Assessment of psychiatric comorbidities and serotonergic or noradrenergic medication use on blood pressure using 24‐hour ambulatory blood pressure monitoring

Abstract In this study, the authors aimed to assess both nighttime and daytime blood pressure (BP) variability using 24‐hour ambulatory BP monitoring (ABPM) in persons with and without psychiatric conditions and with or without selective serotonin reuptake inhibitors (SSRIs) or serotonin‐norepinephrine reuptake inhibitors (SNRIs) treatment. In this retrospective study, patients who underwent psychiatric evaluation and ABPM within 6 months of each other between January 1, 2012 and December 31, 2017 were identified using billing data. Participants were divided into three groups—participants with no psychiatric diagnosis and no psychiatric medicine (−Diagnosis/−Medication), those with psychiatric diagnosis and on SSRIs/SNRIs (+Diagnosis/+Medication), and psychiatric diagnosis but no psychiatric medications (+Diagnosis/−Medication). Day and nighttime systolic and diastolic BPs were compared between groups controlling for relevant variables using multivariable linear regression models. A total of 475 participants met inclusion criteria including 135 in the −Diagnosis/−Medication group, 232 in the +Diagnosis/+Medication group, and 108 in the +Diagnosis/−Medication group. In adjusted multivariable analysis, the +Diagnosis/+Medication group had higher nighttime systolic BP (median 120 vs 110 mm (Hg); p = .01) and nighttime diastolic BP (median 68 vs 63 mm (Hg); p = .006) as compared to −Diagnosis/−Medication. No statistically significant differences in BPs between the −Diagnosis/−Medication and +Diagnosis/−Medication groups were observed, after adjustment. Use of SSRIs/SNRIs was associated with significantly higher nocturnal systolic and diastolic BP among patients with psychiatric diagnosis using SSRIs/SNRIs but not associated with psychiatric diagnosis without SSRI/SNRI use. SSRIs/SNRIs use may be associated with higher BP levels and this merits future prospective studies using ABPM to assess day and nighttime BP changes with SSRIs/SNRIs use.


INTRODUCTION
In 2017, about 7% of all US adults (17.3 million) reported at least one episode of major depression and of those, 11 million reported severe impairment from this episode. 1,2 Additionally, 19.1% reported having an anxiety disorder in 2017 and 31.1% reported experiencing an anxiety disorder at some point in their life. 3 Selective serotonin reuptake inhibitors (SSRIs) and serotonin-norepinephrine reuptake inhibitors (SNRIs) are considered first line treatment choice for most patients with major depression 4 and various anxiety disorders. 5 Antidepressant use is common; in 2013, 12% of adults in United States reported using antidepressant medications and eight out of 10 reported long term use. 6 Furthermore, use of any psychiatric medication was more common among those aged 60-85 years of age (25.1%) compared with those who were 18-39 years old (9.0%). 6 Association of depression and anxiety with cardiovascular mortality has been previously reported. [7][8][9][10] This highlights the critical importance of monitoring cardiovascular health and managing any modifiable cardiac and metabolic risk factors in patients with depression and anxiety.
Hypertension (HTN) is an important, common, and potentially modifiable risk factor to consider among those dealing with depression and anxiety. The prevalence of age-adjusted HTN among adults in United States is 45.4% according to data from national health and nutrition examination survey period 2017-2018. 11 A number of complications occur due to elevated blood pressure (BP) including cardiovascular disease and chronic kidney disease (CKD). 12,13 Association of depression by itself and BP has been mixed. It has been linked to hypotension instead of HTN 14 or no association in a well-designed population based study. 15 While anxiety has been positively associated with HTN. 16 One study has reported that anxiety is associated with nocturnal and early morning HTN. 17 An elegant hypothesis was put forward decades ago by Friedman and Bennet, who in their study of 1101 participants did not find any association between depression and HTN but did find association between anxiety and HTN 18 : depression causes stress to an individual and this affective-stress not always but often results in anxiety, which in turn increases BP. 18 Diagnosis of HTN, its causes, and impact of cardiovascular risk requires repeated BP measurement, clinical assessment, and workup including physical examination and laboratory testing. 19 BP measurement is complicated by the spontaneous variation over the course of the same day and marked reduction occurring during sleep among those with or without HTN. 20 24-hour ambulatory BP monitoring (ABPM) is better able to capture BP changes over the course of the day, than the office-based BP or home-based BP measurement. ABPM is a better predictor of changes in response to treatment, 21 is superior in its reproducibility, 22 and less prone to phenomenon of "white coat HTN." 23 Furthermore, nighttime and daytime HTN both need to be considered. A meta-analysis of four large prospective European studies found that nighttime BP was a better predictor of mortality and cardiovascular events. 24 Additionally, other studies have shown the association of nocturnal HTN with adverse cardiovascular outcomes. 25,26 Nighttime BP is not routinely measured, except when utilizing ABPM.
Antidepressant medications predominantly affect the neurotransmitters serotonin, norepinephrine, and dopamine. Variation in BP might be a result of the effects of these neurotransmitters. Furthermore, there is a positive association between the binding affinities of SSRIs and SNRIs for norepinephrine and serotonin transporters and arterial HTN. 27,28 Based on this concept, we designed this retrospective study to assess both nighttime and daytime BP variability using 24-hour ABPM in patients with and without psychiatric conditions and with or without SSRIs/SNRIs treatment. We hypothesized that patients on SSRIs/SNRIs would experience elevated BP, as measured by ABPM.

