Abstract
Objective:
Anti-hypertensive medications have been examined as agents for PTSD prevention in trauma-exposed individuals, given well-documeted associations between PTSD and increased risk of cardiovascular disease and purported trauma-relavent mechanisms of action for these medications. Evidence regarding the effectiveness of such drugs for this purpose remains mixed.
Methods:
We conducted a national population-based cohort study using data from Danish national registries to assess whether four classes of anti-hypertensive drugs (beta-adrenoceptor blockers [beta blockers], angiotensin II receptor blockers [ARBs], angiotensin-converting enzyme [ACE] inhibitors, and calcium channel blockers) were associated with a decreased incidence of PTSD (diagnosed according to ICD-10) over a 22-year study period. Data for this study originated from a population-based cohort of over 1.4 million persons who experienced a traumatic event between 1994 and 2016 in Denmark. We calculated the incidence rate of PTSD per 100,000 person-years among persons who filled a prescription for each class of drug in the 60 days prior to a traumatic event and for corresponding unexposed comparison groups. We then used Cox proportional hazards regression to compare the rate of PTSD among persons who filled an anti-hypertensive medication prescription within 60 days before their trauma to the rate among persons who did not.
Results:
We found evidence that calcium channel blockers were associated with a decreased incidence of PTSD (adjusted hazard ratio = 0.60, 95% confidence interval [CI] = 0.33, 1.1); all other anti-hypertensive medication classes had null or near null associations.
Conclusion:
These findings lay a foundation for additional research focusing on anti-hypertensive medications that appear most effective in reducing PTSD incidence following trauma and for additional replication work aimed at continuing to clarify the disparate findings reported in the literature to date.
Keywords: PTSD, trauma, hypertension, anti-hypertensives, cohort study, pharmacoepidemiology
Introduction
There is a well-documented association between trauma, posttraumatic stress disorder (PTSD), and cardiovascular disease (CVD).1 Accordingly, drugs for treating hypertension have been examined for reduction in PTSD symptoms, and to a lesser extent for PTSD prevention, in trauma-exposed individuals. Antihypertensive medications include drugs across a variety of classes: beta-adrenoceptor blockers (beta blockers), angiotensin II receptor blockers (ARBs), angiotensin-converting enzyme (ACE) inhibitors, and calcium channel blockers. Several studies have reported a possible association between anti-hypertensive medications and posttraumatic stress disorder (PTSD), although results are mixed.2–5 Much of the evidence documenting associations between anti-hypertensive drugs and PTSD come from small clinical (not population-based) studies and animal models, and thus, conflicting findings may be due to true etiologic differences or differences between samples and other methodologic choices.
Meta-analyses and literature reviews summarizing several small studies have concluded that there is weak evidence for a small protective effect of propranolol, a beta blocker, on incident PTSD prevention.2–4 It is hypothesized that propranolol interferes with memory reconsolidation at the time of trauma, thereby preventing reactivation of disturbing memories (a key symptom of PTSD).6 Other beta blockers may also affect PTSD symptoms. For example, atenolol, a non-lipid-soluble beta blocker, is thought to reduce anxiety and is sometimes used to treat PTSD and anxiety disorders.3,7 Additionally, evidence from animal models suggests that calcium channel blockers (e.g., nimodipine, nisolapine) may also play a role in either interrupting memory reconsolidation8 or resolving anxiety symptoms.9,10 A study using cross-sectional data from a hospital biorepository did not document differences in use of calcium channel blockers among people with and without prevalent PTSD,5 however incidence of PTSD was not examined. More recently, a large cohort study using electronic health registry data found that brain penetrant calcium channel blockers were associated with a decreased risk of many incident psychiatric disorders, including anxiety disorders, when compared with a low brain permeability calcium channel blocker (amlodipine), although PTSD was not specifically assessed as an outcome.11
Both human and animal studies offer some evidence that ARBs and ACE inhibitors may be associated with a decreased frequency of PTSD or lower PTSD symptoms. For example, a study using a small clinical sample12 and a cross-sectional study using data from a hospital-based biorepository,5 reported that individuals taking either ARBs or ACE inhibitors had fewer prevalent PTSD symptoms than individuals not taking these drugs and that people on these medications were less likely to have prevalent PTSD.5,13 ARBs or ACE inhibitors affect the renin-angiotensin system and have been found to be associated with reducing anxiety and stress, possibly via promoting fear extinction.14–17 ARBs have been more widely studied than ACE inhibitors, with the ARB, losartan, receiving particular attention.14,17 However, a recent randomized controlled trial did not find evidence that losartan performed better than placebo in reducing prevalent PTSD symptoms.18
Taken together, the literature documents some possible evidence of an effect of anti-hypertensive medications on PTSD, but research has been mixed with regard to focus on prevalent PTSD maintenance or incidence and results have been contradictory. Further, population-based studies of PTSD incidence (i.e., new onset among people initially free from PTSD) are rare. The present study fills some of the gaps in this literature through the use of a national population-based study to assess whether four classes of anti-hypertensive drugs, captured as a filled prescription within 60 days before a traumatic experience, were associated with a decreased incidence of PTSD over a 22-year study period.
