Abstract
Objective:
Depression in patients with Chronic Obstructive Pulmonary Disease (COPD) has been shown to be chronic and potentially increase the burden of symptoms. Selective serotonin reuptake inhibitors (SSRIs) have anti-inflammatory and serotonergic effects that may improve lung function. We hypothesized that participants taking SSRIs have better lung function than those not taking SSRIs. The dataset was the Multi-Ethnic Study of Atherosclerosis (MESA) Lung Study. Use of SSRIs was assessed by medication inventory; spirometry was conducted following standard guidelines; dyspnea ratings were self-reported.
Results:
Contrary to our hypothesis, FEV1 was lower, and odds of dyspnea were higher among participants taking SSRIs as compared with those not taking an antidepressant; these differences persisted even with control for potential confounders including depressive symptoms. We found no evidence of a beneficial association between SSRI use and lung function or dyspnea in a large US-based cohort.
Keywords: Selective serotonin reuptake inhibitors, depression, lung function, FEVI, dyspnea, COPD
SUMMARY
Using the MESA dataset, we analyzed whether SSRIs are associated with improved lung function compared to those not prescribed SSRIs
There was no association with SSRIs and lung function
This investigation was among the first to assess the impact of antidepressants on lung function using epidemiologic methods and use a large, high-quality study
INTRODUCTION
Chronic obstructive pulmonary disease (COPD) is classified as an inflammatory group of diseases that cause airflow obstruction, respiratory muscle weakness, and exacerbations that result in reduced lung function [1]. Prior longitudinal studies have shown that depression in patients with COPD is chronic and inadequately treated, which may increase the burden of symptoms for these patients [2]. Even in those without COPD, higher inflammatory markers are associated with lower lung function [3], suggesting an association between lung function and inflammation predating the clinical development of obstructive disease. Reducing inflammation may be beneficial for subclinical disease, in addition to those with clinically defined obstructive disease.
Some antidepressants, namely selective serotonin reuptake inhibitors (SSRIs), may have off-label uses in treating COPD symptoms. SSRIs are anti-inflammatory and act upon serotonin, which is integral to central breathing control [4, 5]. Several prior studies have suggested an association of SSRI use with better lung health [6, 7] but all have been small and with limited adjustment for potential confounders. Most recently, fluvoxamine has been hypothesized to help patients with COVID-19 due to its anti-inflammatory effects [8]. The purpose of this study is to investigate whether those using prescription SSRIs have better concurrent lung function and dyspnea ratings than those not using SSRIs in a large US cohort.
Methods
This study uses data from the Multi-Ethnic Study of Atherosclerosis (MESA) Lung Study (NCT00843271). MESA is a multi-center, population-based, longitudinal study of 6,814 men and women free of clinical cardiovascular disease at ages 45–84 years at baseline (in 2000–02) across six sites in the United States [9]. The MESA Lung Study is an ancillary study of MESA to test the endothelial hypothesis of COPD and emphysema that recruited 3,965 MESA participants in 2004–06 [10]. The current analysis uses cross sectional data from Exam 4 (Sept 2005 to May 2007). Analyses were duplicated using data from Exam 5 (Apr 2010 to Dec 2011) to check for robustness.
The institutional review boards of all collaborating institutions and the National Heart Lung and Blood Institute (National Institutes of Health, Bethesda, MD, USA) approved the protocols for MESA and all procedures described herein. All participants provided written informed consent.
Outcome variables
Spirometry, including FEV1, was measured in accordance with American Thoracic Society/European Respiratory Society guidelines [11, 12] on a dry-rolling-sealed spirometer (Occupational Marketing, Inc., Houston, TX) as previously described [12]. Trained interviewers assessed dyspnea at both exams. Dyspnea is defined as a positive answer to one of the following questions: “When walking on level ground, do you get more breathless than people your own age?” or “Do you ever have to stop walking due to breathlessness?” Dyspnea was coded dichotomously (yes (1)/no (0)) in logistic regression analyses.
