depression relapse after remission following acute-phase antidepressant treatment using data from our previous randomized controlled clinical trial (n = 494): the SUN☺D trial (Kato et al., 2018).
2.2. Clinical assessments and outcomes Assessments were undertaken at weeks 0, 1, 3, 9, and 25 after starting sertraline in the SUN☺D trial. At week 0, the study psychiatrist made a diagnosis using a semi-structured interview, the Primary Care Evaluation of Mental Disorders (PRIME-MD) (Spitzer et al., 1994). The PRIME-MD allows the calculation of the severity of the index depression episode as measured by the PHQ-9. At weeks 1, 3, 9, and 25, assessors who were trained at and operating from the central office and blind to the treatment assignment and the timing of the assessment administered the PHQ-9 and the Frequency, Intensity, and Burden of Side Effects Rating Scale (FIBSER) by telephone (Pinto-Meza et al., 2005). We previously established the inter-rater reliability of the telephone assessment (Shimodera et al., 2012). The telephone assessments were well blinded, as assessors had no more than chance success at guessing the allocation. The PHQ-9 consists of the nine diagnostic criteria items for major depression from the Diagnostic and Statistical Manual of Mental Disorders, 4th edition (DSM-IV) (Spitzer et al., 1999). Each item is rated from 0 (Not at all) to 3 (Nearly every day). The total score range is 0–27; higher ratings indicate more severe depression. The instrument has demonstrated excellent reliability, validity, and responsiveness (Furukawa, 2010). The FIBSER comprises 7-point scales that measure the frequency, intensity, and burden of side effects. Higher ratings indicate greater severity and the total score range is 0–21 (Rush et al., 2006).
2. Methods 2.1. Participants This study represents a secondary analysis of the SUN☺D trial. The SUN☺D was a multicenter, open-label, assessor-blinded, practical randomized controlled clinical trial that involved two randomization steps to examine first- and second-line treatment strategies for untreated unipolar major depression. Specifically, the trial consisted of two steps: Step 1 was a cluster-randomized trial comparing titration up to the minimum vs. maximum of the recommended dose range among patients starting with sertraline. Step 2 addressed the impact of continuing initial treatment, augmenting with mirtazapine, or switching to mirtazapine for patients who started treatment for major depression with sertraline but had not remitted by 3 weeks. Briefly we recruited adult men and women between 25 and 75 years who had a primary diagnosis of a non-psychotic unipolar major depressive episode within the past month. All recruited persons were also not allowed to be taking any antidepressant, antipsychotic or mood stabilizer. This trial was conducted in the psychiatry departments of 48 hospitals and clinics across Japan between December 2010 and September 2015. The protocol provides additional details of the eligibility criteria (Furukawa et al., 2011) and the main findings have been published previously (Kato et al., 2018). Fig. 1 presents the study flow. First, patients with non-psychotic unipolar major depression for whom the treating physician had judged that starting treatment with sertraline was indicated (week 0) were followed-up and patients who were tolerant of sertraline 25 mg/day for 3–16 days were recruited. In Step 1, participants received first-line treatment with sertraline to be titrated up to either 50 mg/day or 100 mg/day by week 3 according to cluster randomization by site. In Step 2, participants who had not reached remission, defined as a score of 4 or less on the Patient Health Questionnaire (PHQ-9) at week 3, were randomized 1:1:1 (using a web-based central computer system) to continue with sertraline, to augment sertraline with mirtazapine, or to switch to mirtazapine (Fig. 1). The Step 2 individual randomization
2.3. Definition of cases with remission and predictors investigated We defined the remission of depression as a PHQ-9 score less than 5, following a previous study by Kroenke et al. (2001). They provided the following rules of thumb for interpreting continuous PHQ-9 scores: 0–4 no depression; 5–9 mild depression; 10–14 moderate depression; 15–19 moderately severe depression; 20+ severe depression. We divided subjects into two groups: those who maintained remission at week 25 vs. those who relapsed and whose PHQ-9 score was greater than 4 at week 25. In this study, we examined the following variables as potential predictors: PHQ-9 scores for each of the nine criteria (evaluated by PRIME-MD) at week 0; patient demographic and clinical characteristics (age, gender, education, job status, marital status, age of onset at first depressive episode, number of previous depressive episodes, length of current episode at week 1), and FIBSER and total PHQ-9 scores at week 9 (residual symptoms) after initial treatment. 109
Journal of Affective Disorders 250 (2019) 108–113
T. Akechi, et al.
Assessed for eligibility (n=56,261) Non-psychotic unipolar major depression for whom the treating physician had judged that starting treatment with sertraline in indicated (week 0)
Excluded before Step 1 randomization (n=54,250) No major depressive episode (n=48,366) Declined to participate or not meeting other eligibility criteria (n=5,884)
Step 1: Cluster-randomized to sertraline 50 mg/d or 100 mg/d arms (n=2,011)
Excluded before Step 2 randomization (n=365) Unable to be followed up at week 3 (n=58) Followed up but withdrew consent to be randomized (n=76) Protocol violation (n=1) Remitted at week 3 (n=230)
Step 2: Based on assessment of depression severity and side effect at week 3, non-remitters were randomized individually (n=1,647)
Allocated to
Allocated to
Allocated to
continue
augment
switch to
with
sertraline
mirtazapine
sertraline
with
(n=558)
(n=551)
mirtazapine (n=538)
Remission (n=423) and relapse (n=71) at week 25 Fig. 1. Screening, randomization and follow-up.
2.4. Statistical analysis
2015), we avoided bivariate analysis and conducted logistic regression analysis with backward elimination (likelihood ratio) to explore predictors of worsening depression. In this logistic regression model,
In accordance with the PROBAST risk of bias tool (Wolff et al., 110
Journal of Affective Disorders 250 (2019) 108–113
T. Akechi, et al.
Furthermore, 15.4% and 24.3% of subjects discontinued all antidepressants by the end of 12 or 25 weeks, respectively. Among the subjects who did not discontinue antidepressants, 15.6% of subjects experienced relapse.
presence or absence of relapse was entered as a dependent variable, and the patient demographic and clinical characteristics mentioned above were entered as independent variables as potential predictors. In this analysis, we performed multiple imputation to complement missing data including primary outcome data in order to undertake intent-totreat-analyses. Finally to investigate the potential effect of treatment and medication discontinuation on the findings, additional analyses were conducted for controlling for treatment arm and for medication discontinuation. Variance inflation factors (VIFs) was used for multicollinearity diagnostics for the current logistic regression model. Finally additional bivariate analyses were conducted to check the results (e.g. obtained significant predictors and relapse). A p-value of less than 0.05 was used as the significance level in all statistical analyses; all p-values reported were two-tailed. All statistical procedures except multiple imputation done by R were conducted using IBM SPSS Statistics version 25 software for Windows (SPSS Inc., NY, 2017).
3.2. Predictors of worsening depression Since there were total of 10 missing data of follow-up at week 25, these were complemented by using multiple imputation. The stepwise logistic regression analysis showed that lower PHQ-9 depressive mood score at week 0, higher suicidal ideation score at week 0, total PHQ-9 score at week 9, and FIBSER total score at week 9 independently and significantly predicted depression relapse. PHQ-9 scores for the nine criteria at week 0 (except depressive mood and suicidal ideation) and other clinical factors, including patient demographic and clinical characteristics at week 1 (age, gender, education, job status, marital status, age of onset at first depressive episode, number of previous depressive episodes, length of current episode) did not significantly predict depression relapse (Table 2). Additional bivariate analyses showed that lower PHQ-9 depressive mood score at week 0 is borderline significance (p = 0.07) and higher suicidal ideation score at week 0, total PHQ-9 score at week 9, and FIBSER total score at week 9 are significant. Calculation VIFs for multicollinearity diagnostics for the current logistic regression model showed that all VIFs were not high (e.g. less than 4.1) and this suggests validity of the analysis. Even after controlling for treatment arm and for medication discontinuation, the findings are same (data not show,). In addition, there were no interactions of predictors of relapse with treatment arm and discontinuation (data not shown). Finally, when relapse was defined as a PHQ-9 score of 10 or more at week 25, there were no significant predictors in this analysis partly because the number of relapse cases is decreased (N = 11).