Materials and Methods
Data for this study originated from a population-based cohort of over 1.4 million persons who experienced a traumatic event in Denmark between 1994 and 2016. The methods used for this study have been described elsewhere.19 Briefly, persons in the cohort experienced probable traumatic events including: fire/explosion, transportation accident, exposure to a toxic substance/medical complications, traumatic brain injury, assault with or without a weapon, pregnancy-related trauma, suicide death of a family member, and multiple traumatic events (i.e., persons experiencing more than one of these events on the same day). Traumatic event data was obtained from registry-based sources including the Danish National Patient Registry,20 the Danish Psychiatric Central Research Register,21 the Danish Medical Birth Registry,22 the Civil Registration System20 and the Cause-of-Death Registry.23 While we were unable to fully assess the severity of these events given the registry-based nature of the data, we took several steps to ensure that we were capturing traumatic experiences rather than stressors (such as requiring a hospital stay following certain events).19 Previous analyses indicate that we were able to isolate events that were more serious than typical stressors with these definitions; we found risk of all forms of psychopathology were increased following all but one of our defined traumatic events (the exception was pregnancy-related trauma) when compared with risk of psychopathology following a common stressor (non-suicidal death of a loved one).19 All traumatic events were the first recorded traumatic event within the study period. The current analysis was restricted to persons with no recorded psychiatric disorders between 1994 and date of the potentially traumatic experience (International Classification of Diseases, 10th Edition, or ICD-10, F codes).24
We defined exposure as redemption of a prescription for one of four classes of anti-hypertensive medication (calcium channel blockers, Anatomical Therapeutic Chemical or ATC code: C08; ACE inhibitors, ATC code: C09AA; ARBs, ATC code: C09CA; and beta blockers, ATC code: C07) within 60 days prior to a traumatic event. Although other timeframes for medication redemption prior to trauma were possible, we chose 60 days as our timeframe of interest in an effort to balance looking back far enough to capture all of the people who might be on these medications at the time of trauma against looking back so far that some participants may have redeemed the prescription within the time frame but already stopped using it by the time the trauma occurred. Medication data were obtained from the Danish National Prescription Registry covering all Danish pharmacies.20 Use of Danish registry data for the current population-based study confers a major advantage in that medication variables in the Danish National Prescription Registry indicates a prescription that was filled at a pharmacy, rather than just receipt of a prescription issued by a physician – the norm in most population-based registry-based data sources. To create an unexposed comparison cohort, we matched each exposed person, without replacement, to up to three persons who did not fill an anti-hypertensive medication prescription within 60 days prior to their traumatic experience (matched on year of birth, month and year of trauma, and sex). Follow-up ended for unexposed persons if they filled an anti-hypertensive medication prescription in the study period. Sample sizes are presented in Table 1. The outcome was an incident diagnosis of PTSD (ICD-10 code: F43.1)24 over the 22-year follow-up period, as indicated in either the Danish National Patient Registry20 (covering all inpatient and outpatient contacts at non-psychiatric hospitals in Denmark) and/or the Danish Psychiatric Central Research Registrer21 (covering all inpatient and outpatient care at psychiatric hospitals in Denmark). We included all primary and secondary diagnoses in the present analysis. A validation study of PTSD diagnoses in the Psychiatric Central Research Registry showed high validity when compared with independent symptom reassessment.25 Confounders were chosen for adjustment based on the existing literature. They included marital status and income at the time of trauma obtained from the Civil Registration System,20 filled prescriptions for an antidepressant (ATC code: N06A), an NSAID (ATC code: M01A), or a statin (ATC code: C10AA) within 60 prior to the traumatic event, and Charlson Comorbidity Index (CCI) score at any time before trauma26 (based on diagnoses recorded in the Danish National Patient Register) and traumatic event type.