Exposure
Medications were assessed by medication inventory, in which participants were asked to bring in their current medications, which were recorded. The exposure of interest was use of SSRIs and, for secondary analyses, other antidepressants. Antidepressants were categorized according to the Anatomical Therapeutic Chemical classification system (SSRIs, tricyclic antidepressants, and serotonin norepinephrine reuptake inhibitors) [13]. We coded this variable as an indicator variable (yes (1)/no (0)) to indicate whether a participant self-reported use of the drug or not at the time of the visit.
Statistical analysis
Analyses were conducted in SAS 9.4 (Cary, NC). We excluded participants if they a) did not have exposure (medication) information available, b) were on more than one antidepressant or c) were missing data on any confounder included in the model. FEV1 was analyzed using multiple linear regression, dyspnea was analyzed using logistic regression. Results are shown as unadjusted, adjusted model 1 [adjusted for age, sex, race/ethnicity, height, weight, smoking status, pack-years, and depressive symptoms (using the Centers for Epidemiologic Studies Depression (CES-D) scale [14])], and adjusted model 2 (variables from model 1 + serotonin and norepinephrine reuptake inhibitors and tricyclic antidepressants).
As inhaled corticosteroids and bronchodilators are the current medications for low lung function, we performed an additional sensitivity analysis that included an indicator variable for common medications that affect lung function in the model.
Results
The MESA Lung sample consisted of 3,542 participants (Table 1). We excluded 11 participants from further analysis since they were on more than one antidepressant. The mean age of the sample was 66±10 years, 51% were female, 35% White, 16% Asian, 26% Black, and 23% Hispanic. Approximately 9% were current smokers and 39% were former smokers, with an average of 22 pack-years. The mean ± standard deviation of FEV1 was 2387±732mL; dyspnea was reported in 15% of the cohort. The average CES-D score was 8, with 14% classified as depressed according to the CES-D scale (CES-D>16).
Table 1.
Description of cohort at MESA Exam 4
| Sample characteristics | No antidepressant N=3273 |
SSRI N=178 |
SNRI N=42 |
TCA N=38 |
Total N=3542 |
|||||
|---|---|---|---|---|---|---|---|---|---|---|
| N or mean ± SD | % | N or mean ± SD | % | N or mean ± SD | % | N or mean ± SD | % | N or mean ± SD | % | |
| Age, years, mean±SD | 66±10 | 66±10 | 64±10 | 70±10 | 66±10 | |||||
| Gender | ||||||||||
| Male | 1651 | 51 | 65 | 37 | 9 | 21 | 13 | 34 | 1748 | 49 |
| Female | 1617 | 49 | 113 | 64 | 33 | 79 | 25 | 66 | 1794 | 51 |
| Race/Ethnicity | ||||||||||
| White | 1086 | 33 | 113 | 64 | 30 | 71 | 16 | 42 | 1254 | 35 |
| Asian | 559 | 17 | 4 | 2 | 3 | 7 | 1 | 3 | 567 | 16 |
| Black | 876 | 27 | 19 | 11 | 4 | 10 | 7 | 18 | 907 | 26 |
