Abstract
Background
The association between posttraumatic stress disorder (PTSD) and mortality in patients undergoing implantable cardioverter‐defibrillator (ICD) placement has not been evaluated in US veterans.
Hypothesis
PTSD in veterans with ICD is associated with increased mortality.
Methods
We studied a retrospective cohort of 25 678 veterans who underwent ICD implantation between September 30, 2002, and December 31, 2011. Of these subjects, 3280 carried the diagnosis of PTSD prior to ICD implantation. Primary outcome was all‐cause mortality between date of ICD implantation and end of follow‐up (September 30, 2013). We used Cox proportional hazard models to compute multivariable adjusted hazard ratios with corresponding 95% confidence intervals for the relation between PTSD diagnosis and death following ICD placement.
Results
During a mean follow‐up of 4.21 ± 2.62 years, 11 015 deaths were reported. The crude incidence rate of death was 87.8 and 103.9/1000 person‐years for people with and without PTSD, respectively. We did not find an association between presence of PTSD before or after ICD implantation and incident death when adjusted for multiple risk factors (hazard ratio: 1.003, 95% confidence interval: 0.948–1.061). In secondary analysis, no statistically significant association was found.
Conclusions
In this retrospective cohort study among more than 25 000 veterans undergoing ICD implantation, almost 13% had a diagnosis of PTSD. Subjects with PTSD were significantly younger, yet they had a higher incidence of coronary heart disease, major cardiac comorbidities, cancer, and mental health conditions. We found no association between presence of PTSD before or after ICD implantation and incident death when adjusting for all covariates.
Keywords: Implantable Cardioverter‐Defibrillator, Posttraumatic Stress Disorder
1. INTRODUCTION
Posttraumatic stress disorder (PTSD) is a very common disorder, with lifetime prevalence in the general community between 4.5% and 7%.1 The prevalence of PTSD in US Veterans Affairs patients, especially in those who were in combat, is higher than in the general population.2 PTSD stressors or events that could have led to PTSD were traditionally associated with combat experience.3, 4 Later, the range of PTSD stressors was widened to include assaults, motor vehicle accidents, and even psychological trauma such as unexpected death of a family member or friend. These changes have been reflected in more recent editions of the Diagnostic and Statistical Manual of Mental Disorders. 5 Some investigators consider PTSD caused by serious medical illness a special subtype of PTSD with unique characteristics.6 It has been estimated that an individual during his or her lifetime has a 90% chance of encountering a PTSD stressor.7 The risk of developing PTSD, however, differs with different stressor types. Thus, being in combat or being a victim of rape or assault creates a much higher chance of developing PTSD than finding out about the death of a close relative.8
Learning about a diagnosis of a serious, potentially life‐threatening medical condition can cause PTSD. For example, it has been shown that the prevalence of PTSD after an acute myocardial infarction (MI) is about 15%.7
On the other hand, PTSD itself may contribute to other problems, among them coronary heart disease (CHD). Multiple studies have shown that PTSD is associated with increased incidence of CHD in both military veterans and civilian trauma survivors.9, 10, 11
A substantial minority of patients who undergo implantable cardioverter‐defibrillator (ICD) placement have PTSD.12, 13 PTSD prevalence in cardiac clinics ranges from 0% to 24%,14, 15 and in the ICD clinic it has been found to be about 20%.16 PTSD in patients undergoing ICD implantation could be caused by a medical condition leading to ICD implantation, such as acute MI, cardiac arrest, or cardiomyopathy, or may be unrelated to it.17 Predictors of developing PTSD 24 months after ICD implantation were peritraumatic dissociation, feeling of helplessness, depression, female sex, history of being resuscitated, and higher number of cardiac symptoms. Significant predictors of PTSD 4 years after implantation were low educational level, feeling of helplessness, alexithymia, experience of >5 shocks, and previous history of PTSD.18
PTSD is associated with ICD implantation and/or ICD shocks is unique in that the ICD patient is continuously exposed to the defibrillation threat. This type of PTSD is similar to PTSD caused other serious medical illnesses and is considered by some investigators to be different from PTSD caused by intense but time‐limited trauma, such as being in combat.6
When compared with subjects without PTSD following ICD implantation in a prospective study evaluating the general population, those subjects who had PTSD at the time of ICD implantation showed a higher mortality rate: PTSD conferred a 2.4× long‐term age‐ and sex‐adjusted mortality risk for patients with ICD.16 PTSD stressors in that study were limited to cardiac conditions; subjects with PTSD caused by other stressors were not analyzed.
A report published by the American Heart Association (AHA) has suggested that PTSD symptoms in ICD patients are of concern and may be associated with increased mortality at long‐term follow‐up.19
The primary objective of our study was to describe characteristics of veterans with and without PTSD undergoing ICD implantation and to compare mortality in a large cohort of veterans who had a diagnosis of PTSD caused by any stressor prior to implantation of ICD with the mortality of veterans without PTSD undergoing ICD implantation.