Table 1.
Descriptive characteristics of the samples at the time of trauma.
| Calcium Channel Blockers (n = 31441) | Unexposed (n = 81031) | ACE Inhibitors (n = 28303) | Unexposed (n = 74770) | Beta Blockers (n = 39004) | Unexposed (n = 95563) | Angiotensin II Receptor Blockers (n = 12244) | Unexposed (n = 35679) | |
|---|---|---|---|---|---|---|---|---|
| n (%) | n (%) | n (%) | n (%) | n (%) | n (%) | n (%) | n (%) | |
|
| ||||||||
| Male | 14986 (48%) | 39169 (48%) | 15299 (54%) | 39939 (53%) | 18446 (47%) | 44846 (47%) | 5516 (45%) | 16162 (45%) |
| Female | 16455 (52%) | 41862 (52%) | 13004 (46%) | 34831 (47%) | 20558 (53%) | 50717 (53%) | 6728 (55%) | 19517 (55%) |
| Age group | ||||||||
| <30 | 215 (0.7%)a | 648 (0.8%) | 176 (0.6%)a | 520 (0.7%) | 1138 (2.9%)a | 3411 (3.6%) | 48 (0.4%) | 144 (0.4%) |
| 30–50 | 1799 (5.7%)a | 5382 (6.6%) | 1785 (6.3%)a | 5366 (7.2%) | 3319 (8.5%)a | 9876 (10%) | 707 (5.8%) | 2106 (5.9%) |
| 50+ | 29427 (94%)a | 75001 (93%) | 26342 (93%)a | 68884 (92%) | 34547 (89%)a | 82276 (86%) | 11489 (94%) | 33429 (94%) |
| Married/Reg. Part. | 15937 (51%)a | 41900 (52%) | 14554 (51%) | 40010 (54%) | 19904 (51%)a | 49762 (52%) | 6516 (53%) | 18728 (53%) |
| Single | 2605 (8.3%)a | 7642 (9.4%) | 2578 (9.1%) | 6993 (9.4%) | 4177 (11%)a | 11677 (12%) | 923 (7.5%) | 3073 (8.6%) |
| Divorced | 3794 (12%)a | 10919 (14%) | 3517 (12%) | 10086 (14%) | 4462 (11%)a | 12619 (13%) | 1494 (12%) | 4823 (14%) |
| Widowed | 9095 (29%)a | 20539 (25%) | 7649 (27%) | 17634 (24%) | 10448 (27%)a | 21441 (22%) | 3311 (27%) | 9041 (25%) |
| Unknown | 10 (0%)a | 31 (0%) | --c | 47 (0.1%) | 13 (0%)a | 64 (0.1%) | --c | 14 (0%) |
| Income Quartile | ||||||||
| 1 (lowest) | 7154 (23%)a | 16696 (21%) | 6371 (23%)a | 15577 (21%) | 9310 (24%)a | 19695 (21%) | 2793 (23%) | 8452 (24%) |
| 2 | 13847 (44%)a | 32718 (40%) | 12345 (44%)a | 29121 (39%) | 16187 (42%)a | 36218 (38%) | 5015 (41%) | 14126 (40%) |
| 3 | 5682 (18%)a | 15438 (19%) | 5130 (18%)a | 14441 (19%) | 7470 (19%)a | 19088 (20%) | 2307 (19%) | 6606 (19%) |
| 4 (highest) | 4686 (15%)a | 15843 (20%) | 4371 (15%)a | 15265 (20%) | 5900 (15%)a | 19946 (21%) | 2099 (17%) | 6395 (18%) |
| Child (no income) | 23 (0.1%)a | 68 (0.1%) | 22 (0.1%)a | 67 (0.1%) | 38 (0.1%)a | 114 (0.1%) | --c | --c |