3. Results 3.1. Participants Fig. 1 shows the screening, randomization, and follow-up of study participants. As shown, 1647 had not remitted by week 3 and were individually randomized to continue sertraline (n = 551), augment sertraline with mirtazapine (n = 538), or switch to mirtazapine (n = 558). Of these, 494 patients remitted at week 9 and completed week 25 assessments (Table 1). The mean subject age was early 40 s and half the subjects were female. Approximately 40% worked full-time and more than half were married. The median number of previous depressive episodes was one and the number of past episodes was variable. At week 25, 71 subjects (14.4% [95% confidence interval: 12% to 18%]) experienced some degree of relapse and 423 subjects (85.6%) were still in remission. The number and percentage of relapsed subjects were as follows: mild depression (n = 60, 12.1%); moderate depression (n = 8, 1.6%); moderately severe depression (n = 2, 0.4%); and severe depression (n = 1, 0.2%). Regarding cessation of antidepressant treatment, 33.8% and 50.3% of subjects discontinued the antidepressants assigned at Step 2 randomization by 12 and 25 weeks, respectively.
4. Discussion The current study demonstrated that approximately one-seventh of subjects experienced some relapse of their depression within 4 months even after remission following 2 months of acute-phase antidepressant treatment. Most relapses were mild. Our findings also suggested that suicidal ideation upon development of the current depression episode and presence of residual symptoms at remission could predict subsequent depression relapse after remission. The proportion of participants that experienced relapse is generally consistent with findings from previous reports (Kanai et al., 2003; Maj et al., 1992; Melartin et al., 2004; Mueller et al., 1999), although the present study population and setting differed from those in previous studies. Considering that 24.3% of our subjects discontinued all antidepressant use by the end of 25 weeks, and the strong evidence that continuous antidepressant treatment can reduce the risk of relapse, (Geddes et al., 2003) there is a clear need for maintenance therapy for major depression. Many patients suffer from early depression relapse; therefore, physicians should treat patients with depression carefully and monitor them even after they have achieved remission. In addition, most guidelines clearly state the need for continuous treatment, typically for more than 6 months, although this depends on patients’ individual risk of relapse (Bauer et al., 2015; Kennedy et al., 2016; NICE, 2009 last updated April 2016). The implementation and dissemination of a guideline treatment strategy for relapse prevention is recommended. To the best of our knowledge, there are no studies indicating that milder depressive mood, presence of suicidal ideation in the current depressive episode, and greater severity of side effects at remission are significant predictors of subsequent depression relapse. It is not clear why milder depressive mood in the current depression episode is a potential predictor of future relapse. However, our clinical experience
Table 1 Characteristics of the subjects: remission at week 9. All subjects (n = 494) Demographic characteristics* Age, year mean (SD) Female sex, n (%) Education year, mean (SD) Job status, n (%) Employed full-time Employed part-time On medical leave Housewife Student Retired Not employed Missing Marital status, n (%) Single, never married Single, divorced or separated Single, widowed Married Missing Clinical characteristics Age of onset at first episode, years, mean (SD) Number of previous depressive episodes, mean (SD), median, range Length of current episode, months, mean (SD), range Inpatient status at baseline, n (%)
43.5 (12.4) 250 (50.8) 14.3 (2.2) 202 (40.9) 38 (7.7) 137 (27.7) 57 (11.5) 1 (0.2) 7 (1.4) 51 (10.3) 1 (0.2) 140 (28.3) 43 (8.7) 20 (4.0) 289 (58.5) 2 (0.4) 38.7 (13.6) 2.5 (4.3), 1, 1–49 4.5 (9.1), 1–146 2 (0.4)
111
Journal of Affective Disorders 250 (2019) 108–113
T. Akechi, et al.
Table 2 Predictors of relapse of major depression after remission by acute phase pharmacotherapy-multivariate logistic regression analysis with stepwise backward elimination method. characteristics
Beta
SE
Odds ratio
95% CI
p-value
Depressive mood score (item 1) of PHQ-9 score at week 0 Suicidal ideation score (item 9) of PHQ-9 score at week 0 Total PHQ-9 score at week 9 Side effects at week 9
−0.44 0.36 0.49 0.10
0.21 0.16 0.11 0.05
0.64 1.44 1.63 1.11
0.43–0.96 1.06–1.96 1.23–2.04 1.01–1.21
0.03 0.02 <0.001 0.03
with an antidepressant other than sertraline, or chosen to augment with or switch to a drug other than mirtazapine, and excluded from the analysis subjects who remitted early owing to sertraline administration. Second, as our primary endpoint was 25 weeks post-pharmacotherapy, we did not measure transient relapse before 25 weeks. The assessment of relapse at 25 weeks may have been too short. Third we used the stepwise regression method to explore potential predictors of relapse. This may reduce the generalizability of the results. Fourth, considering that different definition of relapse showed the different results, the findings obtained may be sensitive to the specific definition of relapse. Finally, as this was a secondary analysis of our previous study data, we may not have assessed all potentially relevant predictors including treatment resistance (e.g., number of antidepressants failed in past and current episodes). In conclusion, approximately one-seventh of subjects experienced some depression relapse within 4 months, even after remission. Lower depressive mood and higher suicidal ideation upon development of the current depression episode, the presence of residual symptoms, and greater severity of side effects at remission may be relevant predictors of subsequent depression relapse.
suggests that milder depressive mood before the initiation of depression treatment may influence the intensity and/or continuity of antidepressant treatment, as depressive mood is a core symptom of depression (e.g., mild depressive mood may be associated with lower intensity and/or shorter duration of antidepressant treatment). Interestingly, considering that the SUN☺D trial excluded subjects at high risk for imminent suicide (as judged by the treating physician), even mild to moderate suicidal ideation may predict subsequent relapse. Patients with suicidal ideation may benefit from longer antidepressant treatment to prevent early depression relapse. Our findings demonstrate that greater severity of side effects at remission can be a useful predictor of relapse. This suggests clinical relevance of continuous check of side effects experienced by patients even though treatment response is good. Although it is well known that demographic factors contribute to the development of major depression (Kessler et al., 2003), clinical factors rather than demographic characteristics are more important determinants of depression outcome. (Hardeveld et al., 2010) On the other hand, because the findings with regard to the milder depressive mood and presence of suicidal ideation in the current depressive episode, and greater severity of side effects at remission as predictors of relapse were not firmly replicated in our additional analyses, further studies are needed to confirm the findings. Our findings also indicate that the presence of residual symptoms at remission is a significant predictor of relapse. This is consistent with many clinical practice guidelines, which report that residual symptoms predict depression relapse (Bauer et al., 2015; Kennedy et al., 2016; NICE, 2009 last updated April 2016). Our findings furthermore support the relevance of residual symptoms as predictors of relapse. Other clinical factors, such as age of onset at first depressive episode, number of previous episodes, length of current episode, and adverse pharmacotherapy events (as measured by the FIBSER) were not significant predictors, which supports previous study findings (Bauer et al., 2015; Hardeveld et al., 2010; Kennedy et al., 2016; NICE, 2009, last updated April 2016). Although most clinical guidelines and research studies have consistently shown an association between number of previous depressive episodes and subsequent relapse (Bauer et al., 2015; Hardeveld et al., 2010; Kennedy et al., 2016; NICE, 2009, last updated April 2016), the present findings did not. Most of our subjects did not have recurrent depressive episodes (the median number of previous depressive episodes was one, as shown in Table 1); therefore, differences between study subjects may explain these contrasting findings. Our results suggest that depression is a recurrent and chronic disorder, as indicated by many previous studies (Colman et al., 2011; Hardeveld et al., 2010). However, although it is clear that the main goal of the post-remission treatment stage is to prevent recurrence, this treatment stage is challenging owing to limited evidence to guide clinical management (Wang et al., 2014). Although continuous antidepressant treatment can benefit many depressive patients (Geddes et al., 2003), further clinical studies may be needed to identify treatments that improve the subsequent course of depression. For example, possible predictors could be taken into account during the initial phase of antidepressant treatment. The current study has several limitations. First, we do not know whether the antidepressant treatment regime affected the results. For example, the results may have been different had we chosen to begin
Declaration of interests The authors have completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Dr Akechi has received lectures fees from Astellas, AstraZeneca, Daiichi-Sankyo, Dainippon-Sumitomo, Eizai, Hisamitsu, Lilly, MSD, Meiji-seika Pharma, Mochida, Novartis, Otsuka, Shionogi, TanabeMitsubishi, Terumo, and Yoshitomi. Dr Akechi has received research funds from Daiichi-Sankyo, Eizai, MSD, Pfizer, Novartis, and TanabeMitsubishi. Dr. Mantani has received lecture fees from Mochida, Eli Lilly and Meiji. Dr. Hirota has received lectures fees from Dainippon-Sumitomo, Eizai, Jansen Pharm, Meiji-seika Pharma, Otsuka, Shionogi, TanabeMitsubishi, and Yoshitomi. Dr. Shimodera has received lecture fees from Otsuka, MSD, Meiji, Eli Lilly, Mochida, Tsumura, and Shionogi. Dr. Yamada declares no conflicts of interest associated with this manuscript. Dr. Inagaki has received lectures fees from Meiji, Mochida, Takeda, Novartis, Yoshitomi, and Pfizer and personal fees from Technomics. Dr Inagaki has received research funds from Novartis and Otsuka. His institute has received research funds from Eisai, Astellas, DainipponSumitomo, Pfizer, Daiichi-Sankyo, and Takeda. Dr. Watanabe has received royalties from Sogensha, Paquet and Akatsuki. Dr. Kato has received lectures fee from Tanabe-Mitsubishi. Dr. Furukawa has received lecture fees from Meiji, MitsubishiTanabe, MSD and Pfizer. He has received research support from Mitsubishi-Tanabe. Funding The study was funded by the Ministry of Health, Labor and Welfare, 112
