Post-traumatic stress disorder (PTSD) is gaining increasing recognition as a risk factor for morbidity and mortality. The aim of this study was to examine the impact of PTSD and abnormal cardiovascular biomarkers on mortality in military veterans. Eight hundred ninety-one patients presenting for routine echocardiography were enrolled. Baseline clinical data and serum samples for biomarker measurement were obtained and echocardiography was performed at the time of enrollment. Patients were followed for up to 7.5 years for the end point of all-cause mortality. Ninety-one patients had PTSD at the time of enrollment. There were 33 deaths in patients with PTSD and 221 deaths in those without PTSD. Patients with PTSD had a trend toward worse survival on Kaplan-Meier analysis (p = 0.057). Among patients with elevated B-type natriuretic peptide (>60 pg/ml), those with PTSD had significantly increased mortality (p = 0.024). Among patients with PTSD, midregional proadrenomedullin (MR-proADM), creatinine, and C-terminal proendothelin-1 were significant univariate predictors of mortality (p = 0.006, p = 0.024, and p = 0.003, respectively). In a multivariate model, PTSD, B-type natriuretic peptide, and MR-proADM were independent predictors of mortality. In patients with PTSD, MR-proADM was a significant independent predictor of mortality after adjusting for B-type natriuretic peptide, cardiovascular risk factors, cancer, and sleep apnea. Adding MR-proADM to clinical predictors of mortality increased the C-statistic from 0.572 to 0.697 (p = 0.007). In conclusion, this study demonstrates an association among PTSD, abnormal cardiac biomarker levels, and increased mortality.
The aims of this study were to evaluate the association between post-traumatic stress disorder (PTSD) and mortality and to determine clinical parameters and biomarkers that can help identify patients with PTSD at high risk for mortality. Biomarkers evaluated in PTSD-diagnosed and non-PTSD-diagnosed subjects were B-type natriuretic peptide (BNP), C-terminal preprovasopressin (copeptin), midregional proadrenomedullin (MR-proADM), and C-terminal proendothelin-1 (CT-proET1).
Methods
A total of 891 patients presenting to the VA San Diego Medical Center for routine outpatient echocardiography from June 2003 to November 2006 were enrolled in this study. This study was approved by the institutional review board of the University of California, San Diego. Patients were excluded if they were referred solely for ruling out intracardiac thrombus or vegetation. All patients were aged >18 years. Informed consent was obtained before enrollment. Echocardiograms, baseline medical histories, and serum samples for biomarker measurements were obtained on the day of enrollment. PTSD diagnoses were made by Veterans Affairs psychiatrists before enrollment. The patients were followed for up to 7.5 years by review of electronic medical records for the primary end point of all-cause mortality.
All blood samples were collected in ethylenediaminetetraacetic acid–containing plastic tubes by venipuncture. Plasma was separated within 60 minutes by centrifuge and stored at −70°C in plastic freezer vials. All biomarker analyses were performed at a central laboratory at the end of the study, with the exception of BNP, which was analyzed in monthly batches throughout the study. All blood samples were processed by personnel blinded to patient data.
BNP was measured using the ADVIA Centaur fully automated 2-site sandwich immunoassay (Bayer, Tarrytown, New York). The BNP assay had a limit of quantitation of 0.5 pg/ml and within-run imprecision of 3.1%. MR-proADM was measured using an automated sandwich chemiluminescence immunoassay on the KRYPTOR System (BRAHMS GmbH, Berlin, Germany). The MR-proADM assay had a limit of quantitation of 0.23 nmol/L and within-run imprecision of <5%. Copeptin was measured using a sandwich immunoluminometric assay (BRAHMS LUMItest CT-proAVP; BRAHMS GmbH). The copeptin assay had a limit of quantitation of 0.4 pmol/L and within-run imprecision of <10%. CT-proET1 was measured using a chemiluminescence sandwich immunoassay (CT-proET1 LIA; BRAHMS GmbH). The CT-proET1 assay had a limit of quantitation of 0.4 pmol/L and within-run imprecision of <5%.
Echocardiography was performed according to the guidelines of the American Society of Echocardiography. Two-dimensional echocardiograms were analyzed for wall motion abnormalities, left ventricular systolic and diastolic function, as well as the ejection fraction. All echocardiograms were interpreted by experienced cardiologists who were blinded to biomarker levels and PTSD status.
Values are expressed as mean ± SD or as counts and percentages as appropriate. All analyses were exploratory and used a p value of 0.05 for significance. Survival curves for patients with and without PTSD as well as in subgroups based on a predetermined BNP cutoff of 60 pg/ml were constructed using the Kaplan-Meier method and compared using log-rank tests. The BNP cutoff of 60 pg/ml was selected for increased risk for heart failure on the basis of results of previous screening studies of preclinical ventricular systolic and diastolic dysfunction. A multivariate Cox regression analysis for the mortality end point was conducted for PTSD, BNP, MR-proADM, copeptin, creatinine, and CT-proET1. The study biomarkers were also analyzed in separate univariate analyses for the mortality end point in patients with PTSD. To identify significant predictors of mortality in patients with PTSD in our small patient population, stepwise multivariate analyses were performed with MR-proADM, BNP, cardiovascular risk factors, cancer history, and history of sleep apnea.