MATERIALS AND METHODS
The retrospective study was approved by our institutional review

RESULTS
A total of 475 participants met the inclusion criteria; the −Diagnosis/−Medication group included 135 participants, the +Diagnosis/+Medication group was comprised of 232 participants, and the +Diagnosis/−Medication group consisted of 108 participants. Baseline demographics of the groups are summarized in Table 1 and ABPM readings are summarized in Table 2 In order to examine whether higher nighttime systolic and diastolic BP in the +Diagnosis/+Medication group was associated with the psychiatric diagnosis or the use of SSRIs/SNRIs, we compared ABPM results between the −Diagnosis/−Medication group and the +Diagnosis/−Medication group using the same model as for the first analysis. In unadjusted analysis, participants in the +Diagnosis/−Medication group had significantly higher mean daytime systolic BP than the −Diagnosis/−Medication group by 4.28 (95% CI: 0.71, 7.86) mm (Hg) on average (p = .02) ( Table 3). However, after adjustment, there was no statistically significant difference between these groups both daytime and nighttime systolic or diastolic BP measurements ( Table 3).
In order to assess associations of SNRIs and SSRIs with BP separately, we divided the +Diagnosis/+Medication group participants into two subgroups: the SNRI subgroup (participants with psychiatric diagnosis and taking SNRIs) and the SSRI subgroup (participants with psychiatric diagnosis and taking SSRIs). Twelve patients were taking both SSRIs and SNRIs and were excluded from subgroup analysis. The demographics and comorbidities of these subgroups are summarized in Table 4 and ABPM readings are summarized in Table 5. No statistically significant differences were observed in unadjusted or adjusted analysis between SNRI and SSRI subgroups for daytime or nighttime systolic and diastolic BP ( Table 6).

P-value
Beta (95% CI) showed a decrease in the mortality rate among patient prescribed SSRIs with concomitant cardiovascular disease. 43 Additionally, there has also been evidence of beneficial effects of SSRIs use in patients with congestive heart failure and HTN. [44][45][46][47] Comorbid depression may lead to poorer adherence to antihypertensive treatment leading to polypharmacy, 48,49 while its treatment can lead to better BP control, 50 and finally, there is preliminary evidence that integrated treatment of depression and HTN can lead to better outcomes for both. 51 Thus, the potential risk of increase in nocturnal BP needs to be balanced against the potential beneficial effects of SSRIs/SNRIs use.

Unadjusted analysis
Our study methodology of using ABPM has implications for future clinical studies evaluating safety profile of serotonergic and/or noradrenergic psychotropic medications. This has already been proposed for HTN related clinical trials. 22 Had we not used ABPM, we would have not found the association of SSRIs and SNRIs with higher nocturnal BP. Thus, any study not using ABPM will likely miss important BP changes, especially the nighttime and early morning BP changes.  Models adjusted for age, sex, race, history of hypertension, history of diabetes, smoking status, number of antihypertensive medications, and chronic kidney disease.
Use of a relatively large sample size of a population that is representative of a general psychiatric clinic population where detailed psychiatric evaluations performed by experienced psychiatrists with specialty and sub-specialty training were available, use of 24-hour ABPM to assess BP variability, use of a comparison groups −Diagnosis/−Medication (kidney transplant donors) and +Diagnosis/−Medication (psychiatric diagnosis with no SSRIs/SNRIs) are strengths of this study.

CONCLUSIONS
SSRIs/SNRIs use was associated with a higher nocturnal systolic and diastolic BP. This association needs to be further studied using an appropriately powered longitudinal cohort study using ABPM at the time of psychiatric diagnosis prior to the start of SSRIs/SNRIs and repeating ABPM while receiving treatment. SSRIs/SNRIs, like any other medications, need to be prescribed after careful risk-benefit analysis, especially in patients with pre-existing HTN, keeping in mind that sub-optimal or no treatment of psychiatric comorbidities in itself is associated with considerable risk.

ACKNOWLEDGMENTS
This study was supported by the Mayo Clinic Robert D. and Patricia E.
Kern Center for the Science of Healthcare Delivery.

CONFLICT OF INTEREST
All authors have no conflicts of interest to report.

AUTHOR CONTRIBUTIONS
Shehzad K. Niazi was responsible for study concept and oversight, interpretation of results, and writing and editing of the manuscript.
Sobia H. Memon was responsible for data preparation and maintenance, and writing and editing of the manuscript. Elizabeth R. Lesser was responsible for statistical analysis and writing of the manuscript.
Emily Brennan was responsible for statistical analysis and editing of the manuscript. Nabeel Aslam was responsible for study concept and oversight, interpretation of results, and writing and editing of the manuscript.