Journal of Affective Disorders 250 (2019) 108–113
T. Akechi, et al.
Japan (H-22-Seishin-Ippan-008) from April 2010 through March 2012, and thereafter by the Japan Foundation for Neuroscience and Mental Health (JFNMH). The JFNMH received donations from Asahi Kasei, EliLilly, GSK, Janssen, MSD, Meiji Seika, Mochida, Otsuka, Pfizer, Shionogi, Taisho, and Tanabe-Mitsubishi. The Ministry of Health, Labor and Welfare, Japan and the Japan Foundation for Neuroscience and Mental Health had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.
Depression: SUN(^_^)D study protocol. Trials 12, 116. Geddes, J.R., Carney, S.M., Davies, C., Furukawa, T.A., Kupfer, D.J., Frank, E., Goodwin, G.M., 2003. Relapse prevention with antidepressant drug treatment in depressive disorders: a systematic review. Lancet 361, 653–661. Hardeveld, F., Spijker, J., De Graaf, R., Nolen, W.A., Beekman, A.T., 2010. Prevalence and predictors of recurrence of major depressive disorder in the adult population. Acta Psychiatr. Scand. 122, 184–191. Harkness, K.L., Bagby, R.M., Kennedy, S.H., 2012. Childhood maltreatment and differential treatment response and recurrence in adult major depressive disorder. J. Consult. Clin. Psychol. 80, 342–353. Jang, S., Jung, S., Pae, C., Kimberly, B.P., Craig Nelson, J., Patkar, A.A., 2013. Predictors of relapse in patients with major depressive disorder in a 52-week, fixed dose, double blind, randomized trial of selegiline transdermal system (STS). J. Affect. Disord. 151, 854–859. Kanai, T., Takeuchi, H., Furukawa, T.A., Yoshimura, R., Imaizumi, T., Kitamura, T., Takahashi, K., 2003. Time to recurrence after recovery from major depressive episodes and its predictors. Psychol. Med. 33, 839–845. Kato, T., Furukawa, T.A., Mantani, A., Kurata, K., Kubouchi, H., Hirota, S., Sato, H., Sugishita, K., Chino, B., Itoh, K., Ikeda, Y., Shinagawa, Y., Kondo, M., Okamoto, Y., Fujita, H., Suga, M., Yasumoto, S., Tsujino, N., Inoue, T., Fujise, N., Akechi, T., Yamada, M., Shimodera, S., Watanabe, N., Inagaki, M., Miki, K., Ogawa, Y., Takeshima, N., Hayasaka, Y., Tajika, A., Shinohara, K., Yonemoto, N., Tanaka, S., Zhou, Q., Guyatt, G.H., 2018. Optimising first- and second-line treatment strategies for untreated major depressive disorder - the SUND study: a pragmatic, multi-centre, assessor-blinded randomised controlled trial. BMC Med. 16, 103. Kawakami, N., Shimizu, H., Haratani, T., Iwata, N., Kitamura, T., 2004. Lifetime and 6month prevalence of DSM-III-R psychiatric disorders in an urban community in Japan. Psychiatry Res. 121, 293–301. Kennedy, S.H., Lam, R.W., McIntyre, R.S., Tourjman, S.V., Bhat, V., Blier, P., Hasnain, M., Jollant, F., Levitt, A.J., MacQueen, G.M., McInerney, S.J., McIntosh, D., Milev, R.V., Muller, D.J., Parikh, S.V., Pearson, N.L., Ravindran, A.V., Uher, R., 2016. Canadian Network for Mood and Anxiety Treatments (CANMAT) 2016 Clinical Guidelines for the Management of Adults with Major Depressive Disorder: section 3. Pharmacological Treatments. Can. J. Psychiatry Revue canadienne de psychiatrie 61, 540–560. Kessler, R.C., Berglund, P., Demler, O., Jin, R., Koretz, D., Merikangas, K.R., Rush, A.J., Walters, E.E., Wang, P.S., 2003. The epidemiology of major depressive disorder: results from the National Comorbidity Survey Replication (NCS-R). JAMA 289, 3095–3105. Kroenke, K., Spitzer, R.L., Williams, J.B., 2001. The PHQ-9: validity of a brief depression severity measure. J. Gen. Intern. Med. 16, 606–613. Maj, M., Veltro, F., Pirozzi, R., Lobrace, S., Magliano, L., 1992. Pattern of recurrence of illness after recovery from an episode of major depression: a prospective study. Am. J. Psychiatry 149, 795–800. Melartin, T.K., Rytsala, H.J., Leskela, U.S., Lestela-Mielonen, P.S., Sokero, T.P., Isometsa, E.T., 2004. Severity and comorbidity predict episode duration and recurrence of DSM-IV major depressive disorder. J. Clin. Psychiatry 65, 810–819. Mueller, T.I., Leon, A.C., Keller, M.B., Solomon, D.A., Endicott, J., Coryell, W., Warshaw, M., Maser, J.D., 1999. Recurrence after recovery from major depressive disorder during 15 years of observational follow-up. Am. J. Psychiatry 156, 1000–1006. NICE, 2009. last updated April 2016. National institute for health and care excellence depression in adults: recognition and management. clinical guideline. Pinto-Meza, A., Serrano-Blanco, A., Penarrubia, M.T., Blanco, E., Haro, J.M., 2005. Assessing depression in primary care with the PHQ-9: can it be carried out over the telephone? J. Gen. Intern. Med. 20, 738–742. Ramana, R., Paykel, E.S., Cooper, Z., Hayhurst, H., Saxty, M., Surtees, P.G., 1995. Remission and relapse in major depression: a two-year prospective follow-up study. Psychol. Med. 25, 1161–1170. Rush, A.J., Trivedi, M.H., Wisniewski, S.R., Stewart, J.W., Nierenberg, A.A., Thase, M.E., Ritz, L., Biggs, M.M., Warden, D., Luther, J.F., Shores-Wilson, K., Niederehe, G., Fava, M., 2006. Bupropion-SR, sertraline, or venlafaxine-XR after failure of SSRIs for depression. N. Engl. J. Med. 354, 1231–1242. Shimodera, S., Kato, T., Sato, H., Miki, K., Shinagawa, Y., Kondo, M., Fujita, H., Morokuma, I., Ikeda, Y., Akechi, T., Watanabe, N., Yamada, M., Inagaki, M., Yonemoto, N., Furukawa, T.A., 2012. The first 100 patients in the SUN(^_^)D trial (strategic use of new generation antidepressants for depression): examination of feasibility and adherence during the pilot phase. Trials 13, 80. Spitzer, R.L., Kroenke, K., Williams, J.B., 1999. Validation and utility of a self-report version of PRIME-MD: the PHQ primary care study. Primary Care Evaluation of Mental Disorders. Patient Health Questionnaire. JAMA 282, 1737–1744. Spitzer, R.L., Williams, J.B., Kroenke, K., Linzer, M., deGruy, F.V., Hahn, S.R., Brody, D., Johnson, J.G., 1994. Utility of a new procedure for diagnosing mental disorders in primary care. The PRIME-MD 1000 study. JAMA 272, 1749–1756. Wang, J.L., Patten, S., Sareen, J., Bolton, J., Schmitz, N., MacQueen, G., 2014. Development and validation of a prediction algorithm for use by health professionals in prediction of recurrence of major depression. Depress. Anxiety 31, 451–457. WHO, 2008. The global burden of disease: 2004 update. Williams, N., Simpson, A.N., Simpson, K., Nahas, Z., 2009. Relapse rates with long-term antidepressant drug therapy: a meta-analysis. Hum. Psychopharmacol. 24, 401–408. Wolff, R., Whiting, P., Mallett, S., Riley, R., Westwood, M., Kleijnen, J., Moons, K., 2015. PROBAST-Prediction model risk of bias assessment tool. In: G-I-N Conference.