Results
Among the 891 patients enrolled in the study, 91 patients were diagnosed with PTSD by their Veterans Affairs psychiatrists before enrollment. There were 33 deaths among the 91 patients with PTSD (36.3%) and 221 deaths among the 800 patients without PTSD (27.6%) during the 7.5-year follow-up period. Patients with PTSD were younger. Patients with PTSD also had less incidence of heart failure and fewer cardiovascular risk factors ( Table 1 ). However, patients with PTSD had a trend toward worse survival (p = 0.057; Figure 1 ) .
| Parameter | PTSD at Baseline | p Value | |
|---|---|---|---|
| No | Yes | ||
| Age (years) | 67 ± 12 | 66 ± 13 | 0.191 |
| Body mass index (kg/m 2 ) | 29 ± 6 | 29 ± 6 | 0.752 |
| Body surface area (m 2 ) | 2.1 ± 0.2 | 2.1 ± 0.2 | 0.857 |
| Systolic blood pressure (mm Hg) | 136 ± 20 | 134 ± 18 | 0.299 |
| Diastolic blood pressure (mm Hg) | 78 ± 12 | 79 ± 11 | 0.609 |
| Mean arterial blood pressure (mm Hg) | 95 ± 13 | 95 ± 12 | 0.986 |
| Serum creatinine (mg/dl) | 1.3 ± 1.5 | 1.1 ± 0.3 | 0.199 |
| Blood urea nitrogen (mg/dl) | 18 ± 10 | 16 ± 7 | 0.016 |
| BNP (pg/ml) | 134 ± 381 | 69 ± 82 | 0.127 |
| Log MR-proADM (nmol/L) | −0.14 ± 0.18 | −0.18 ± 0.16 | 0.087 |
| Log CT-proET1 (pmol/L) | 1.89 ± 0.18 | 1.86 ± 0.17 | 0.153 |
| Log copeptin (pmol/L) | 0.9 ± 0.4 | 0.8 ± 0.3 | 0.010 |
| Left ventricular end-diastolic diameter (cm) | 5.2 ± 0.9 | 5.1 ± 0.7 | 0.252 |
| Left ventricular end-systolic diameter (cm) | 3.5 ± 1.1 | 3.2 ± 0.7 | 0.010 |
| Ejection fraction (%) | 59.6 ± 13.2 | 63.4 ± 10.7 | 0.010 |
| Men | 780 (98%) | 89 (98%) | 0.860 |
| White | 635 (80%) | 70 (77%) | |
| Black | 70 (9%) | 10 (11%) | |
| Hispanic | 57 (7%) | 6 (7%) | |
| Asian | 28 (4%) | 2 (2%) | |
| Other | 9 (1%) | 3 (3%) | |
| Obesity ⁎ | 178 (22%) | 19 (21%) | 0.752 |
| Coronary artery disease † | 317 (40%) | 35 (39%) | 0.815 |
| Previous myocardial infarction | 193 (24%) | 16 (18%) | 0.159 |
| Angina pectoris | 92 (12%) | 10 (11%) | 0.878 |
| Arrhythmia | 214 (27%) | 15 (17%) | 0.033 |
| Cor pulmonale and right-sided heart failure | 5 (0.6%) | 0 (0.0%) | 0.449 |
| Previous coronary artery bypass grafting | 129 (16%) | 8 (9%) | 0.065 |
| Previous percutaneous intervention and stenting | 79 (10%) | 9 (10%) | 0.998 |
| Previous acute coronary syndromes | 26 (3%) | 4 (4%) | 0.555 |
| Hypertension | 570 (72%) | 71 (78%) | 0.190 |
| Pulmonary hypertension | 13 (2%) | 1 (1%) | 0.877 |
| Stroke | 61 (8%) | 4 (4%) | 0.259 |
| Chronic obstructive pulmonary disease | 129 (16%) | 11 (12%) | 0.312 |
| Hyperlipidemia ‡ | 471 (59%) | 53 (58%) | 0.886 |
| Diabetes mellitus | 265 (33%) | 28 (30%) | 0.634 |
| Renal dysfunction | 83 (10%) | 9 (10%) | 0.880 |
| Heart failure § | 216 (27%) | 14 (15%) | 0.016 |
| Diastolic dysfunction | 342 (43%) | 42 (46%) | 0.648 |
| Wall motion abnormality | 160 (20%) | 10 (11%) | 0.034 |
| Systolic dysfunction | 158 (20%) | 7 (8%) | 0.017 |
| Angiotensin-converting enzyme inhibitors | 383 (48%) | 35 (39%) | 0.273 |
| β blockers | 378 (48%) | 41 (45%) | 0.660 |
| Calcium channel blockers | 202 (25%) | 23 (25%) | 0.988 |
| Antiarrhythmic agents | 29 (4%) | 0 (0%) | 0.064 |
| Diuretics | 366 (46%) | 38 (42%) | 0.450 |
| Digoxin | 95 (12%) | 8 (9%) | 0.377 |
| Angiotensin blockers | 59 (7%) | 8 (9%) | 0.635 |
| Aldosterone inhibitors | 7 (1%) | 0 (0%) | 0.369 |
| Nitrates | 142 (18%) | 13 (14%) | 0.398 |
| Antilipidemic agents | 407 (51%) | 38 (42%) | 0.090 |
| Diabetic medications | 174 (22%) | 18 (20%) | 0.648 |
| Anticoagulants | 255 (32%) | 33 (36%) | 0.414 |
| Dyspnea | 487 (61%) | 59 (65%) | 0.480 |
| Edema | 271 (34%) | 29 (32%) | 0.689 |
| Smoking history | 336 (55%) | 38 (54%) | 0.088 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