2. METHODS
2.1. Study population
By conducting a retrospective cohort study using data from the Veterans Affairs (VA) National Patient Care Database and regional healthcare databases within the New England Veterans Integrated Service Network (VISN‐1), we identified 26 639 veterans who underwent ICD implantation between September 30, 2002, and December 31, 2011. Three veterans were excluded because their age was missing, and 958 subjects were excluded because they had a diagnosis of schizophrenia. The remaining 25 678 subjects were used for the analysis. For these subjects, 26 539 observations were made, because 861 subjects developed PTSD following ICD implantation and were crossed over. Those subjects contributed to 2 observations, 1 for person‐time prior to developing PTSD and 1 after.
We used International Classification of Diseases, Ninth Revision (ICD‐9), as well as Current Procedural Terminology (CPT) codes and Healthcare Common Procedure Coding System (HCPCS) device codes to capture inclusion diagnoses, outcomes, and comorbid conditions. Of the identified veterans, 3280 carried a diagnosis of established PTSD (Table 1). The earliest date of the diagnosis of PTSD available was October 1, 1997.
Table 1.
Basic characteristics and group comparison
| Characteristic | Total | PTSD+ | PTSD– | P Value |
|---|---|---|---|---|
| No. of participants | 25 678 | 3280 | 22 398 | |
| Age, y | 65.5 ± 10.3 | 62.1 ± 9.0 | 66.2 ± 10.4 | <0.0001 |
| Sex | ||||
| M | 25 365 (98.8) | 3223 (98.3) | 22 142 (98.9) | |
| F | 313 (1.2) | 57 (1.7) | 256 (1.1) | |
| Alcohol dependence | ||||
| Present | 4714 (18.4) | 1189 (36.3) | 3525 (15.7) | <0.0001 |
| Absent | 20 964 (81.6) | 2091 (63.8) | 18 873 (84.3) | <0.0001 |
| Anxiety | ||||
| Present | 4987 (19.4) | 1479 (45.1) | 3508 (15.7) | <0.0001 |
| Absent | 20 691 (80.6) | 1801 (54.9) | 18 890 (84.3) | <0.0001 |
| Bipolar disorder | ||||
| Present | 792 (3.1) | 336 (10.2) | 456 (2.0) | <0.0001 |
| Absent | 24 886 (96.9) | 2944 (89.8) | 21 942 (98.0) | <0.0001 |
| Depression | ||||
| Present | 11 408 (44.4) | 2668 (81.4) | 8740 (39.0) | <0.0001 |
| Absent | 14 270 (55.6) | 612 (18.6) | 13 658 (61.0) | <0.0001 |
| Drug abuse | ||||
| Present | 2444 (9.5) | 786 (24.0) | 1658 (7.4) | <0.0001 |
| Absent | 23 234 (90.5) | 2494 (76.0) | 20 740 (92.6) | <0.0001 |
| CHD | ||||
| Present | 23 846 (92.9) | 3080 (93.9) | 20 766 (92.7) | 0.0135 |
| Absent | 1832 (7.1) | 200 (6.1) | 1632 (7.3) | 0.0135 |
| CHF | ||||
| Present | 22 123 (86.2) | 2797 (85.3) | 19 326 (86.3) | 0.1177 |
| Absent | 3555 (13.8) | 483 (14.7) | 3072 (13.7) | 0.1177 |
| DM | ||||
| Present | 12 936 (50.4) | 1780 (54.3) | 11 156 (49.8) | <0.0001 |
| Absent | 12 742 (49.6) | 1500 (45.7) | 11 242 (50.2) | <0.0001 |
| Cardiac arrest (or VF) | ||||
| Present | 2762 (10.8) | 371 (11.3) | 2391 (10.7) | 0.2723 |
| Absent | 22 916 (89.2) | 2909 (88.7) | 20 007 (89.3) | 0.2723 |
| HTN | ||||
| Present | 23 788 (92.6) | 3108 (94.8) | 20 680 (92.3) | <0.0001 |
| Absent | 1890 (7.4) | 172 (5.2) | 1718 (7.7) | <0.0001 |
| Hyperlipidemia | ||||
| Present | 22 365 (87.1) | 2969 (90.5) | 19 396 (86.6) | <0.0001 |
| Absent | 3313 (12.9) | 311 (9.5) | 3002 (13.4) | <0.0001 |
| Paroxysmal VT | ||||
| Present | 10 109 (39.4) | 1241 (37.8) | 8868 (39.6) | 0.0543 |
| Absent | 15 569 (60.6) | 2039 (62.2) | 13 530 (60.4) | 0.0543 |
| Smoking | ||||
| Present | 13 140 (51.2) | 2010 (61.3) | 11 130 (49.7) | <0.0001 |
| Absent | 12 538 (48.8) | 1270 (38.7) | 11 268 (50.3) | <0.0001 |
| Stroke | ||||
| Present | 5981 (23.3) | 791 (24.1) | 5190 (23.2) | 0.2322 |
| Absent | 19 697 (76.7) | 2489 (75.9) | 17 208 (76.8) | 0.2322 |
| TIA | ||||
| Present | 2210 (8.6) | 330 (10.1) | 1880 (8.4) | 0.0015 |
| Absent | 23 468 (91.4) | 2950 (89.9) | 20 518 (91.6) | 0.0015 |
| Renal failure | ||||
| Present | 3290 (12.8) | 359 (10.9) | 2931 (13.1) | 0.0006 |
| Absent | 22 388 (87.2) | 2921 (89.1) | 19 467 (86.9) | 0.0006 |
| Cancer | ||||
| Present | 12 493 (48.7) | 1740 (53.0) | 10 753 (48.0) | <0.0001 |
| Absent | 13 185 (51.3) | 1540 (47.0) | 11 645 (52.0) | <0.0001 |
Abbreviations: CHD, coronary heart disease; CHF, congestive heart failure; DM, diabetes mellitus; F, female; HTN, hypertension; M, male; PTSD, posttraumatic stress disorder; SD, standard deviation; TIA, transient ischemic attack; VF, ventricular fibrillation; VT, ventricular tachycardia.
