Late-life coronary heart disease mortality of Finnish war veterans in the TAMRISK study, a 28-year follow-up

There exist a multitude of studies that have investigated the impact of combat exposure on long-term health. A study from the Netherlands has provided evidence for increased late-life mortality among World War II military veterans approximately six decades later. Particularly wounded war survivors and those with posttraumatic stress disorder (PTSD) were at risk for death by any cause [1]. Vietnam veterans had statistically significant excesses of deaths from cancer and external causes, including motor vehicle accidents and accidental poisonings, in a 24-year retrospective analysis [2]. Among veterans of the Persian Gulf War, there was a significantly higher mortality rate than among veterans deployed elsewhere, but most of the increase was due to motor vehicle accidents rather than disease [3], an effect which dissipated after seven years of follow-up [4]. The same pattern has been observed also in UK Gulf War veterans [5]. The death rates of Australian Vietnam veterans were higher than those for other veterans for diseases of the digestive system, possibly indicating increased incidence of stress-related disorders [6].

There are only a few studies addressing the long-term cardiovascular consequences of participating in war combat. Furthermore, the results are inconclusive. Follow-up mortality rates of male traumatic lower limb amputees of the Israeli army and Vietnam veterans with PTSD had higher mortality rates due to cardiovascular disease, compared with the general population [7, 8]. However, recent results from the ARIC study reported that middle-aged veterans after an average of 36 years from combat exposure were not at increased cardiovascular risk compared to controls [9, 10]. Furthermore, death rates from circulatory system diseases did not differ between Vietnam veterans and their peers [11] and female Vietnam veterans had significant lower-than-expected deaths from circulatory diseases [12]. On the other hand, results from the studies by Modan, et al [7] and Boscarino [8] are not necessarily conflicting with the studies by Johnson et al [9, 10], Boehmer et al [11] and Thomas et al [12], as the former studies looked only at the subset of veterans with physical injury or PTSD, while the latter studies assessed all veterans exposed to combat.

We therefore wanted to pursue the hypothesis that wounded veterans who were alive at the age of 55 are more at long-term risk for coronary death than other veterans. The men in the present study were all born in 1925, and most of them had been involved in war-time activities when they were 17-20 years old. Now, over 60 years later, we have examined whether physical injuries in combat had long-term consequences on coronary mortality in this cohort.

Background characteristics of those wounded or injured (n = 102) and for those not wounded (n = 565) are shown in Table 1. Men who had been wounded or injured in action were taller and had slightly higher BMI, had more self-reported depression experienced during the last six months, and were more on disability pension, as compared with the comparison group at the age of 55.

In Cox proportional hazards analysis, wounded or injured men had a 1.7-fold (95% CI, 1.1-2.5; p = 0.01) increased risk of CHD death. The Kaplan-Meier survival curve illustrates the better survival of the comparison group compared with those who had been wounded or injured in action (Figure 1; p = 0.01). There were significant losses in the group of wounded men in the first 1-2 years of the study followed by a recovery towards years 5-10.

Since the wounded men were significantly taller and heavier and more depressed than their non-wounded peers, we also adjusted for BMI and depression in Cox proportional hazards analysis. After this adjustment, wounded or injured men still had a 1.7-fold (95% CI 1.1-2.5; p = 0.02) increased risk, while BMI only slightly increased the risk for CHD death (RR 1.1; 95% CI 1.0-1.1; p = 0.04), and depression had no effect (RR 0.8; 95% CI 0.5-1.2; p = NS). In men who had been wounded or injured in action the crude CHD mortality rate per 10,000 population was 2843, while in the comparison group the corresponding figure was 1961.

The effect of educational level on the results was also examined. When the information on being wounded or injured in action and the degree of education were both included in Cox proportional hazards analysis, wounded or injured men still had a 1.7-fold (95% CI 1.1-2.6; p = 0.01) increased risk of CHD, while the degree of education had no effect on CHD mortality (RR 0.90, 95% CI 0.77-1.01; p = NS).

Men from Tampere, Finland, who had been wounded or injured in action were 1.7 times more likely to die from CHD than men in the comparison group when over 55 years old. There are only a few previous studies touching on the same subject. Despite small sample size, we feel that this study is an important contribution to the literature because most studies to date have looked at the effects of exposure to combat-related PTSD rather than exposure to combat wound or injury. Further, this study assesses associations within a Finnish population, while the majority of other studies have looked at veterans from other countries (e.g., American, Australian and Middle Eastern populations). Our results are in accordance with a study from the Netherlands, which followed from 1992 to 2002 the overall late-life mortality levels of World War II military veterans, men and women, with birth years from 1920 to 1929. War survivors who had been seriously wounded were at a 1.7-fold risk to live shorter lives [1].