Authors' contributions Study concept and design: Tatsuo Akechi Acquisition, analysis or interpretation of data: Tatsuo Akechi, Akio Mantani, Ken'ichi Kurata, Susumu Hirota, Shinji Shimodera, Mitsuhiko Yamada, Masatoshi Inagaki, Norio Watanabe, Tadashi Kato, Toshi A. Furukawa Drafting of the manuscript: Akechi T Critical revision of the manuscript for important intellectual content: Tatsuo Akechi, Akio Mantani, Ken'ichi Kurata, Susumu Hirota, Shinji Shimodera, Mitsuhiko Yamada, Masatoshi Inagaki, Norio Watanabe, Tadashi Kato, Toshi A. Furukawa Statistical analysis: Tatsuo Akechi Obtained funding: Toshi A. Furukawa, Mitsuhiko Yamada Acknowledgments We thank Ms. K. Maki, K. Kobori, Y. Yanase, N. Kato, A. Nomura, and K. Tojima for their support. We thank Diane Williams, Ph.D., from Edanz Group (www.edanzediting.com/ac) for editing a draft of this manuscript. We also thank Dr. S. Osaga for his statistical support. Supplementary materials Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.jad.2019.03.004. References APA, 2010. Practice guideline for the treatment of patients with major depressive disorder, third edition. Am. J. Psychiatry 167 (Suppl), 1–118. Bauer, M., Pfennig, A., Severus, E., Whybrow, P.C., Angst, J., Moller, H.J., 2013. World Federation of Societies of Biological Psychiatry (WFSBP) guidelines for biological treatment of unipolar depressive disorders, part 1: update 2013 on the acute and continuation treatment of unipolar depressive disorders. World J. Biol. Psychiatry 14, 334–385. Bauer, M., Severus, E., Kohler, S., Whybrow, P.C., Angst, J., Moller, H.J., 2015. World Federation of Societies of Biological Psychiatry (WFSBP) guidelines for biological treatment of unipolar depressive disorders. part 2: maintenance treatment of major depressive disorder-update 2015. World J. Biol. Psychiatry 16, 76–95. Cho, H.J., Lavretsky, H., Olmstead, R., Levin, M.J., Oxman, M.N., Irwin, M.R., 2008. Sleep disturbance and depression recurrence in community-dwelling older adults: a prospective study. Am. J. Psychiatry 165, 1543–1550. Cleare, A., Pariante, C.M., Young, A.H., Anderson, I.M., Christmas, D., Cowen, P.J., Dickens, C., Ferrier, I.N., Geddes, J., Gilbody, S., Haddad, P.M., Katona, C., Lewis, G., Malizia, A., McAllister-Williams, R.H., Ramchandani, P., Scott, J., Taylor, D., Uher, R., 2015. Evidence-based guidelines for treating depressive disorders with antidepressants: a revision of the 2008 British Association for Psychopharmacology guidelines. J. Psychopharmacol. 29, 459–525. Colman, I., Naicker, K., Zeng, Y., Ataullahjan, A., Senthilselvan, A., Patten, S.B., 2011. Predictors of long-term prognosis of depression. CMAJ 183, 1969–1976. Cuijpers, P., Hollon, S.D., van Straten, A., Bockting, C., Berking, M., Andersson, G., 2013. Does cognitive behaviour therapy have an enduring effect that is superior to keeping patients on continuation pharmacotherapy? A meta-analysis. BMJ Open 3. Furukawa, T.A., 2010. Assessment of mood: guides for clinicians. J. Psychosom. Res. 68, 581–589. Furukawa, T.A., Akechi, T., Shimodera, S., Yamada, M., Miki, K., Watanabe, N., Inagaki, M., Yonemoto, N., 2011. Strategic Use of New generation antidepressants for
113
Contents lists available at ScienceDirect
Journal of Affective Disorders journal homepage: www.elsevier.com/locate/jad
Research paper
Predicting relapse in major depression after successful initial pharmacological treatment
T
⁎
Tatsuo Akechia, , Akio Mantanib, Ken'ichi Kuratac, Susumu Hirotad, Shinji Shimoderae, Mitsuhiko Yamadaf, Masatoshi Inagakig, Norio Watanabeh, Tadashi Katoi, Toshi A. Furukawah, for the SUND Investigators a
Department of Psychiatry and Cognitive-Behavioral Medicine, Nagoya City University Graduate School of Medical Sciences, Nagoya, Japan Mantani Mental Clinic, Hiroshima, Japan c Kabe Mental Health Clinic, Hiroshima, Japan d Hirota Clinic, Hiroshima, Japan e Department of Neuropsychiatry, Kochi Medical School, Kochi, Japan f Department of Neuropsychopharmacology, National Institute of Mental Health, National Center of Neurology and Psychiatry, Tokyo, Japan g Department of Neuropsychiatry, Okayama University Hospital, Okayama, Japan h Kyoto University Graduate School of Medicine, School of Public Health Department of Health Promotion of Human Behavior, Kyoto, Japan i Aratama Kokorono Clinic, Nagoya, Japan b
A R T I C LE I N FO
A B S T R A C T
Keywords: Major depression Antidepressant Relapse Predictor
Background: Identifying the predictors of relapse could help to develop more individualized treatment strategies for major depression. The study aim was to explore predictors of depression relapse after remission using data from our previous multicenter randomized practical trial of patients with major depression. Methods: Our cohort comprised subjects with Patient Health Questionnaire (PHQ-9) scores less than 5 after antidepressant treatment for 9 weeks. Relapse was defined as a PHQ-9 score of 5 or more at week 25. We examined patient demographic and clinical characteristics at baseline (age, sex education, job status, marital status, onset age at first depressive episode, number of previous episodes, length of current episode, scores on the nine PHQ-9 criteria at week 0) and Frequency, Intensity, and Burden of Side Effects Rating Scale and PHQ-9 total scores at week 9 (residual symptoms) as potential predictors of depression relapse at week 25. Results: Of 494 patients remitted at week 9, 71 (14.4%) experienced relapse at week 25. Logistic regression analysis showed that lower PHQ-9 depressive mood score at week 0, higher suicidal ideation score at week 0, and total PHQ-9 score at week 9, and greater severity of side effects at week 9 were significant predictors. On the other hand, when relapse was defined as a PHQ-9 score of 10 or more at week 25, there were no significant predictors. Limitations: There may be other important predictors that this study failed to identify and the findings obtained may be sensitive to the specific definition of relapse. Conclusions: Approximately one-seventh of subjects who remitted after 2 months of acute-phase treatment experienced depression relapse within 4 months of remission. Lower depressive mood and higher suicidal ideation upon development of the current depression episode, the presence of residual symptoms, and greater severity of side effects at remission may predict subsequent depression relapse.