| Unknown | 58 (0.2%)a | 268 (0.3%) | 64 (0.2%)a | 299 (0.4%) | 99 (0.3%)a | 502 (0.5%) | 30 (0.2%) | 100 (0.3%) |
| Antidepressantsb | 3734 (12%)a | 6893 (8.5%) | 3208 (11%) | 6315 (8.4%) | 4334 (11%)a | 7620 (8.0%) | 1423 (12%) | 3211 (9.0%) |
| NSAIDs | 4972 (16%) | 10950 (14%) | 4150 (15%) | 10302 (14%) | 4990 (13%) | 12577 (13%) | 1968 (16%) | 4825 (14%) |
| Statins | 7088 (23%)a | 6551 (8.1%) | 7338 (26%)a | 6110 (8.2%) | 10087 (26%)a | 7014 (7.3%) | 3236 (26%)a | 4354 (12%) |
| CCI Score | ||||||||
| 0 | 10506 (33%)a | 42027 (52%) | 7840 (28%)a | 39093 (52%) | 12807 (33%)a | 52717 (55%) | 4064 (33%)a | 16771 (47%) |
| 1–2 | 12907 (41%)a | 27332 (34%) | 12099 (43%)a | 25154 (34%) | 15243 (39%)a | 30095 (32%) | 4825 (39%)a | 12508 (35%) |
| 3+ | 8028 (26%)a | 11672 (14%) | 8364 (30%)a | 10523 (14%) | 10954 (28%)a | 12751 (13%) | 3355 (27%)a | 6400 (18%) |
| Type of Trauma | ||||||||
| Fire/explosion | 982 (3.1%)a | 4045 (5.0%) | 925 (3.3%)a | 3734 (5.0%) | 1000 (2.6%)a | 4694 (4.9%) | 452 (3.7%)a | 1506 (4.2%) |
| Trans. accident | 950 (3.0%)a | 3014 (3.7%) | 800 (2.8%)a | 3094 (4.1%) | 898 (2.3%)a | 3718 (3.9%) | 390 (3.2%)a | 1475 (4.1%) |
| Exp. to toxic substance/Medical complications |
23025 (73%)a | 51929 (64%) | 21041 (74%)a | 48008 (64%) | 27759 (71%)a | 58086 (61%) | 8962 (73%)a | 24000 (67%) |
| TBI | 5808 (19%)a | 18874 (23%) | 5100 (18%)a | 17379 (23%) | 7061 (18%)a | 20881 (22%) | 2217 (18%)a | 7560 (21%) |
| Assault | 110 (0.3%)a | 616 (0.8%) | 119 (0.4%)a | 614 (0.8%) | 145 (0.4%)a | 791 (0.8%) | 60 (0.4%)a | 291 (0.8%) |
| Pregnancy-related | 209 (0.7%)a | 1156 (1.4%) | 21 (0.1%)a | 709 (0.9%) | 1768 (4.5%)a | 5883 (6.2%) | --c | 289 (0.8%) |
| Family suicide | 171 (0.5%)a | 758 (0.9%) | 108 (0.4%)a | 630 (0.8%) | 166 (0.4%)a | 778 (0.8%) | 77 (0.6%)a | 271 (0.8%) |
| Multiple events | 186 (0.6%)a | 639 (0.8%) | 189 (0.7%)a | 602 (0.8%) | 207 (0.5%)a | 732 (0.8%) | 81 (0.7%)a | 287 (0.8%) |
Note.
Standardized difference compared exposed to unexposed for overall category was greater than 0.1.
Although we restricted the sample to persons without psychiatric disorders prior to trauma, some cohort members had been prescribed pre-trauma antidepressants, (e.g., non-psychiatric indications such as chronic pain and smoking cessation).
Values under 10 not displayed.