Data are presented as n (%) or mean ± SD.
2.2. Outcome
The primary outcome was mortality from any cause between the date of ICD implantation and the end of the study, which was September 30, 2013.
2.3. Other variables
Data on demographics included sex and age; comorbidities included CHD, congestive heart failure (CHF), diabetes mellitus (DM), cardiac arrest (or ventricular fibrillation [VF]), hypertension (HTN), hyperlipidemia, paroxysmal ventricular tachycardia (VT), smoking, stroke, transient ischemic attack (TIA), renal failure, cancer, alcohol dependence, anxiety, bipolar disorder, depression, and drug abuse.
2.4. Statistical analysis
We computed follow‐up person‐time from the exposure assessment (ICD implantation) until death or censoring date (ie, the date of last available follow‐up). Baseline characteristics of the cohort were compared. For each categorical variable, if the condition was coded for at a clinic visit, it was assigned a value of 1; otherwise, it was coded as zero. We used Cox proportional hazard models to compute multivariable adjusted hazard ratios (HR) with corresponding 95% confidence intervals (CI), using participants without PTSD at baseline as the reference group. First we adjusted for age alone, then we added variables to the model based on their potential to be confounders, including major cardiac comorbidities such as CHD, CHF, cardiac arrest, DM, HTN, hyperlipidemia, paroxysmal VT, smoking, stroke, TIA, and VF. In additional models, we added renal insufficiency and cancer separately, and then together with cardiac comorbidities, as potential cofounders of the relation between PTSD and mortality. All analyses were conducted using SAS software, version 9.2 (SAS Institute, Inc., Cary, NC). The significance level was set at 0.05.
3. RESULTS
Table 1 shows the baseline demographics of 25 678 subjects (25 365 males and 313 females), among them 3280 veterans with PTSD. Mean age of the study participants was 65.5 ± 10.3 years. Subjects with PTSD were significantly younger than those without PTSD (62.1 ± 9.0 vs 66.2 ± 10.4 years; P < 0.0001). PTSD was more frequent among women undergoing ICD implantation (18.2%) than among men (12.7%). Veterans with PTSD had significantly higher rates of CHD, DM, HTN, hyperlipidemia, smoking, TIA, and cancer (P < 0.05 for each comorbidity). The frequency of CHF was high in our study population (86.2%) and not significantly different between the cohorts (P = 0.1). There was no statistically significant difference in frequency of cardiac arrest (or VF) and stroke. Veterans with PTSD, compared with veterans without PTSD, had a lower rate of renal insufficiency (10.6% vs 13.1%, P < 0.0001) and paroxysmal VT (37.8% vs 39.6%, P = 0.05).
The rate of mental health comorbidities was much higher in subjects with PTSD. The veterans with PTSD had >2‐fold frequency of alcohol dependence and depression (36.3% vs 15.7% for alcohol dependence and 81.4% vs 39.0% for depression), an almost 3‐fold frequency of anxiety (45.1% vs 15.7%), and a >5× higher rate of bipolar disorder (10.2% vs 2.0%). The rate of drug abuse was >3× higher in the PTSD population (24.0% vs 7.4%).
During a mean follow‐up of 4.21 ± 2.62 years, 11 015 deaths were reported. The crude incidence rate of death was 87.8 and 103.9/1000 person‐years for people with and without PTSD, respectively (Table 2). There was a reversed association between presence of PTSD at or before the time of ICD implantation intake and incident death in an unadjusted model (HR: 0.836, 95% CI: 0.791–0.884). In all other models, the association was no longer present (Table 2).
Table 2.
HRs and 95% CIs for death by PTSD status
| Cases/Person‐Years | Crude Incidence Rate (1000 Person‐Years) | HR (95% CI) | ||||||
|---|---|---|---|---|---|---|---|---|
| Unadjusted | Model 1a | Model 2b | Model 3c | Model 4d | Model 5e | |||
| PTSD absent | 9834/94676 | 103.87 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 |
| PTSD present | 1181/13449 | 87.81 | 0.836 (0.791–0.884) | 1.076 (0.958–1.072) | 1.001 (0.946–1.059) | 1.002 (0.947–1.061) | 1.000 (0.945–1.058) | 1.003 (0.948–1.061) |
Abbreviations: CHD, coronary heart disease; CHF, congestive heart failure; CI, confidence interval; DM, diabetes mellitus; HR, hazard ratio; HTN, hypertension; PTSD, posttraumatic stress disorder; TIA, transient ischemic attack; VF, ventricular fibrillation; VT, ventricular tachycardia.