1. Introduction Major depression is a leading cause of disability and its burden is expected to increase during the next 10 years (WHO, 2008). As major depression is one of the most prevalent psychiatric disorders in most
developed countries (Kawakami et al., 2004; Kessler et al., 2003), the establishment of effective health strategies to manage and prevent depression is important. Although numerous treatments are available for major depression and meta-analysis indicates that cognitive behavioral therapy has an
⁎ Correspondence to: Department of Psychiatry and Cognitive-Behavioral Medicine, Nagoya City University Graduate School of Medical Sciences, 1 Kawasumi, Mizuho-cho, Mizuho-ku, Nagoya, Aichi 467-8601, Japan. E-mail address: [email protected] (T. Akechi).
https://doi.org/10.1016/j.jad.2019.03.004 Received 9 August 2018; Received in revised form 12 February 2019; Accepted 3 March 2019 Available online 05 March 2019 0165-0327/ © 2019 Elsevier B.V. All rights reserved.
Journal of Affective Disorders 250 (2019) 108–113
T. Akechi, et al.
used the minimization method, adjusting for site, whether 50% or greater reduction on PHQ-9 score had been achieved, and whether patients reported moderate or greater impairment owing to side effects. In Step 2, in the continue-sertraline arm, sertraline was administered at 50 or 100 mg/day according to the initial randomization (The maximum sertraline dose was 100 mg daily, half of the recommend maximum dose in the U.S because the maximum of the licensed dosage in Japan is 100 mg/day). In the augmentation arm, sertraline was continued and mirtazapine was added at between 7.5 and 45 mg/day at the discretion of the study psychiatrist. In the mirtazapine switch arm, mirtazapine at between 7.5 mg and 45 mg/day was administered; sertraline was tapered and discontinued by week 7. Co-administration of benzodiazepines, but not of other antidepressants, antipsychotics, or mood stabilizers, was allowed up to week 9. After week 9, treatment was at the physician's discretion. All treatments were open-label. Institutional review boards at each participating site approved the study. The lead site, Kyoto University, hosted the regulatory and data management cores. An independent data monitoring committee oversaw the trial. All participants provided written informed consent. This study used the data of 494 subjects who remitted at week 9 and could be followed-up at week 25.
enduring effect following termination of the acute treatment (Cuijpers et al., 2013), pharmacotherapy is one of the most widely used standard therapies in clinical practice and is mentioned in most clinical practice guidelines. Most guidelines recommend at least 6–9 months of continuous treatment for maintenance therapy (preventing relapse and recurrence) after remission(APA, 2010; Bauer et al., 2013, 2015; Cleare et al., 2015; Geddes et al., 2003; Kennedy et al., 2016; NICE, 2009, last updated April 2016), because many patients (typically 20%–40%) experience depression relapse within 1 year even after depression remission (Geddes et al., 2003; Kanai et al., 2003; Maj et al., 1992; Melartin et al., 2004; Mueller et al., 1999; Ramana et al., 1995). Metaanalysis demonstrates that prophylactic antidepressant drug therapy is useful for preventing future relapses (Williams et al., 2009). Several guidelines and studies suggest that clinical factors (particularly number of previous depression episodes and subclinical residual symptoms) rather than patient demographic factors (e.g., age and gender) can predict depression relapse(Bauer et al., 2015; Cho et al., 2008; Hardeveld et al., 2010; Harkness et al., 2012; Jang et al., 2013; Kennedy et al., 2016; NICE, 2009, last updated April 2016). However, most of these studies have tended to use small samples, typically 60–80 subjects,(Kanai et al., 2003; Maj et al., 1992; Ramana et al., 1995) except for two studies conducted by Melartin et al. (2004) (n = 198) and Mueller et al. (1999) (n = 380). As the identification of possible predictors of relapse could help to develop more individualized management strategies for major depression, we investigated the proportion and predictors of depression relapse after remission following acute-phase antidepressant treatment using data from our previous randomized controlled clinical trial (n = 494): the SUN☺D trial (Kato et al., 2018).
2.2. Clinical assessments and outcomes Assessments were undertaken at weeks 0, 1, 3, 9, and 25 after starting sertraline in the SUN☺D trial. At week 0, the study psychiatrist made a diagnosis using a semi-structured interview, the Primary Care Evaluation of Mental Disorders (PRIME-MD) (Spitzer et al., 1994). The PRIME-MD allows the calculation of the severity of the index depression episode as measured by the PHQ-9. At weeks 1, 3, 9, and 25, assessors who were trained at and operating from the central office and blind to the treatment assignment and the timing of the assessment administered the PHQ-9 and the Frequency, Intensity, and Burden of Side Effects Rating Scale (FIBSER) by telephone (Pinto-Meza et al., 2005). We previously established the inter-rater reliability of the telephone assessment (Shimodera et al., 2012). The telephone assessments were well blinded, as assessors had no more than chance success at guessing the allocation. The PHQ-9 consists of the nine diagnostic criteria items for major depression from the Diagnostic and Statistical Manual of Mental Disorders, 4th edition (DSM-IV) (Spitzer et al., 1999). Each item is rated from 0 (Not at all) to 3 (Nearly every day). The total score range is 0–27; higher ratings indicate more severe depression. The instrument has demonstrated excellent reliability, validity, and responsiveness (Furukawa, 2010). The FIBSER comprises 7-point scales that measure the frequency, intensity, and burden of side effects. Higher ratings indicate greater severity and the total score range is 0–21 (Rush et al., 2006).
2. Methods 2.1. Participants This study represents a secondary analysis of the SUN☺D trial. The SUN☺D was a multicenter, open-label, assessor-blinded, practical randomized controlled clinical trial that involved two randomization steps to examine first- and second-line treatment strategies for untreated unipolar major depression. Specifically, the trial consisted of two steps: Step 1 was a cluster-randomized trial comparing titration up to the minimum vs. maximum of the recommended dose range among patients starting with sertraline. Step 2 addressed the impact of continuing initial treatment, augmenting with mirtazapine, or switching to mirtazapine for patients who started treatment for major depression with sertraline but had not remitted by 3 weeks. Briefly we recruited adult men and women between 25 and 75 years who had a primary diagnosis of a non-psychotic unipolar major depressive episode within the past month. All recruited persons were also not allowed to be taking any antidepressant, antipsychotic or mood stabilizer. This trial was conducted in the psychiatry departments of 48 hospitals and clinics across Japan between December 2010 and September 2015. The protocol provides additional details of the eligibility criteria (Furukawa et al., 2011) and the main findings have been published previously (Kato et al., 2018). Fig. 1 presents the study flow. First, patients with non-psychotic unipolar major depression for whom the treating physician had judged that starting treatment with sertraline was indicated (week 0) were followed-up and patients who were tolerant of sertraline 25 mg/day for 3–16 days were recruited. In Step 1, participants received first-line treatment with sertraline to be titrated up to either 50 mg/day or 100 mg/day by week 3 according to cluster randomization by site. In Step 2, participants who had not reached remission, defined as a score of 4 or less on the Patient Health Questionnaire (PHQ-9) at week 3, were randomized 1:1:1 (using a web-based central computer system) to continue with sertraline, to augment sertraline with mirtazapine, or to switch to mirtazapine (Fig. 1). The Step 2 individual randomization
2.3. Definition of cases with remission and predictors investigated We defined the remission of depression as a PHQ-9 score less than 5, following a previous study by Kroenke et al. (2001). They provided the following rules of thumb for interpreting continuous PHQ-9 scores: 0–4 no depression; 5–9 mild depression; 10–14 moderate depression; 15–19 moderately severe depression; 20+ severe depression. We divided subjects into two groups: those who maintained remission at week 25 vs. those who relapsed and whose PHQ-9 score was greater than 4 at week 25. In this study, we examined the following variables as potential predictors: PHQ-9 scores for each of the nine criteria (evaluated by PRIME-MD) at week 0; patient demographic and clinical characteristics (age, gender, education, job status, marital status, age of onset at first depressive episode, number of previous depressive episodes, length of current episode at week 1), and FIBSER and total PHQ-9 scores at week 9 (residual symptoms) after initial treatment. 109
Journal of Affective Disorders 250 (2019) 108–113
T. Akechi, et al.
Assessed for eligibility (n=56,261) Non-psychotic unipolar major depression for whom the treating physician had judged that starting treatment with sertraline in indicated (week 0)
Excluded before Step 1 randomization (n=54,250) No major depressive episode (n=48,366) Declined to participate or not meeting other eligibility criteria (n=5,884)
Step 1: Cluster-randomized to sertraline 50 mg/d or 100 mg/d arms (n=2,011)
Excluded before Step 2 randomization (n=365) Unable to be followed up at week 3 (n=58) Followed up but withdrew consent to be randomized (n=76) Protocol violation (n=1) Remitted at week 3 (n=230)
Step 2: Based on assessment of depression severity and side effect at week 3, non-remitters were randomized individually (n=1,647)
Allocated to
Allocated to
Allocated to
continue
augment
switch to
with
sertraline
mirtazapine
sertraline
with
(n=558)
(n=551)
mirtazapine (n=538)
Remission (n=423) and relapse (n=71) at week 25 Fig. 1. Screening, randomization and follow-up.