| Hispanic | 752 | 23 | 42 | 24 | 5 | 12 | 14 | 36 | 814 | 23 |
| Height, cm, mean±SD | 166±10 | 165±9 | 167±9 | 164±7 | 166±10 | |||||
| Weight, lb, mean±SD | 171±39 | 173±38 | 190±43 | 175±46 | 172±39 | |||||
| BMI, kg/m2, mean±SD | 28±5 | 29±6 | 31±6 | 29±7 | 28±5 | |||||
| Health Insurance | ||||||||||
| None | 181 | 6 | 5 | 3 | 2 | 5 | 1 | 3 | 189 | 5 |
| Medicaid | 268 | 8 | 16 | 9 | 5 | 12 | 7 | 18 | 297 | 8 |
| Medicare | 325 | 10 | 10 | 6 | 3 | 7 | 3 | 8 | 341 | 10 |
| HMO | 2147 | 66 | 128 | 72 | 30 | 71 | 19 | 50 | 2331 | 66 |
| VA/Other | 352 | 11 | 19 | 11 | 2 | 5 | 8 | 21 | 384 | 11 |
| Cigarette smoking status | ||||||||||
| Never-smokers | 1720 | 53 | 80 | 45 | 17 | 42 | 16 | 42 | 1835 | 52 |
| Former smokers | 1257 | 39 | 69 | 39 | 20 | 49 | 17 | 45 | 1371 | 39 |
| Current smokers | 286 | 9 | 29 | 16 | 4 | 10 | 5 | 13 | 325 | 9 |
| Smoking history pack-years# | 22±25 | 27±36 | 18±15 | 34±36 | 22±26 | |||||
| FEV1, mL, mean±SD | 2394±736 | 2305±711 | 2375±555 | 2301±637 | 2387±732 | |||||
| FEV1 % predicted, mean±SD | 94±18 | 90±18 | 91±12 | 97±16 | 94±18 | |||||
| COPD Stage | ||||||||||
| None | 2551 | 78 | 139 | 78 | 31 | 74 | 29 | 76 | 2757 | 78 |
| Mild | 420 | 13 | 17 | 10 | 6 | 14 | 6 | 16 | 449 | 13 |
| Moderate | 265 | 8 | 19 | 11 | 5 | 12 | 3 | 8 | 295 | 8 |
| Severe | 32 | 1 | 2 | 1 | 0 | 0 | 0 | 0 | 35 | 1 |
| Very Severe | 5 | .15 | 1 | .60 | 0 | 0 | 0 | 0 | 6 | .1 |
| Dyspnea | 465 | 14 | 42 | 24 | 14 | 33 | 14 | 37 | 541 | 15 |
| CES-D | ||||||||||
| Score, mean±SD | 7±7 | 13±11 | 11±11 | 10±8 | 8±8 | |||||
| Depressed (CES-D>16) | 413 | 13 | 54 | 30 | 10 | 24 | 8 | 21 | 489 | 14 |
| Corticosteroid use | ||||||||||
| Yes | 135 | 4 | 17 | 10 | 2 | 5 | 5 | 13 | 159 | 5 |
| No | 3138 | 96 | 161 | 91 | 40 | 95 | 33 | 87 | 3383 | 96 |
SSRI: selective serotonin reuptake inhibitors, TCA: tricyclic antidepressants, SNRI: serotonin-norepinephrine reuptake inhibitors; cm: centimeters; kg: kilogram; m: meter; CES-D: Center for Epidemiological Studies-Depression
among ever-smokers. All numbers reported as number and percentage unless otherwise stated. Shaded variables or values were not applicable for mean±SD when reporting percentage.
There were 178 participants on SSRIs, 42 on serotonin norepinephrine reuptake inhibitors, and 38 on tricyclic antidepressants. The mean age was roughly similar across antidepressant groups, with those on tricyclic antidepressants being slightly older. The SSRI group had the highest proportion of current smokers and the lowest FEV1% at baseline. This group also had the largest proportion of participants who were depressed.
Forced expiratory volume in one second
The unadjusted and adjusted findings from the linear regression models between SSRIs and FEV1 for Exam 4 are in Table 2. At both exams, findings indicate an inverse relationship between SSRIs and FEV1. 95% confidence intervals excluded a strong association in the direction hypothesized.
Table 2.