Model 1: adjusted for age.
Model 2: adjusted for age, CHD, CHF, cardiac arrest, DM, HTN, hyperlipidemia, paroxysmal VT, smoking, stroke, TIA, and VF.
Model 3: adjusted for age, CHD, CHF, cardiac arrest, DM, HTN, hyperlipidemia, paroxysmal VT, smoking, stroke, TIA, VF, and renal failure.
Model 4: adjusted for age, CHD, CHF, cardiac arrest, DM, HTN, hyperlipidemia, paroxysmal VT, smoking, stroke, TIA, VF, and cancer.
Model 5: adjusted for age, CHD, CHF, cardiac arrest, DM, HTN, hyperlipidemia, paroxysmal VT, smoking, stroke, TIA, VF, renal failure, and cancer.
In secondary analysis, we investigated an association between the presence of PTSD at or before the time of ICD implantation intake and incident death in various subgroups of subjects separately, among them males, females, subjects with and without alcohol dependence, with and without anxiety, with and without bipolar disorder, with and without depression, and with and without drug abuse (Table 3). For 313 female study participants there was a reverse, but not significant, association between PTSD and incident death (HR: 0.560, 95% CI: 0.228–1.372 when adjusted for all covariates; Table 3). For all other subgroups of subjects, no significant association was found (Table 3).
Table 3.
HRs and 95% CIs for death by PTSD status in different subgroups of subjects
| Cases/Person‐Years | Crude Incidence Rate (1000 Person‐Years) | Males | ||||||
|---|---|---|---|---|---|---|---|---|
| Unadjusted | Model 1a | Model 2b | Model 3c | Model 4d | Model 5e | |||
| PTSD absent | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | ||
| PTSD present | 0.84 (0.795–0.889) | 1.016 (0.961–1.076) | 1.004 (0.948–1.062) | 1.006 (0.95–1.065) | 1.003 (0.948–1.062) | 1.006 (0.951–1.065) | ||
| Females | ||||||||
| PTSD absent | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | ||
| PTSD present | 0.438 (0.190–1.013) | 0.589 (0.251–1.379) | 0.562 (0.230–1.373) | 0.559 (0.228–1.371) | 0.563 (0.230–1.374) | 0.560 (0.228–1.372) | ||
| Younger than mean age of 65.5 years | ||||||||
| PTSD absent | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | ||
| PTSD present | 1.002 (0.93–1.078) | N/A | 0.962 (0.893–1.036) | 0.961 (0.893–1.035) | 0.957 (0.888–1.030) | 0.957 (0.889–1.031) | ||
| Older than mean age of 65.5 years | ||||||||
| PTSD absent | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | ||
| PTSD present | 1.058 (0.969–1.156) | N/A | 1.06 (0.970–1.158) | 1.061 (0.971–1.159) | 1.058 (0.968–1.156) | 1.060 (0.970–1.158) | ||
| Alcohol dependence absent | ||||||||
| PTSD absent | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | ||
| PTSD present | 0.813 (0.760–0.871) | 0.964 (0.900–1.032) | 0.954 (0.891–1.022) | 0.954 (0.891–1.022) | 0.954 (0.891–1.022) | 0.955 (0.891–1.022) | ||
| Alcohol dependence present | ||||||||
| PTSD absent | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | ||
| PTSD present | 0.915 (0.826–1.013) | 1.026 (0.925–1.138) | 1.016 (0.915–1.128) | 1.023 (0.921–1.135) | 1.02 (0.927–1.122) | 1.023 (0.921–1.136) | ||
| Anxiety absent | ||||||||
| PTSD absent | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | ||
| PTSD present | 0.844 (0.787–0.906) | 1.035 (0.963–1.111) | 1.029 (0.958–1.105) | 1.033 (0.962–1.109) | 1.029 (0.958–1.105) | 1.033 (0.962–1.110) | ||
| Anxiety present | ||||||||
| PTSD absent | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | ||
| PTSD present | 0.781 (0.709–0.860) | 0.912 (0.827–1.006) | 0.913 (0.827–1.007) | 0.914 (0.828–1.008) | 0.913 (0.827–1.007) | 0.914 (0.829–1.009) | ||
| Bipolar disorder absent | ||||||||
| PTSD absent | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | ||
| PTSD present | 0.841 (0.794–0.891) | 1.014 (0.956–1.075) | 1.003 (0.946–1.063) | 1.003 (0.946–1.064) | 1.002 (0.945–1.063) | 1.004 (0.947–1.065) | ||
| Bipolar disorder present | ||||||||
| PTSD absent | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | ||
| PTSD present | 0.807 (0.642–1.015) | 0.889 (0.705–1.122) | 0.859 (0.679–1.086) | 0.868 (0.686–1.098) | 0.859 (0.679–1.086) | 0.868 (0.686–1.098) | ||
| Depression absent | ||||||||
| PTSD absent | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | ||
| PTSD present | 0.883 (0.793–1.983) | 1.047 (0.940–1.166) | 1.058 (0.950–1.178) | 1.042 (0.935–1.160) | 1.058 (0.950–1.179) | 1.043 (0.936–1.162) | ||
| Depression present | ||||||||
| PTSD absent | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | ||