2.4. Statistical analysis
2015), we avoided bivariate analysis and conducted logistic regression analysis with backward elimination (likelihood ratio) to explore predictors of worsening depression. In this logistic regression model,
In accordance with the PROBAST risk of bias tool (Wolff et al., 110
Journal of Affective Disorders 250 (2019) 108–113
T. Akechi, et al.
Furthermore, 15.4% and 24.3% of subjects discontinued all antidepressants by the end of 12 or 25 weeks, respectively. Among the subjects who did not discontinue antidepressants, 15.6% of subjects experienced relapse.
presence or absence of relapse was entered as a dependent variable, and the patient demographic and clinical characteristics mentioned above were entered as independent variables as potential predictors. In this analysis, we performed multiple imputation to complement missing data including primary outcome data in order to undertake intent-totreat-analyses. Finally to investigate the potential effect of treatment and medication discontinuation on the findings, additional analyses were conducted for controlling for treatment arm and for medication discontinuation. Variance inflation factors (VIFs) was used for multicollinearity diagnostics for the current logistic regression model. Finally additional bivariate analyses were conducted to check the results (e.g. obtained significant predictors and relapse). A p-value of less than 0.05 was used as the significance level in all statistical analyses; all p-values reported were two-tailed. All statistical procedures except multiple imputation done by R were conducted using IBM SPSS Statistics version 25 software for Windows (SPSS Inc., NY, 2017).
3.2. Predictors of worsening depression Since there were total of 10 missing data of follow-up at week 25, these were complemented by using multiple imputation. The stepwise logistic regression analysis showed that lower PHQ-9 depressive mood score at week 0, higher suicidal ideation score at week 0, total PHQ-9 score at week 9, and FIBSER total score at week 9 independently and significantly predicted depression relapse. PHQ-9 scores for the nine criteria at week 0 (except depressive mood and suicidal ideation) and other clinical factors, including patient demographic and clinical characteristics at week 1 (age, gender, education, job status, marital status, age of onset at first depressive episode, number of previous depressive episodes, length of current episode) did not significantly predict depression relapse (Table 2). Additional bivariate analyses showed that lower PHQ-9 depressive mood score at week 0 is borderline significance (p = 0.07) and higher suicidal ideation score at week 0, total PHQ-9 score at week 9, and FIBSER total score at week 9 are significant. Calculation VIFs for multicollinearity diagnostics for the current logistic regression model showed that all VIFs were not high (e.g. less than 4.1) and this suggests validity of the analysis. Even after controlling for treatment arm and for medication discontinuation, the findings are same (data not show,). In addition, there were no interactions of predictors of relapse with treatment arm and discontinuation (data not shown). Finally, when relapse was defined as a PHQ-9 score of 10 or more at week 25, there were no significant predictors in this analysis partly because the number of relapse cases is decreased (N = 11).
3. Results 3.1. Participants Fig. 1 shows the screening, randomization, and follow-up of study participants. As shown, 1647 had not remitted by week 3 and were individually randomized to continue sertraline (n = 551), augment sertraline with mirtazapine (n = 538), or switch to mirtazapine (n = 558). Of these, 494 patients remitted at week 9 and completed week 25 assessments (Table 1). The mean subject age was early 40 s and half the subjects were female. Approximately 40% worked full-time and more than half were married. The median number of previous depressive episodes was one and the number of past episodes was variable. At week 25, 71 subjects (14.4% [95% confidence interval: 12% to 18%]) experienced some degree of relapse and 423 subjects (85.6%) were still in remission. The number and percentage of relapsed subjects were as follows: mild depression (n = 60, 12.1%); moderate depression (n = 8, 1.6%); moderately severe depression (n = 2, 0.4%); and severe depression (n = 1, 0.2%). Regarding cessation of antidepressant treatment, 33.8% and 50.3% of subjects discontinued the antidepressants assigned at Step 2 randomization by 12 and 25 weeks, respectively.
4. Discussion The current study demonstrated that approximately one-seventh of subjects experienced some relapse of their depression within 4 months even after remission following 2 months of acute-phase antidepressant treatment. Most relapses were mild. Our findings also suggested that suicidal ideation upon development of the current depression episode and presence of residual symptoms at remission could predict subsequent depression relapse after remission. The proportion of participants that experienced relapse is generally consistent with findings from previous reports (Kanai et al., 2003; Maj et al., 1992; Melartin et al., 2004; Mueller et al., 1999), although the present study population and setting differed from those in previous studies. Considering that 24.3% of our subjects discontinued all antidepressant use by the end of 25 weeks, and the strong evidence that continuous antidepressant treatment can reduce the risk of relapse, (Geddes et al., 2003) there is a clear need for maintenance therapy for major depression. Many patients suffer from early depression relapse; therefore, physicians should treat patients with depression carefully and monitor them even after they have achieved remission. In addition, most guidelines clearly state the need for continuous treatment, typically for more than 6 months, although this depends on patients’ individual risk of relapse (Bauer et al., 2015; Kennedy et al., 2016; NICE, 2009 last updated April 2016). The implementation and dissemination of a guideline treatment strategy for relapse prevention is recommended. To the best of our knowledge, there are no studies indicating that milder depressive mood, presence of suicidal ideation in the current depressive episode, and greater severity of side effects at remission are significant predictors of subsequent depression relapse. It is not clear why milder depressive mood in the current depression episode is a potential predictor of future relapse. However, our clinical experience
Table 1 Characteristics of the subjects: remission at week 9. All subjects (n = 494) Demographic characteristics* Age, year mean (SD) Female sex, n (%) Education year, mean (SD) Job status, n (%) Employed full-time Employed part-time On medical leave Housewife Student Retired Not employed Missing Marital status, n (%) Single, never married Single, divorced or separated Single, widowed Married Missing Clinical characteristics Age of onset at first episode, years, mean (SD) Number of previous depressive episodes, mean (SD), median, range Length of current episode, months, mean (SD), range Inpatient status at baseline, n (%)
43.5 (12.4) 250 (50.8) 14.3 (2.2) 202 (40.9) 38 (7.7) 137 (27.7) 57 (11.5) 1 (0.2) 7 (1.4) 51 (10.3) 1 (0.2) 140 (28.3) 43 (8.7) 20 (4.0) 289 (58.5) 2 (0.4) 38.7 (13.6) 2.5 (4.3), 1, 1–49 4.5 (9.1), 1–146 2 (0.4)
111
Journal of Affective Disorders 250 (2019) 108–113
T. Akechi, et al.
Table 2 Predictors of relapse of major depression after remission by acute phase pharmacotherapy-multivariate logistic regression analysis with stepwise backward elimination method. characteristics
Beta
SE
Odds ratio
95% CI
p-value
Depressive mood score (item 1) of PHQ-9 score at week 0 Suicidal ideation score (item 9) of PHQ-9 score at week 0 Total PHQ-9 score at week 9 Side effects at week 9
−0.44 0.36 0.49 0.10
0.21 0.16 0.11 0.05
0.64 1.44 1.63 1.11
0.43–0.96 1.06–1.96 1.23–2.04 1.01–1.21
0.03 0.02 <0.001 0.03
with an antidepressant other than sertraline, or chosen to augment with or switch to a drug other than mirtazapine, and excluded from the analysis subjects who remitted early owing to sertraline administration. Second, as our primary endpoint was 25 weeks post-pharmacotherapy, we did not measure transient relapse before 25 weeks. The assessment of relapse at 25 weeks may have been too short. Third we used the stepwise regression method to explore potential predictors of relapse. This may reduce the generalizability of the results. Fourth, considering that different definition of relapse showed the different results, the findings obtained may be sensitive to the specific definition of relapse. Finally, as this was a secondary analysis of our previous study data, we may not have assessed all potentially relevant predictors including treatment resistance (e.g., number of antidepressants failed in past and current episodes). In conclusion, approximately one-seventh of subjects experienced some depression relapse within 4 months, even after remission. Lower depressive mood and higher suicidal ideation upon development of the current depression episode, the presence of residual symptoms, and greater severity of side effects at remission may be relevant predictors of subsequent depression relapse.