It is notable that there were significant losses in the group of wounded men in the first 1-2 years of the study followed by a recovery towards years 5-10, which might suggest bias towards severely ill individuals during the recruiting of cases in 1980. However, this seems unlikely, since the final cohort included 79% of all the 55-year-old men invited. On the other hand, it is highly probable that we have not seen the whole effect, since the present follow-up was started three decades after the traumatic experience. A follow-up of World War II veterans reported that already within 15 years after the war a subject would experience increased physical decline or death [13]. Likewise, 24-year follow-up mortality rates of male traumatic lower limb amputees of the Israeli army were 1.8-fold higher due to cardiovascular disease, compared with the general population [7]. In a 30-year follow-up, psychotraumatic events among Vietnam veterans were associated with postwar cardiovascular mortality with a hazard ratio of 1.7 [8].

Physical injury is particularly susceptible to concerns about whether trauma influences health outcomes because of some direct injury to the relevant health system, or via biologic damage due to ongoing stress resulting from the trauma. It may be that being wounded is a marker of exposure to severe trauma and only affects health when it induces PTSD or other forms of chronic stress that in turn impose physiologic wear and tear, particularly on the cardiovascular system. The wounded men in our study were more depressed than those in the comparison group, supporting this hypothesis. Effects of early life traumatic events on an individual's health have previously been proposed to extend across the life-course. Greater exposure to deaths of military comrades and exposure to war trauma at a younger age were associated with increased signs of physician-diagnosed cardiac, gastrointestinal, and nervous disease and more unique disease ailments across the life of American Civil War veterans [14]. The siege of Leningrad experienced at pre-pubertal age had also long-term effects leading to higher mortality, especially from ischemic heart disease and circulatory disease [15]. War veterans in this study were very young when exposed to war as soldiers. This may imply that many of them were exposed to war also when too young for the army as civilians, which often means bad nutrition in wartime.

The wounded men in our study had higher BMI and were more often on disability pension, as compared with the comparison group. In addition to depression, also these factors could possibly be intermediary variables between trauma/stress and a CHD death.

Socioeconomic status (SES) has been identified as a major risk factor for coronary disease [16]. A frequent measure of SES is education because it does not usually change after young adulthood. The degree of training had no effect on CHD risk in the present study. A confounding factor might be that over half of the men had no more than basic compulsory education.

In the present study, men who had been wounded or injured in action had higher BMI, as measured three decades after the war, compared to the comparison group. Continuous stressful conditions have previously been reported to lead to distinctive abdominal obesity, as observed in competing sailors during five month offshore sailing races [17]. Since early life traumatic experience may alter systems important in physiological stress regulation, a role for these systems has been speculated in explaining similar findings. In fact, former Finnish child evacuees of Word War II had higher prevalences of cardiovascular disease and type 2 diabetes at the age of 60 [18]. Subjects under continuous stress may develop metabolic alterations towards the metabolic syndrome despite regular exercise and a proper diet. This was especially seen in traumatic lower limb amputees of war. Surviving amputees had hyperinsulinemia, increased coagulability, and increased sympathetic and parasympathetic responses. These established CHD risk factors may thus explain the excess mortality due to CHD [7]. In line with these earlier observations, there was a trend towards an increased prevalence of diabetes and myocardial infarction in our subjects with physical war injuries, compared to the comparison group.

Complete information from 79% (667/843) participants and the long-term follow-up enhance the reliability of our findings. Another strength of the study is that there was no confounding by age or sex because the study group was restricted to Finnish men aged 55 years in 1980; however, this strength also poses a challenge to how broadly one can apply the findings. There are also concerns with exposure measurement, since there was minimal information on the type of injury. There may also be survivorship bias, since the present follow-up was started three decades after the traumatic experience.

Being physically wounded or injured in action predicted higher CHD mortality compared to the comparison group in a follow-up of Finnish war veterans starting three decades and ending six decades after World War II. Our finding supports the concept that the effects of traumatic events at early adulthood may have long-term consequences on an individual's health. Future studies should delve more into what mechanisms are at work in the observed associations.