| PTSD present | 0.773 (0.722–1.827) | 0.930 (0.868–0.996) | 0.938 (0.876–1.006) | 0.950 (0.886–1.018) | 0.938 (0.875–1.005) | 0.950 (0.886–1.018) | ||
| Drug abuse absent | ||||||||
| PTSD absent | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | ||
| PTSD present | 0.847 (0.796–0.901) | 1.005 (0.945–1.070) | 0.991 (0.931–1.055) | 0.993 (0.933–1.057) | 0.991 (0.931–1.055) | 0.994 (0.934–1.058) | ||
| Drug abuse present | ||||||||
| PTSD absent | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | ||
| PTSD present | 0.982 (0.850–1.134) | 0.977 (0.846–1.129) | 0.999 (0.864–1.155) | 1.000 (0.865–1.157) | 0.987 (0.853–1.143) | 0.991 (0.857–1.147) | ||
Abbreviations: CHD, coronary heart disease; CHF, congestive heart failure; CI, confidence interval; DM, diabetes mellitus; HR, hazard ratio; HTN, hypertension; N/A, not available; PTSD, posttraumatic stress disorder; TIA, transient ischemic attack; VF, ventricular fibrillation; VT, ventricular tachycardia.
Model 1: adjusted for age.
Model 2: adjusted for age, CHD, CHF, cardiac arrest, DM, HTN, hyperlipidemia, paroxysmal VT, smoking, stroke, TIA, and VF.
Model 3: adjusted for age, CHD, CHF, cardiac arrest, DM, HTN, hyperlipidemia, paroxysmal VT, smoking, stroke, TIA, VF, and renal failure.
Model 4: adjusted for age, CHD, CHF, cardiac arrest, DM, HTN, hyperlipidemia, paroxysmal VT, smoking, stroke, TIA, VF, and cancer.
Model 5: adjusted for age, CHD, CHF, cardiac arrest, DM, HTN, hyperlipidemia, paroxysmal VT, smoking, stroke, TIA, VF, renal failure, and cancer.
4. DISCUSSION
We sought to describe the characteristics and mortality of veterans with and without PTSD undergoing ICD implantation. Among >25 000 veterans undergoing ICD implantation, almost 13% had a diagnosis of PTSD. PTSD was more frequent among women undergoing ICD implantation than among men, which is consistent with what has been reported previously.18
We observed a higher rate of CHD and different CHD risk factors such as DM, HTN, and hyperlipidemia among subjects with PTSD undergoing ICD implantation, despite PTSD veterans being 4 years younger than those without the disorder. Several previous studies have demonstrated a higher risk of CHD among subjects with PTSD.9, 10, 20, 21 A study analyzing US veterans undergoing coronary angiography 22 reported that patients with PTSD were younger, which is similar to our observations. The same study found that patients undergoing coronary angiography who had a diagnosis of PTSD were not at increased risk for subsequent cardiovascular events or death, which the authors explained by the younger age of PTSD subjects and the importance of age in estimating global risk of CHD.
Our findings do not suggest an association between PTSD before or after the time of implantation of an ICD and increased mortality among US veterans.
To the best of our knowledge, this is the first study to assess the effects of PTSD on mortality in US veterans undergoing ICD implantation. Although multiple studies have evaluated PTSD development in subjects following ICD implantation,12, 23, 24 ICD shocks,12 and cardiac events (stressors) leading to ICD implantation,16 we could not find any previous study that would specifically address subjects who have known PTSD caused by any stressor before ICD implantation.
Current literature tends to suggest increased overall mortality in subjects with PTSD.20 PTSD is associated with various changes in inflammatory pathways as well as in the sympathetic nervous system.25 Additionally, subjects with PTSD have a higher rate of mental health comorbidities, such as depression, anxiety, and bipolar disorder, and a higher rate of adverse lifestyle habits, such as alcohol and drug abuse and, possibly, decreased level of physical activity.26, 27 As mentioned earlier, an association between PTSD symptoms and CHD has been shown in veterans and in civilians in multiple studies,9, 10, 11, 28 including a prospective cohort study.29 Other studies have shown that negative childhood experiences such as adverse social conditions, abuse, neglect, household dysfunction, and others contribute to development of obesity, CHD, DM, and stroke later in life.30 Several studies showed an association between PTSD and metabolic syndrome.31, 32 Mediation effects of inflammatory markers are the most likely biological explanations of these associations 33 and may contribute to a negative effect of PTSD on cardiovascular health and overall mortality.