suggests that milder depressive mood before the initiation of depression treatment may influence the intensity and/or continuity of antidepressant treatment, as depressive mood is a core symptom of depression (e.g., mild depressive mood may be associated with lower intensity and/or shorter duration of antidepressant treatment). Interestingly, considering that the SUN☺D trial excluded subjects at high risk for imminent suicide (as judged by the treating physician), even mild to moderate suicidal ideation may predict subsequent relapse. Patients with suicidal ideation may benefit from longer antidepressant treatment to prevent early depression relapse. Our findings demonstrate that greater severity of side effects at remission can be a useful predictor of relapse. This suggests clinical relevance of continuous check of side effects experienced by patients even though treatment response is good. Although it is well known that demographic factors contribute to the development of major depression (Kessler et al., 2003), clinical factors rather than demographic characteristics are more important determinants of depression outcome. (Hardeveld et al., 2010) On the other hand, because the findings with regard to the milder depressive mood and presence of suicidal ideation in the current depressive episode, and greater severity of side effects at remission as predictors of relapse were not firmly replicated in our additional analyses, further studies are needed to confirm the findings. Our findings also indicate that the presence of residual symptoms at remission is a significant predictor of relapse. This is consistent with many clinical practice guidelines, which report that residual symptoms predict depression relapse (Bauer et al., 2015; Kennedy et al., 2016; NICE, 2009 last updated April 2016). Our findings furthermore support the relevance of residual symptoms as predictors of relapse. Other clinical factors, such as age of onset at first depressive episode, number of previous episodes, length of current episode, and adverse pharmacotherapy events (as measured by the FIBSER) were not significant predictors, which supports previous study findings (Bauer et al., 2015; Hardeveld et al., 2010; Kennedy et al., 2016; NICE, 2009, last updated April 2016). Although most clinical guidelines and research studies have consistently shown an association between number of previous depressive episodes and subsequent relapse (Bauer et al., 2015; Hardeveld et al., 2010; Kennedy et al., 2016; NICE, 2009, last updated April 2016), the present findings did not. Most of our subjects did not have recurrent depressive episodes (the median number of previous depressive episodes was one, as shown in Table 1); therefore, differences between study subjects may explain these contrasting findings. Our results suggest that depression is a recurrent and chronic disorder, as indicated by many previous studies (Colman et al., 2011; Hardeveld et al., 2010). However, although it is clear that the main goal of the post-remission treatment stage is to prevent recurrence, this treatment stage is challenging owing to limited evidence to guide clinical management (Wang et al., 2014). Although continuous antidepressant treatment can benefit many depressive patients (Geddes et al., 2003), further clinical studies may be needed to identify treatments that improve the subsequent course of depression. For example, possible predictors could be taken into account during the initial phase of antidepressant treatment. The current study has several limitations. First, we do not know whether the antidepressant treatment regime affected the results. For example, the results may have been different had we chosen to begin
Declaration of interests The authors have completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Dr Akechi has received lectures fees from Astellas, AstraZeneca, Daiichi-Sankyo, Dainippon-Sumitomo, Eizai, Hisamitsu, Lilly, MSD, Meiji-seika Pharma, Mochida, Novartis, Otsuka, Shionogi, TanabeMitsubishi, Terumo, and Yoshitomi. Dr Akechi has received research funds from Daiichi-Sankyo, Eizai, MSD, Pfizer, Novartis, and TanabeMitsubishi. Dr. Mantani has received lecture fees from Mochida, Eli Lilly and Meiji. Dr. Hirota has received lectures fees from Dainippon-Sumitomo, Eizai, Jansen Pharm, Meiji-seika Pharma, Otsuka, Shionogi, TanabeMitsubishi, and Yoshitomi. Dr. Shimodera has received lecture fees from Otsuka, MSD, Meiji, Eli Lilly, Mochida, Tsumura, and Shionogi. Dr. Yamada declares no conflicts of interest associated with this manuscript. Dr. Inagaki has received lectures fees from Meiji, Mochida, Takeda, Novartis, Yoshitomi, and Pfizer and personal fees from Technomics. Dr Inagaki has received research funds from Novartis and Otsuka. His institute has received research funds from Eisai, Astellas, DainipponSumitomo, Pfizer, Daiichi-Sankyo, and Takeda. Dr. Watanabe has received royalties from Sogensha, Paquet and Akatsuki. Dr. Kato has received lectures fee from Tanabe-Mitsubishi. Dr. Furukawa has received lecture fees from Meiji, MitsubishiTanabe, MSD and Pfizer. He has received research support from Mitsubishi-Tanabe. Funding The study was funded by the Ministry of Health, Labor and Welfare, 112
Journal of Affective Disorders 250 (2019) 108–113
T. Akechi, et al.
Japan (H-22-Seishin-Ippan-008) from April 2010 through March 2012, and thereafter by the Japan Foundation for Neuroscience and Mental Health (JFNMH). The JFNMH received donations from Asahi Kasei, EliLilly, GSK, Janssen, MSD, Meiji Seika, Mochida, Otsuka, Pfizer, Shionogi, Taisho, and Tanabe-Mitsubishi. The Ministry of Health, Labor and Welfare, Japan and the Japan Foundation for Neuroscience and Mental Health had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.
Depression: SUN(^_^)D study protocol. Trials 12, 116. Geddes, J.R., Carney, S.M., Davies, C., Furukawa, T.A., Kupfer, D.J., Frank, E., Goodwin, G.M., 2003. Relapse prevention with antidepressant drug treatment in depressive disorders: a systematic review. Lancet 361, 653–661. Hardeveld, F., Spijker, J., De Graaf, R., Nolen, W.A., Beekman, A.T., 2010. Prevalence and predictors of recurrence of major depressive disorder in the adult population. Acta Psychiatr. Scand. 122, 184–191. Harkness, K.L., Bagby, R.M., Kennedy, S.H., 2012. Childhood maltreatment and differential treatment response and recurrence in adult major depressive disorder. J. Consult. Clin. Psychol. 80, 342–353. Jang, S., Jung, S., Pae, C., Kimberly, B.P., Craig Nelson, J., Patkar, A.A., 2013. Predictors of relapse in patients with major depressive disorder in a 52-week, fixed dose, double blind, randomized trial of selegiline transdermal system (STS). J. Affect. Disord. 