Forced expiratory volume in one second (FEV1) at MESA Exam 4: unadjusted and adjusted models
| A. Unadjusted model | B. Adjusted model 1 | C. Adjusted model 2 | |||||||
|---|---|---|---|---|---|---|---|---|---|
| β | 95% CI | p-value | β | 95% CI | p-value | β | 95% CI | p-value | |
| SSRI | −88.6 | −196.2 to 19.0 | 0.11 | −63.0 | −128.0 to 2.1 | 0.06 | −66.0 | −131.2 to −0.8 | 0.048* |
| SNRI | −511 | −182.1 to 79.9 | 0.44 | ||||||
| TCA | 72.2 | −51.5 to 195.9 | 0.25 | ||||||
| Cigarette smoking status | |||||||||
| Never-smokers | Ref | Ref | Ref | Ref | Ref | Ref | |||
| Former smokers | 1.4 | −33.7 to 36.6 | 0.94 | 1.6 | −33.6 to 36.8 | 0.93 | |||
| Current smokers | −116.0 | −172.8 to −59.3 | <.0001* | −115.9 | −172.7 to −59.1 | <.0001* | |||
| Smoking pack-years | −3.8 | −4.6 to −3.0 | <.0001* | −3.8 | −4.6 to −3.0 | <.0001* | |||
| CES-D Score | −1.8 | −3.7 to 0.04 | 0.06 | −1.8 | −3.7 to 0.04 | 0.06 | |||
CES-D: Center for Epidemiological Studies-Depression, SNRI: serotonin-norepinephrine reuptake inhibitors, SSRI: selective serotonin reuptake inhibitors, TCA: tricyclic antidepressants. Shaded variables or values were not included in the model. Beta shows the mean difference. Adjusted model 1: cigarette smoking status (Never, former, and current smokers), smoking pack years, CES-D score, age, sex, race/ethnicity, height and weight. Adjusted model 2: model 1 + serotonin and norepinephrine reuptake inhibitors and tricyclic antidepressants
Dyspnea
Results from the logistic regression models assessing the relationship between SSRIs and dyspnea are in Table 3. SSRIs were associated with an increased odds of dyspnea in both the unadjusted and adjusted models. 95% confidence intervals excluded a strong association in the direction hypothesized.
Table 3.
Dyspnea at MESA Exam 4: unadjusted and adjusted models
| A. Unadjusted model | B. Adjusted model 1 | C. Adjusted model 2 | |||||||
|---|---|---|---|---|---|---|---|---|---|
| OR | 95% CI | p-value | OR | 95% CI | p-value | OR | 95% CI | p-value | |
| SSRI | 1.99 | 1.41 to 2.80 | <.0001* | 1.18 | 0.80 to 1.74 | 0.40 | 1.15 | 0.78 to 1.70 | 0.48 |
| SNRI | 1.51 | 0.72 to 3.18 | 0.28 | ||||||
| TCA | 2.66 | 1.40 to 5.04 | 0.003* | ||||||
| Cigarette smoking status | |||||||||
| Never-smokers | Ref | Ref | Ref | Ref | Ref | Ref | |||
| Former smokers | 0.99 | 0.78 to 1.27 | 0.15 | 0.99 | 0.77 to 1.26 | 0.14 | |||
| Current smokers | 1.40 | 0.96 to 2.03 | 0.052 | 1.38 | 0.95 to 2.01 | 0.06 | |||
| Smoking pack-years | 1.01 | 1.00 to 1.01 | 0.002* | 1.01 | 1.00 to 1.01 | 0.002* | |||
| CES-D Score | 1.07 | 1.06 to 1.08 | <.0001* | 1.07 | 1.06 to 1.08 | <.0001* | |||
CES-D: Center for Epidemiological Studies-Depression, SNRI: serotonin-norepinephrine reuptake inhibitors, SSRI: selective serotonin reuptake inhibitors, TCA: tricyclic antidepressants. Adjusted model 1: age, sex, race/ethnicity/ height, weight, cigarette smoking status (never, former, and current smokers), smoking pack-years, and CES-D score. Adjusted model 2: model 1 + serotonin and norepinephrine reuptake inhibitors and tricyclic antidepressants
Sensitivity analysis
Adding lung function medications to the models decreased the effect sizes slightly, but the direction and results remained consistent.
Discussion
We hypothesized that SSRIs might be associated with better lung function through their anti-inflammatory and serotonergic effects. The results did not support our hypothesis; indeed, the associations found were in the opposite direction. Using a large population-based prospective dataset, we found that SSRIs were inversely associated with FEV1 and dyspnea, after controlling for confounders. Furthermore, this association persisted after further controlling for serotonin norepinephrine reuptake inhibitors and tricyclic antidepressants. This indicates that a better understanding of the effects of anti-depressants on lung function may be warranted.