Subjects undergoing ICD implantation could be especially prone to developing or worsening of PTSD. Although the majority of our study population did not experience cardiac arrest and/or VF prior to ICD implantation (Table 1), an overwhelming majority of the subjects had a history of CHD and/or MI, which qualify as PTSD stressors. A review study summarizing PTSD in the wake of heart disease concluded that patients with coronary artery disease are at increased risk of PTSD, with a prevalence rate of about 15% in MI patients.7 Therefore, it is possible that the PTSD symptoms in some of our subjects were worsened by cardiac problems leading to the decision to implant an ICD. Some studies suggested that the stress of ICD implantation itself, or even pre‐ICD implantation discussion about a risk for a spontaneous life‐threatening arrhythmia resulting in death, could cause PTSD.34, 35
Following an ICD implantation, ICD shocks could occur, preventing sudden cardiac death but at the same time exposing shock recipients to develop some form of psychological distress.36 A comprehensive review of literature focusing on the impact of ICD shocks on various physiological outcomes was conducted.37 This review found 5 studies evaluating PTSD, all of which used self‐reported questionnaires. A statistically significant effect of ICD shocks on PTSD or PTSD symptoms was found in 3 studies.
Our study showed a reverse association between PTSD at the time of implantation of ICD and mortality in an unadjusted model that was no longer seen following adjustment for age. How could we explain these findings? The diagnosis of PTSD may lead to an earlier introduction to the VA healthcare system. A prior study suggests that subjects with PTSD use the medical system more frequently.38 This could lead to better medical surveillance, which may have caused an earlier recognition of cardiac problems requiring an ICD implantation. This, in turn could have counteracted the effect of PTSD on mortality. An alternative interpretation is that PTSD may be associated with cardiac conditions that require implantation of ICD earlier in life.
In our cohort of veterans, PTSD subjects were significantly younger but also had a greater incidence of major cardiac comorbidities, higher incidence of cancer, and higher incidence of mental health comorbidities such as alcohol dependence, anxiety, bipolar disorder, depression, and drug abuse. The latter risk factors, which have higher prevalence in PTSD, may respond to mental health interventions and may improve the quality of life in this population as well as reduce mortality.
Because there is a complicated relationship between PTSD and mental health comorbidities such as alcohol dependence, anxiety, bipolar disorder, depression, and drug abuse, we felt that adjustment for these comorbidities was not warranted. Additionally, we excluded subjects with schizophrenia from our analysis because the concurrent diagnosis of schizophrenia makes it much more difficult to establish a reliable diagnosis of PTSD.
Previous studies have assessed multiple risk factors for increased anxiety, reduced quality of life, and compromised adjustment to the device, among them age, sex, number and frequency of shocks, perception of support and family response, anger and anxiety, and optimism and positive health expectations. These studies were summarized in an article by Bostwick and Sola.39 In our large retrospective cohort, many of PTSD details were not available, which precluded us from further subanalysis based on type and number of traumatic events, severity of PTSD, PTSD timing in relationship with development of CHD, or age of ICD implantation. It has been described that psychosocial adjustment of younger ICD recipients is unique and quite different from that of older patients undergoing ICD implantation.40
4.1. Study limitations
Our study has several limitations. First, our population consisted only of military veterans, predominantly males, thus making it difficult to generalize our findings to the general population. Further studies that include higher percentage of women and nonveteran subjects are warranted and may corroborate our findings. Second, the study did not distinguish between different types of PTSD, which could have been caused by the stress of cardiac disease leading to the decision for ICD implantation, combat experience, other unrelated stressors, or a combination of several stressors together. Our study did not grade levels of PTSD based on severity. Among the factors that are known to increase anxiety, reduce quality of life, and compromise adjustment to the device, we only analyzed age and sex; we did not have the information, and therefore did not analyze, number and frequency of shocks, perception of support and family response, anger and anxiety, and optimism and positive health expectations.
Our study has several strengths. We have a large sample size, which improves the statistical power to detect smaller effects, and a long follow‐up period. The study includes both male and female veterans of a wide age range and from different geographic regions.
5. CONCLUSION
Among veterans undergoing ICD implantation, almost 13% had a diagnosis of PTSD. Subjects with PTSD were significantly younger, yet they had higher incidence of CHD, major cardiac comorbidities, cancer, and mental health conditions. Some of these results are consistent with previously reported findings, suggesting that PTSD is an independent risk factor for developing CHD. Our study found no association between PTSD and increased risk for mortality in this cohort of veterans. Future investigations, including prospective studies, are needed to confirm our results as well as to assess the effect of PTSD on other endpoints such as quality of life, frequency of hospitalizations, and cardiovascular death. If a positive association is found, it might be possible to improve these adverse outcomes by timely screening for PTSD and aggressive treatment for PTSD subjects.
Conflicts of interest
The authors declare no potential conflicts of interest.