Abbreviations
CCI = Charlson Comorbidity Index
Exp. to toxic substance = Exposure to toxic substance
NSAIDs = Non-steroidal anti-inflammatory drugs
Reg. Part. = Registered partner
TBI = Traumatic brain injury
Trans. Accident = Transportation accident
We conducted descriptive analyses comparing demographic characteristics (sex, age group, marital status, income quartile, and type of trauma) and individual CCI diagnoses in the exposed and matched unexposed groups and calculated standardized differences between the groups for each variable. Next, we calculated the incidence rate of PTSD per 100,000 person-years, and the cumulative incidence of PTSD over the entire study period, among persons in the exposed and unexposed groups. Incidence rates are a measure of the number of outcome cases that occur over a period of time that is equivalent across all groups (i.e., 100,000 person-years) and can be interpreted as the number of cases that would occur if you followed 100,000 people for one year or 50,000 people for two years, etc. Finally, we conducted Cox proportional hazards regression to compare the PTSD rate among persons who filled an anti-hypertensive medication prescription within 60 days prior to their traumatic experience to matched persons who did not fill a prescription for these medications. We estimated associations based on age, trauma date, and sex-matched groups (i.e., the minimally adjusted models) and associations additionally adjusted for the confounders noted above. In accordance with recent scientific recommendations against using null-hypothesis significance testing to determine the importance of results,27,28 and concerns raised regarding the role of null-hypothesis significance testing in the reproducibility crisis,29,30 we evaluate our results without regard to whether their confidence intervals include the null. Rather, we present confidence intervals as information regarding the range of likely values for the evaluated associations. Further, we interpret our results with regard to the level of evidence provided by these point estimates and confidence intervals. That is, point estimates and confidence intervals that indicate a range of consistent estimates as most likely (either protective or harmful) will be interpreted as potential findings, while near null point estimates and confidence intervals that provide relatively equal evidence of protective and harmful associations will be interpreted as less informative evidence. Importantly, we provide all results so that readers can make their own determinations about our findings. All analyses were conducted in SAS version 9.4. This work was approved by the Institutional Review Board at Boston University Medical Campus and the Danish Data Protection Agency [2015-57-0002]. Given data security requirements, all data and code can only be accessed by request to the Danish government.
Results
Descriptive characteristics of the sample are presented in Table 1. The sample as a whole was relatively evenly split by sex and the large majority of persons were over 50 years of age. This is consistent with the general timing of prescriptions for anti-hypertensive medications. Among persons exposed and unexposed to one of the four types of anti-hypertensive medications, the biggest differences were in CCI score at the time of trauma – persons receiving anti-hypertensive medications had more comorbid conditions. The average length of follow-up time for participants in the sample was 5–6 years and varied slightly based on the type of anti-hypertensive medication.
Results of all regression analyses are presented in Table 2. For persons who received calcium channel blockers within 60 days prior to trauma, the rate of PTSD was 10.9/100,000 person-years and the cumulative incidence in the study period was 0.06%. The PTSD rate among the matched unexposed cohort was 15.7/100,000 person-years and the cumulative incidence in the study period was 0.08%. This corresponded to a minimally adjusted association of 0.70 (95% confidence interval [CI] = 0.41, 1.2). After adjusting for marital status, income, other medications prescribed and redeemed at the time of trauma, CCI score, and trauma type the observed association was 0.63 (95% CI = 0.34, 1.2).
Table 2.
Associations between Categories of Prescribed Anti-hypertensive Medications at the Time of Trauma and Incident PTSD.
| Calcium Channel Blockers | ACE Inhibitors | Beta Blockers | Angiotensin II Receptor Blockers | |
|---|---|---|---|---|
|
|
||||
| Minimally adjusted modelsa | 0.70 (0.41, 1.2) | 1.1 (0.70, 1.8) | 0.94 (0.66, 1.3) | 0.72 (0.36, 1.5) |
| Adjusted modelsb | 0.63 (0.34, 1.2) | 1.2 (0.67, 2.1) | 0.88 (0.60, 1.3) | 1.1 (0.48, 2.4) |
|
|
||||
Note.
Minimally adjusted models are adjusted for age, sex and trauma date due to the matching used to create the analytic cohorts.
Adjusted models additional controlled for martial status, prescriptions for antidepressant, non-steroidal anti-inflammatory drugs, or statin, and Charlson comorbidity index score.