Prior research on the associations of antidepressants and lung function have been in small studies [6, 7]. Momtaz et al. studied two groups of age- and sex-matched severe COPD participants (n=50), one of which received fluoxetine, an SSRI [6]. After 3 months on the antidepressant, the treatment group increased their FEV1 from 1.10±0.49 to 1.20±0.44 liters (p=0.01); the group that did not receive antidepressant had no change in FEV1. In a study by Perna et al., six participants were treated with citalopram, an SSRI, and after 1 month, their FEV1 significantly increased from 0.91±0.17 to 1.12±0.15 liters (p<0.05) [7].
Although we controlled for depression in this study using the CES-D, we have no measure of the severity of depression prior to being on an antidepressant. Even in individuals who were on antidepressants, the depression scores were still elevated, and a significant proportion had scores indicating depression. Since depression is linked to increased levels of inflammation [15], it is possible that the increased inflammation seen in depression overrides the potential anti-inflammatory effects of SSRIs. Therefore, we may be seeing an increase in inflammation from depression leading to a decrease in lung function rather than the SSRIs causing a decrease in lung function. Supporting this are results from two randomized, placebo-controlled studies by Brown et al. involving patients with asthma and major depressive disorder. The authors found there was no change in the dyspnea scale after treatment with SSRIs [16, 17]; however, participants who were able to achieve a depression remission had greater reductions in dyspnea than those that did not. Further studies may assess inflammation as a mediator between antidepressant and lung function.
There are several strengths to this study. This investigation was among the first to assess the impact of antidepressants on lung function using epidemiologic methods and use a large, high-quality study. Prior studies have been limited to small sample sizes and often did not consider potential confounding; this study included measures of demographics, depressive symptoms and smoking status.
Limitations
Limitations of the current study include the non-randomized, cross-sectional design and lack of information on antidepressant dose. Confounding by indication is a possibility, but due to the cross-sectional design, we are unable to assess temporality. We do not know the true reason for antidepressant prescription as they can be prescribed for conditions other than depression, such as anxiety and neurogenic pain, and we have no data on the adherence and dose of the antidepressants. Lastly, there were few participants in severe or very severe stages of COPD; it is therefore unknown if they would benefit from antidepressants.
Conclusion
In conclusion, although small, poorly controlled studies have shown an association between SSRIs and better lung endpoints, we were unable to replicate these findings in a large observational cohort.
ACKNOWLEDGEMENTS
This research was supported by contracts HHSN268201500003I, N01-HC-95159, N01-HC-95160, N01-HC-95161, N01-HC-95162, N01-HC-95163, N01-HC-95164, N01-HC-95165, N01-HC-95166, N01-HC-95167, N01-HC-95168 and N01-HC-95169 from the National Heart, Lung, and Blood Institute, and by grants UL1-TR-000040, UL1-TR-001079, and UL1-TR-001420 from the National Center for Advancing Translational Sciences (NCATS). The authors thank the other investigators, the staff, and the participants of the MESA study for their valuable contributions. A full list of participating MESA investigators and institutions can be found at http://www.mesa-nhlbi.org.
List of abbreviations
- CES-D
Centers for Epidemiologic studies depression
- COPD
Chronic obstructive pulmonary disease
- FEV1
forced expiratory volume in one second
- MESA
Multi-Ethic Study of Atherosclerosis
- SSRI
Selective serotonin reuptake inhibitor
Footnotes
Ethics approval- The institutional review boards of all collaborating institutions and the National Heart Lung and Blood Institute (National Institutes of Health, Bethesda, MD, USA) approved the protocols for MESA and all procedures described herein. All participants provided written informed consent
Competing interests- None
Code availability- available upon request
Contributor Information
Hilary F. Armstrong, Columbia University Medical Center, New York.
David Lederer, Columbia University Medical Center, New York.
Gina S. Lovasi, Drexel Dornsife School of Public Health, Philadelphia.
Grant Hiura, Columbia University Medical Center, New York.
Corey E. Ventetuolo, Brown University, Providence.
R Graham Barr, Columbia University Medical Center, New York.
Availability of data and materials-
The datasets generated and/or analyzed during the current study are available in the MESA Lung repository, https://www.mesa-nhlbi.org/
REFERENCES
- 1.(GOLD), G.I.f.C.O.L.D., Global Strategy for the Diagnosis, Management and Prevention of COPD. 2015.