Ofman P, Hoffmeister P, Kaloupek DG, et al. Posttraumatic stress disorder and mortality in VA patients with implantable cardioverter‐defibrillators. Clin Cardiol. 2018;41:652–659. 10.1002/clc.22945
REFERENCES
- 1. Pietrzak RH, Goldstein RB, Southwick SM, et al. Prevalence and Axis I comorbidity of full and partial posttraumatic stress disorder in the United States: results from Wave 2 of the National Epidemiologic Survey on Alcohol and Related Conditions. J Anxiety Disord. 2011;25:456–465. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Dohrenwend BP, Turner JB, Turse NA, et al. The psychological risks of Vietnam for U.S. veterans: a revisit with new data and methods. Science. 2006;313:979–982. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Fairbank JA, Nicholson RA. Theoretical and empirical issues in the treatment of post‐traumatic stress disorder in Vietnam veterans. J Clin Psychol. 1987;43:44–55. [DOI] [PubMed] [Google Scholar]
- 4. Foy DW, Carroll EM, Donahoe CP Jr. Etiological factors in the development of PTSD in clinical samples of Vietnam combat veterans. J Clin Psychol. 1987;43:17–27. [DOI] [PubMed] [Google Scholar]
- 5. Diagnostic and Statistical Manual of Mental Disorders. 5th ed Washington, DC: American Psychiatric Association; 2013. [Google Scholar]
- 6. Shalev AY, Schreiber S, Galai T, et al. Post‐traumatic stress disorder following medical events. Br J Clin Psychol. 1993;32(part 2):247–253. [DOI] [PubMed] [Google Scholar]
- 7. Spindler H, Pedersen SS. Posttraumatic stress disorder in the wake of heart disease: Prevalence, risk factors, and future research directions. Psychosom Med. 2005;67:715–723. [DOI] [PubMed] [Google Scholar]
- 8. Norris FH, Friedman MJ, Watson PJ. 60,000 disaster victims speak: Part II. Summary and implications of the disaster mental health research. Psychiatry. 2002;65:240–260. [DOI] [PubMed] [Google Scholar]
- 9. Vaccarino V, Goldberg J, Rooks C, et al. Post‐traumatic stress disorder and incidence of coronary heart disease: a twin study. J Am Coll Cardiol. 2013;62:970–978. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Boscarino JA. A prospective study of PTSD and early‐age heart disease mortality among Vietnam veterans: implications for surveillance and prevention. Psychosom Med. 2008;70:668–676. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Edmondson D, Kronish IM, Shaffer JA, et al. Posttraumatic stress disorder and risk for coronary heart disease: a meta‐analytic review. Am Heart J. 2013;166:806–814. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Kapa S, Rotondi‐Trevisan D, Mariano Z, et al. Psychopathology in patients with ICDs over time: results of a prospective study. Pacing Clin Electrophysiol. 2010;33:198–208. [DOI] [PubMed] [Google Scholar]
- 13. Marshall P, Ketchell A, Maclean J. Comparison of male and female psychological outcomes related to implantable cardioverter defibrillators (COMFORTID). Eur J Cardiovasc Nurs. 2012;11:313–321. [DOI] [PubMed] [Google Scholar]
- 14. Kutz I, Shabtai H, Solomon Z, et al. Post‐traumatic stress disorder in myocardial infarction patients: prevalence study. Isr J Phychiatry Relat Sci. 1994;31:48–56. [PubMed] [Google Scholar]
- 15. Cotter G, Milo‐Cotter O, Rubinstein D, et al. Posttraumatic stress disorder: a missed link between psychiatric and cardiovascular morbidity? CNS Spectr. 2006;11:129–136. [DOI] [PubMed] [Google Scholar]
- 16. Ladwig KH, Baumert J, Marten‐Mittag B, et al. Posttraumatic stress symptoms and predicted mortality in patients with implantable cardioverter‐defibrillators: results from the prospective Living With an Implanted Cardioverter‐Defibrillator study. Arch Gen Psychiatry. 2008;65:1324–1330. [DOI] [PubMed] [Google Scholar]
- 17. Tulloch H, Greenman PS, Tassé V. Post‐traumatic stress disorder among cardiac patients: prevalence, risk factors, and considerations for assessment and treatment. Behav Sci (Basel). 2014;5:27–40. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. von Känel R, Baumert J, Kolb C, et al. Chronic posttraumatic stress and its predictors in patients living with an implantable cardioverter defibrillator. J Affect Disord. 2011;131:344–352. [DOI] [PubMed] [Google Scholar]
- 19. Dunbar SB, Dougherty CM, Sears SF, et al; American Heart Association Council on Cardiovascular Nursing, Council on Clinical Cardiology, and Council on Cardiovascular Disease in the Young. Educational and psychological interventions to improve outcomes for recipients of implantable cardioverter defibrillators and their families: a scientific statement from the American Heart Association. Circulation. 2012;126:2146–2172. [DOI] [PubMed] [Google Scholar]