151, 854–859. Kanai, T., Takeuchi, H., Furukawa, T.A., Yoshimura, R., Imaizumi, T., Kitamura, T., Takahashi, K., 2003. Time to recurrence after recovery from major depressive episodes and its predictors. Psychol. Med. 33, 839–845. Kato, T., Furukawa, T.A., Mantani, A., Kurata, K., Kubouchi, H., Hirota, S., Sato, H., Sugishita, K., Chino, B., Itoh, K., Ikeda, Y., Shinagawa, Y., Kondo, M., Okamoto, Y., Fujita, H., Suga, M., Yasumoto, S., Tsujino, N., Inoue, T., Fujise, N., Akechi, T., Yamada, M., Shimodera, S., Watanabe, N., Inagaki, M., Miki, K., Ogawa, Y., Takeshima, N., Hayasaka, Y., Tajika, A., Shinohara, K., Yonemoto, N., Tanaka, S., Zhou, Q., Guyatt, G.H., 2018. Optimising first- and second-line treatment strategies for untreated major depressive disorder - the SUND study: a pragmatic, multi-centre, assessor-blinded randomised controlled trial. BMC Med. 16, 103. Kawakami, N., Shimizu, H., Haratani, T., Iwata, N., Kitamura, T., 2004. Lifetime and 6month prevalence of DSM-III-R psychiatric disorders in an urban community in Japan. Psychiatry Res. 121, 293–301. Kennedy, S.H., Lam, R.W., McIntyre, R.S., Tourjman, S.V., Bhat, V., Blier, P., Hasnain, M., Jollant, F., Levitt, A.J., MacQueen, G.M., McInerney, S.J., McIntosh, D., Milev, R.V., Muller, D.J., Parikh, S.V., Pearson, N.L., Ravindran, A.V., Uher, R., 2016. Canadian Network for Mood and Anxiety Treatments (CANMAT) 2016 Clinical Guidelines for the Management of Adults with Major Depressive Disorder: section 3. Pharmacological Treatments. Can. J. Psychiatry Revue canadienne de psychiatrie 61, 540–560. Kessler, R.C., Berglund, P., Demler, O., Jin, R., Koretz, D., Merikangas, K.R., Rush, A.J., Walters, E.E., Wang, P.S., 2003. The epidemiology of major depressive disorder: results from the National Comorbidity Survey Replication (NCS-R). JAMA 289, 3095–3105. Kroenke, K., Spitzer, R.L., Williams, J.B., 2001. The PHQ-9: validity of a brief depression severity measure. J. Gen. Intern. Med. 16, 606–613. Maj, M., Veltro, F., Pirozzi, R., Lobrace, S., Magliano, L., 1992. Pattern of recurrence of illness after recovery from an episode of major depression: a prospective study. Am. J. Psychiatry 149, 795–800. Melartin, T.K., Rytsala, H.J., Leskela, U.S., Lestela-Mielonen, P.S., Sokero, T.P., Isometsa, E.T., 2004. Severity and comorbidity predict episode duration and recurrence of DSM-IV major depressive disorder. J. Clin. Psychiatry 65, 810–819. Mueller, T.I., Leon, A.C., Keller, M.B., Solomon, D.A., Endicott, J., Coryell, W., Warshaw, M., Maser, J.D., 1999. Recurrence after recovery from major depressive disorder during 15 years of observational follow-up. Am. J. Psychiatry 156, 1000–1006. NICE, 2009. last updated April 2016. National institute for health and care excellence depression in adults: recognition and management. clinical guideline. Pinto-Meza, A., Serrano-Blanco, A., Penarrubia, M.T., Blanco, E., Haro, J.M., 2005. Assessing depression in primary care with the PHQ-9: can it be carried out over the telephone? J. Gen. Intern. Med. 20, 738–742. Ramana, R., Paykel, E.S., Cooper, Z., Hayhurst, H., Saxty, M., Surtees, P.G., 1995. Remission and relapse in major depression: a two-year prospective follow-up study. Psychol. Med. 25, 1161–1170. Rush, A.J., Trivedi, M.H., Wisniewski, S.R., Stewart, J.W., Nierenberg, A.A., Thase, M.E., Ritz, L., Biggs, M.M., Warden, D., Luther, J.F., Shores-Wilson, K., Niederehe, G., Fava, M., 2006. Bupropion-SR, sertraline, or venlafaxine-XR after failure of SSRIs for depression. N. Engl. J. Med. 354, 1231–1242. Shimodera, S., Kato, T., Sato, H., Miki, K., Shinagawa, Y., Kondo, M., Fujita, H., Morokuma, I., Ikeda, Y., Akechi, T., Watanabe, N., Yamada, M., Inagaki, M., Yonemoto, N., Furukawa, T.A., 2012. The first 100 patients in the SUN(^_^)D trial (strategic use of new generation antidepressants for depression): examination of feasibility and adherence during the pilot phase. Trials 13, 80. Spitzer, R.L., Kroenke, K., Williams, J.B., 1999. Validation and utility of a self-report version of PRIME-MD: the PHQ primary care study. Primary Care Evaluation of Mental Disorders. Patient Health Questionnaire. JAMA 282, 1737–1744. Spitzer, R.L., Williams, J.B., Kroenke, K., Linzer, M., deGruy, F.V., Hahn, S.R., Brody, D., Johnson, J.G., 1994. Utility of a new procedure for diagnosing mental disorders in primary care. The PRIME-MD 1000 study. JAMA 272, 1749–1756. Wang, J.L., Patten, S., Sareen, J., Bolton, J., Schmitz, N., MacQueen, G., 2014. Development and validation of a prediction algorithm for use by health professionals in prediction of recurrence of major depression. Depress. Anxiety 31, 451–457. WHO, 2008. The global burden of disease: 2004 update. Williams, N., Simpson, A.N., Simpson, K., Nahas, Z., 2009. Relapse rates with long-term antidepressant drug therapy: a meta-analysis. Hum. Psychopharmacol. 24, 401–408. Wolff, R., Whiting, P., Mallett, S., Riley, R., Westwood, M., Kleijnen, J., Moons, K., 2015. PROBAST-Prediction model risk of bias assessment tool. In: G-I-N Conference.
Authors' contributions Study concept and design: Tatsuo Akechi Acquisition, analysis or interpretation of data: Tatsuo Akechi, Akio Mantani, Ken'ichi Kurata, Susumu Hirota, Shinji Shimodera, Mitsuhiko Yamada, Masatoshi Inagaki, Norio Watanabe, Tadashi Kato, Toshi A. Furukawa Drafting of the manuscript: Akechi T Critical revision of the manuscript for important intellectual content: Tatsuo Akechi, Akio Mantani, Ken'ichi Kurata, Susumu Hirota, Shinji Shimodera, Mitsuhiko Yamada, Masatoshi Inagaki, Norio Watanabe, Tadashi Kato, Toshi A. Furukawa Statistical analysis: Tatsuo Akechi Obtained funding: Toshi A. Furukawa, Mitsuhiko Yamada Acknowledgments We thank Ms. K. Maki, K. Kobori, Y. Yanase, N. Kato, A. Nomura, and K. Tojima for their support. We thank Diane Williams, Ph.D., from Edanz Group (www.edanzediting.com/ac) for editing a draft of this manuscript. We also thank Dr. S. Osaga for his statistical support. Supplementary materials Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.jad.2019.03.004. References APA, 2010. Practice guideline for the treatment of patients with major depressive disorder, third edition. Am. J. Psychiatry 167 (Suppl), 1–118. Bauer, M., Pfennig, A., Severus, E., Whybrow, P.C., Angst, J., Moller, H.J., 2013. World Federation of Societies of Biological Psychiatry (WFSBP) guidelines for biological treatment of unipolar depressive disorders, part 1: update 2013 on the acute and continuation treatment of unipolar depressive disorders. World J. Biol. Psychiatry 14, 334–385. Bauer, M., Severus, E., Kohler, S., Whybrow, P.C., Angst, J., Moller, H.J., 2015. World Federation of Societies of Biological Psychiatry (WFSBP) guidelines for biological treatment of unipolar depressive disorders. part 2: maintenance treatment of major depressive disorder-update 2015. World J. Biol. Psychiatry 16, 76–95. Cho, H.J., Lavretsky, H., Olmstead, R., Levin, M.J., Oxman, M.N., Irwin, M.R., 2008. Sleep disturbance and depression recurrence in community-dwelling older adults: a prospective study. Am. J. Psychiatry 165, 1543–1550. Cleare, A., Pariante, C.M., Young, A.H., Anderson, I.M., Christmas, D., Cowen, P.J., Dickens, C., Ferrier, I.N., Geddes, J., Gilbody, S., Haddad, P.M., Katona, C., Lewis, G., Malizia, A., McAllister-Williams, R.H., Ramchandani, P., Scott, J., Taylor, D., Uher, R., 2015. Evidence-based guidelines for treating depressive disorders with antidepressants: a revision of the 2008 British Association for Psychopharmacology guidelines. J. Psychopharmacol. 29, 459–525. Colman, I., Naicker, K., Zeng, Y., Ataullahjan, A., Senthilselvan, A., Patten, S.B., 2011. Predictors of long-term prognosis of depression. CMAJ 183, 1969–1976. Cuijpers, P., Hollon, S.D., van Straten, A., Bockting, C., Berking, M., Andersson, G., 2013. Does cognitive behaviour therapy have an enduring effect that is superior to keeping patients on continuation pharmacotherapy? A meta-analysis. BMJ Open 3. Furukawa, T.A., 2010. Assessment of mood: guides for clinicians. J. Psychosom. Res. 68, 581–589. Furukawa, T.A., Akechi, T., Shimodera, S., Yamada, M., Miki, K., Watanabe, N., Inagaki, M., Yonemoto, N., 2011. Strategic Use of New generation antidepressants for
113