- 2.Yohannes AM, Müllerová H, Hanania NA, et al. Long-term course of depression trajectories in patients with COPD. Chest. 2016;149(4):916–926. doi: 10.1016/j.chest.2015.10.081 [DOI] [PubMed] [Google Scholar]
- 3.Hancox RJ, et al. , Systemic inflammation and lung function in young adults. Thorax, 2007. 62(12): p. 1064–1068. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Tynan RJ, et al. , A comparative examination of the anti-inflammatory effects of SSRI and SNRI antidepressants on LPS stimulated microglia. Brain, Behavior, and Immunity, 2012. 26(3): p. 469–479. [DOI] [PubMed] [Google Scholar]
- 5.Hilaire G, et al. , The role of serotonin in respiratory function and dysfunction. Respiratory Physiology & Neurobiology, 2010. 174(1–2): p. 76–88. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Momtaz OM, et al. Effect of treatment of depression and anxiety on physiological state of severe COPD patients. Egyptian Journal of Chest Diseases and Tuberculosis, 2015. 64, 29–34 DOI: 10.1016/j.ejcdt.2014.08.006. [DOI] [Google Scholar]
- 7.Perna G, Cogo R, and Bellodi L, Selective serotonin re-uptake inhibitors beyond psychiatry: Are they useful in the treatment of severe, chronic, obstructive pulmonary disease? Depression and Anxiety, 2004. 20(4): p. 203–204. [DOI] [PubMed] [Google Scholar]
- 8.Sukhatme VP, Reiersen AM, Vayttaden SJ, Sukhatme VV. Fluvoxamine: A review of its mechanism of action and its role in covid-19. Frontiers in Pharmacology. 2021;12. doi: 10.3389/fphar.2021.652688 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Bild DE, et al. , Multi-Ethnic Study of Atherosclerosis: Objectives and Design. American Journal of Epidemiology, 2002. 156(9): p. 871–881. [DOI] [PubMed] [Google Scholar]
- 10.Rodriguez Josanna. “The Association of Pipe and Cigar Use With Cotinine Levels, Lung Function, and Airflow Obstruction.” Annals of Internal Medicine, vol. 152, no. 4, 2010, p. 201., doi: 10.7326/0003-4819-152-4-201002160-00004. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Miller MR, et al. , Standardisation of spirometry. European Respiratory Journal, 2005. 26(2): p. 319–338. [DOI] [PubMed] [Google Scholar]
- 12.Wanger J, et al. , Standardisation of the measurement of lung volumes. European Respiratory Journal, 2005. 26(3): p. 511–522. [DOI] [PubMed] [Google Scholar]
- 13.Hankinson JL, et al. , Performance of american thoracic society-recommended spirometry reference values in a multiethnic sample of adults: The multi-ethnic study of atherosclerosis (mesa) lung study. CHEST Journal, 2010. 137(1): p. 138–145. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Maurer J, et al. , Anxiety and depression in copd: Current understanding, unanswered questions, and research needs. Chest, 2008. 134(4_suppl): p. 43S–56S. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Cepeda MS, Stang P, and Makadia R, Depression Is Associated With High Levels of C-Reactive Protein and Low Levels of Fractional Exhaled Nitric Oxide: Results From the 2007–2012 National Health and Nutrition Examination Surveys. J Clin Psychiatry, 2016. 77(12): p. 1666–1671. [DOI] [PubMed] [Google Scholar]
- 16.Brown ES, et al. , Escitalopram for Severe Asthma and Major Depressive Disorder: A Randomized, Double-Blind, Placebo-Controlled Proof-of-Concept Study. Psychosomatics, 2012. 53(1): p. 75–80. [DOI] [PubMed] [Google Scholar]
- 17.Brown ES, et al. , A Randomized Trial of Citalopram versus Placebo in Outpatients with Asthma and ajor Depressive Disorder: A Proof of Concept Study. Biological Psychiatry, 2005. 58(11): p. 865–870. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
The datasets generated and/or analyzed during the current study are available in the MESA Lung repository, https://www.mesa-nhlbi.org/