In contrast, we found weak or no evidence of an association with PTSD incidence for ACE inhibitors, ARBs, and beta blockers. Among persons who received ACE inhibitors, the rate of PTSD was 17.2/100,000 person-years and the cumulative incidence was 0.09%, while among the matched unexposed group the rate of PTSD was comparable, at 17.4/100,000 person-years, with a cumulative incidence of 0.09%. The minimally adjusted association between ACE inhibitors and PTSD incidence was 1.1 (95% CI = 0.70, 1.8) and after adjusting for additional confounders, the observed association was 1.2 (95% CI = 0.67, 2.1). Among persons who received beta blockers, the rate of PTSD was 20.9/100,000 person-years with a cumulative incidence of 0.12%. In the matched unexposed comparison group, the rate of PTSD was 24.5/100,000 person-years with a cumulative incidence of 0.13%. These rates corresponded to a minimally adjusted association of 0.94 (95% CI: 0.66, 1.3) and an association of 0.88 (95% CI: 0.60, 1.3), after adjusting for additional confounders. Finally, persons who filled a prescription for ARBs within 60 days of their trauma had a PTSD rate of 15.9/100,000 person-years with a cumulative incidence of 0.08%, while the matched comparison group without a prescription for any anti-hypertensive medication had a PTSD rate of 21.8/100,000 person-years and a cumulative incidence of 0.10%. This corresponded to a minimally adjusted association of 0.72 (95% CI: 0.36, 1.5) between ARBs and PTSD incidence and an adjusted association of 1.1 (95% CI: 0.48, 2.4).
Discussion
We found evidence that a specific class of anti-hypertensive medications – calcium channel blockers – was associated with a decreased incidence of PTSD following trauma, while ACE inhibitors, ARBs, and beta blockers showed little to no effect on PTSD incidence in our sample. This is in contrast to a recent study by Seligowski et al. (2021) using data from a hospital biorepository which found that people on ACE inhibitors and ARBs were less likely to have prevalent PTSD, but not calcium channel blockers.5 There may be methodologic explanations for these discrepant findings. While we were able to confirm that participants had at least filled a prescription for an anti-hypertensive medication within 60 days prior to their traumatic experience, the study by Seligowski et al. (2021) used electronic health record data to confirm medications but did not explore timing of medications relative to traumatic experiences.5 Further, our study used longitudinal data to observe PTSD incidence (i.e., the sample was PTSD free at the time of trauma and at the time they filled the anti-hypertensive prescription) while the study by Seligowski et al. (2021) was a cross-sectional assessment of the co-occurrence of medication prescriptions and PTSD diagnosis.5 Consistent with our findings, recent work by Colbourne and Harrison (2022) found that brain permiating calcium channel blockers were protective against incident anxiety disorders when compared with ARBs among a sample with no prior psychiatric disorders, although PTSD was not examined as a specific outcome.11 Taken together, these findings provide a foundation for additional longitudinal research on anti-hypertensive medications that appear most effective in reducing PTSD incidence following trauma and for additional work aimed at clarifying the disparate findings regarding these associations in the literature to date.
Knowledge is accumulating regarding the potential mechanisms through which antihypertensive medications may impact PTSD incidence. There is a growing literature implicating the renin-angiotensin system, which is involved in blood pressure regulation, in the link between PTSD and CVD.5 Studies have found that circulating levels of renin are higher in people who are trauma-exposed, and persons who are on ACEs and ARBs report fewer PTSD symptoms.12 In addition, recent work has highlighted differences in the brain permeability of antihypertensive medications as a potential mechiansm through which some antihypertensive medications prevent psychopathology while others do not.11 In tandem with research that is needed elucidate the nuances of these associations in population-based and clinical samples, additional mechanistic research will be an important component in furthering our understanding of differences in the effects of these medications on PTSD incidence. The strengths of this study include its large population-based design, allowing us to examine four anti-hypertensive medication classes within the same sample; use of Danish registries in which comprehensive data captured for a full population minimizes selection bias; and access to a prescription registry that documents prescription redemption rather than prescription receipt. There are also several important limitations to consider. We focused on presciptions received in the 60 days prior to the traumatic event in an effort to balance sensitivity and specificity of this classification, but it is possible that there are other etiologically important time frames for antihypertensive medication prescription fills prior to trauma. In addition, while data on prescription redemption rather than prescription receipt provides an additional layer of confidence regarding medication consumption compared with traditional population-based registry data, we cannot be sure that persons in the study were taking these medications routinely at the time when the traumatic event occurred. In addition, despite the large original cohort, the variable definitions required for our analysis decreased the sample size considerably, making some estimates imprecise (i.e., with large confidence intervals). Although the captured traumatic events were the first in the period of this study, it is possible that participants had trauma history before the study period and we were unable to capture these data. Relatedly, this study relies strictly on diagnostic codes, and it is possible that some participants had subsyndromal psychopathology that would not be captured. “New user” designs have become an important methodologic approach in pharmacoepidemiology, but we were unable to implement this approach because the proportion of people who started an anti-hypertensive medication and happened to experience a traumatic event within the same 60 days was small, even within our large sample. For these reasons, replication of our findings using other larger samples from diverse populations and more rigorous causal methods (e.g., propensity score matching) is warranted. In addition, certain characteristics of Denmark (e.g., universal healthcare) may limit generalizability of these findings, and calls for replication in different populations.