- 20. Ahmadi N, Hajsadeghi F, Mirshkarlo HB, et al. Post‐traumatic stress disorder, coronary atherosclerosis, and mortality. Am J Cardiol. 2011;108:29–33. [DOI] [PubMed] [Google Scholar]
- 21. Xue Y, Taub PR, Iqbal N, et al. Cardiac biomarkers, mortality, and post‐traumatic stress disorder in military veterans. Am J Cardiol. 2012;109:1215–1218. [DOI] [PubMed] [Google Scholar]
- 22. Bradley SM, Stanislawski MA, Bekelman DB, et al. Invasive coronary procedure use and outcomes among veterans with posttraumatic stress disorder: insights from the Veterans Affairs Clinical Assessment, Reporting, and Tracking Program. Am Heart J. 2014;168:381.e6–390.e6. [DOI] [PubMed] [Google Scholar]
- 23. Hamner M, Hunt N, Gee J, et al. PTSD and automatic implantable cardioverter defibrillators. Psychosomatics. 1999;40:82–85. [DOI] [PubMed] [Google Scholar]
- 24. Pedersen SS, van den Broek KC, Erdman RA, et al. Pre‐implantation implantable cardioverter defibrillator concerns and type D personality increase the risk of mortality in patients with an implantable cardioverter defibrillator. Europace. 2010;12:1446–1452. [DOI] [PubMed] [Google Scholar]
- 25. Baker DG, Nievergelt CM, O'Connor DT. Biomarkers of PTSD: neuropeptides and immune signaling. Neuropharmacology. 2012;62:663–673. [DOI] [PubMed] [Google Scholar]
- 26. Schnurr PP, Green BL. Understanding relationships among trauma, post‐traumatic stress disorder, and health outcomes. Adv Mind Body Med. 2004;20:18–29. [PubMed] [Google Scholar]
- 27. von Känel R, Hepp U, Kraemer B, et al. Evidence for low‐grade systemic proinflammatory activity in patients with posttraumatic stress disorder. J Psychiatr Res. 2007;41:744–752. [DOI] [PubMed] [Google Scholar]
- 28. Crum‐Cianflone NF, Bagnell ME, Schaller E, et al. Impact of combat deployment and posttraumatic stress disorder on newly reported coronary heart disease among US active duty and reserve forces. Circulation. 2014;129:1813–1820. [DOI] [PubMed] [Google Scholar]
- 29. Kubzansky LD, Koenen KC, Spiro A 3rd, et al. Prospective study of posttraumatic stress disorder symptoms and coronary heart disease in the Normative Aging Study. Arch Gen Psychiatry. 2007;64:109–116. [DOI] [PubMed] [Google Scholar]
- 30. Nandi A, Glymour MM, Kawachi I, et al. Using marginal structural models to estimate the direct effect of adverse childhood social conditions on onset of heart disease, diabetes, and stroke. Epidemiology. 2012;23:223–232. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Farr OM, Sloan DM, Keane TM, et al. Stress‐ and PTSD‐associated obesity and metabolic dysfunction: a growing problem requiring further research and novel treatments. Metabolism. 2014;63:1463–1468. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32. Bartoli F, Carrà G, Crocamo C, et al. Metabolic syndrome in people suffering from posttraumatic stress disorder: a systematic review and meta‐analysis. Metab Syndr Relat Disord. 2013;11:301–308. [DOI] [PubMed] [Google Scholar]
- 33. Su S, Wang X, Kapuku GK, et al. Adverse childhood experiences are associated with detrimental hemodynamics and elevated circulating endothelin‐1 in adolescents and young adults. Hypertension. 2014;64:201–207. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34. Prudente LA. Psychological disturbances, adjustment, and the development of phantom shocks in patients with an implantable cardioverter defibrillator. J Cardiovasc Nurs. 2005;20:288–293. [DOI] [PubMed] [Google Scholar]
- 35. Prudente LA, Reigle J, Bourguignon C, et al. Psychological indices and phantom shocks in patients with ICD. J Interv Card Electrophysiol. 2006;15:185–190. [DOI] [PubMed] [Google Scholar]
- 36. Sears SF, Kirian K. Shock and patient‐centered outcomes research: is an ICD shock still a critical event? Pacing Clin Electrophysiol. 2010;33:1437–1441. [DOI] [PubMed] [Google Scholar]
- 37. Manzoni GM, Castelnuovo G, Compare A, et al. Psychological effects of implantable cardioverter defibrillator shocks: a review of study methods. Front Psychol. 2015;6:39. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38. Stein MB, McQuaid JR, Pedrelli P, et al. Posttraumatic stress disorder in the primary care medical setting. Gen Hosp Psychiatry. 2000;22:261–269. [DOI] [PubMed] [Google Scholar]
- 39. Bostwick JM, Sola CL. An updated review of implantable cardioverter/defibrillators, induced anxiety, and quality of life. Psychiatr Clin North Am. 2007;30:677–688. [DOI] [PubMed] [Google Scholar]
- 40. Sears SF Jr, Burns JL, Handberg E, et al. Young at heart: understanding the unique psychosocial adjustment of young implantable cardioverter defibrillator recipients. Pacing Clin Electrophysiol. 2001;24:1113–1117. [DOI] [PubMed] [Google Scholar]