Despite these limitations, this study contributes to the literature as the first longitudinal examination of associations between four classes of anti-hypertensive medications and PTSD incidence in a single population-based sample and can fruitfully point the way to future research in the field.
Clinical Points.
Evidence regarding a potential association between antihypertensive medications and PTSD is mixed, likely due to across study differences in medications examined, focus on PTSD incidence versus prevalence, and other methodologic choices.
Clinicians should be aware the different antihypertensive medications may have differential effects on posttraumatic psychopathology, with some medications potentially more beneficial than others.
sources of funding:
This work was supported by a grant from the National Institute of Mental Health (R01MH110453; PI: Gradus) and the National Heart, Lung and Blood Institute (R01HL160850; PIs: Sumner and Gradus). The funders had no role in the conduct or publication of this study.
Footnotes
Conflicts of interest: We have no conflicts of interest to disclose.
Relevant financial relationships: None.
References
- 1.Sumner JA, Cleveland S, Chen T, Gradus JL. Psychological and biological mechanisms linking trauma with cardiovascular disease risk. Transl Psychiatry. Jan 27 2023;13(1):25. doi: 10.1038/s41398-023-02330-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Argolo FC, Cavalcanti-Ribeiro P, Netto LR, Quarantini LC. Prevention of posttraumatic stress disorder with propranolol: A meta-analytic review. Journal of Psychosomatic Research. 2015;79(2):89–93. [DOI] [PubMed] [Google Scholar]
- 3.Cojocariu SA, Maștaleru A, Sascău RA, Stătescu C, Mitu F, Leon-Constantin MM. Neuropsychiatric consequences of lipophilic beta-blockers. Medicina. 2021;57(2):155. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Pitman RK, Sanders KM, Zusman RM, et al. Pilot study of secondary prevention of posttraumatic stress disorder with propranolol. Biological psychiatry. 2002;51(2):189–192. [DOI] [PubMed] [Google Scholar]
- 5.Seligowski AV, Duffy LA, Merker JB, et al. The renin–angiotensin system in PTSD: a replication and extension. Neuropsychopharmacology. 2021;46(4):750–755. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.American Psychological Association. Diagnostic and statistical manual of mental disorders (5 ed.). 2018; [Google Scholar]
- 7.Armstrong C, Kapolowicz MR. A preliminary investigation on the effects of atenolol for treating symptoms of anxiety. Military Medicine. 2020;185(11–12):e1954–e1960. [DOI] [PubMed] [Google Scholar]
- 8.Garakani A, Mathew S, Charney DS. Neurobiology of anxiety disorders and implications for treatment. Mount Sinai Journal of Medicine. 2006;73(7):941–949. [PubMed] [Google Scholar]
- 9.Levy A, Kadar T, Dachir S. An animal model for studying therapeutic drugs against post-traumatic stress disorder. Military Medicine. 2001;166(suppl_2):74–75. [PubMed] [Google Scholar]
- 10.Verma M, Bali A, Singh N, Jaggi AS. Investigating the role of nisoldipine in foot-shock-induced post-traumatic stress disorder in mice. Fundamental & Clinical Pharmacology. 2016;30(2):128–136. [DOI] [PubMed] [Google Scholar]
- 11.Colbourne L, Harrison PJ. Brain-penetrant calcium channel blockers are associated with a reduced incidence of neuropsychiatric disorders. Molecular Psychiatry. 2022/09/01 2022;27(9):3904–3912. doi: 10.1038/s41380-022-01615-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Khoury NM, Marvar PJ, Gillespie CF, et al. The renin-angiotensin pathway in posttraumatic stress disorder: angiotensin-converting enzyme inhibitors and angiotensin receptor blockers are associated with fewer traumatic stress symptoms. The Journal of Clinical Psychiatry. 2012;73(6):17451. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Nylocks K, Michopoulos V, Rothbaum AO, et al. An angiotensin-converting enzyme (ACE) polymorphism may mitigate the effects of angiotensin-pathway medications on posttraumatic stress symptoms. American Journal of Medical Genetics Part B: Neuropsychiatric Genetics. 2015;168(4):307–315. [DOI] [PubMed] [Google Scholar]
- 14.Marvar PJ, Goodman J, Fuchs S, Choi DC, Banerjee S, Ressler KJ. Angiotensin type 1 receptor inhibition enhances the extinction of fear memory. Biological Psychiatry. 2014;75(11):864–872. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Quiñones MM, Maldonado L, Velazquez B, Porter JT. Candesartan ameliorates impaired fear extinction induced by innate immune activation. Brain, Behavior, and Immunity. 2016;52:169–177. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Saavedra JM, Sánchez-Lemus E, Benicky J. Blockade of brain angiotensin II AT1 receptors ameliorates stress, anxiety, brain inflammation and ischemia: therapeutic implications. Psychoneuroendocrinology. 2011;36(1):1–18. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Shekhar A Angiotensin type 1 receptor antagonists—a novel approach to augmenting posttraumatic stress disorder and phobia therapies? Biological psychiatry. 2014;11(75):836–837. [DOI] [PubMed] [Google Scholar]
- 18.Stein MB, Jain S, Simon NM, et al. Randomized, placebo-controlled trial of the angiotensin receptor antagonist losartan for posttraumatic stress disorder. Biological Psychiatry. 2021;90(7):473–481. [DOI] [PubMed] [Google Scholar]
- 19.Gradus JL, Rosellini AJ, Szentkúti P, et al. Using Danish national registry data to understand psychopathology following potentially traumatic experiences. Journal of Traumatic Stress. 2022;35(2):619–630. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Schmidt M, Schmidt SAJ, Sandegaard JL, Ehrenstein V, Pedersen L, Sørensen HT. The Danish National Patient Registry: a review of content, data quality, and research potential. Clinical Epidemiology. 2015;7:449. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Mors O, Perto GP, Mortensen PB. The Danish psychiatric central research register. Scandinavian Journal of Public Health. 2011;39(7_suppl):54–57. [DOI] [PubMed] [Google Scholar]
- 22.Bliddal M, Broe A, Pottegård A, Olsen J, Langhoff-Roos J. The Danish medical birth register. European Journal of Epidemiology. 2018;33(1):27–36. [DOI] [PubMed] [Google Scholar]
- 23.Helweg-Larsen K The Danish register of causes of death. Scandinavian Journal of Public Health. 2011;39(7_suppl):26–29. [DOI] [PubMed] [Google Scholar]
- 24.World Health Organization. International statistical classification of diseases and related health problems: instruction manual. World Health Organization. 2004;2 [Google Scholar]
- 25.Svensson E, Lash TL, Resick PA, Hansen JG, Gradus JL. Validity of reaction to severe stress and adjustment disorder diagnoses in the Danish Psychiatric Central Research Registry. Clinical Epidemiology. 2015;7:235. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Charlson ME, Pompei P, Ales KL, MacKenzie CR. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. Journal of Chronic Diseases. 1987;40(5):373–383. [DOI] [PubMed] [Google Scholar]
- 27.Wasserstein RL, Schirm AL, Lazar NA. Moving to a world beyond “p < 0.05”. The American Statistician. 2019/03/29 2019;73(sup1):1–19. [Google Scholar]
- 28.Amrhein V, Greenland S, McShane B. Scientists rise up against statistical significance. Nature. Mar 2019;567(7748):305–307. [DOI] [PubMed] [Google Scholar]
- 29.Lash TL. The harm done to reproducibility by the culture of null hypothesis significance testing. American Journal of Epidemiology. Sep 15 2017;186(6):627–635. [DOI] [PubMed] [Google Scholar]
- 30.McShane BB, Gal D, Gelman A, Robert C, Tackett JL. Abandon statistical significance. The American Statistician. 2019/03/29 2019;73(sup1):235–245. [Google Scholar]