Another Hidden Cost of War?
The armed conflicts in Iraq and Afghanistan have brought new realities to military medicine in the types of battlefield injuries experienced by our soldiers, transitioning from the traditional predominance of penetrating wounds to a high proportion of blast injuries leading to multisystem trauma. Another change was a clear victory—startling success to unprecedented heights in the survival of battlefield casualties resulting from a highly coordinated and rapid escalation of care from the conflict theater to definitive life-saving interventions. As these conflicts now wind down, a reasonable question turns to the long-term medical impact to the fighting force to whom we, as a nation, pledge our support in sustaining their care in gratitude for their personal sacrifices. Among injuries exacting a long-term toll, traumatic brain injury has received considerable attention, as it should. However, the impacts of traumatic brain injury tend to be identifiable and relatively immediately apparent. In contrast, more chronic health conditions arising long after the sounds of war have faded and the physical injuries healed receive less attention as they blur into the fabric of ordinary medical conditions. Yet to fully honor the sacrifice and understand the true human impact of armed conflict, a reasonable question is whether traumatic injuries sustained in combat exact a longer term medical toll.
Article see p 2126
In this issue of Circulation, Stewart and colleagues,1 from a number of military medical organizations, report on the relationship between markers of injury severity and subsequent chronic cardiovascular risk factors and outcomes. Over a 9-year period, 3846 individuals sustaining critical injuries in Iraq or Afghanistan were identified and followed through administrative databases retrospectively for the incidence of 4 conditions—diabetes mellitus, hypertension, coronary artery disease (CAD), and chronic kidney disease. The study’s principle finding relates a measurement of the severity of the initial trauma (the injury severity score [ISS]) to these outcomes. Adjusting for a small number of potential confounding variables, higher values on the ISS (each 5-point increment on the scale ranging from up to a maximum value of 75) portended small to modest increases in the incidence of hypertension (6%), diabetes mellitus (13%), broadly-defined CAD (13%), and chronic kidney disease (15%). A supplemental observation included a 5-fold increase in the risk for chronic kidney disease if acute kidney injury occurred in the context of the trauma. Further compelling is that these outcomes were observed over a relatively brief period of 1.1 to 2.8 years of administrative follow-up.
The authors deserve credit for a careful effort to further define the potentially important, longer term cardiovascular impact of armed conflict. Although the relationship between the ISS and cardiovascular outcomes was small to modest across 5-point increments of the ISS, across the span of critical injury severity the impact has quantitative relevance relative to other common cardiovascular risk factors. Data on this topic are challenging to develop because of a number of factors, not the least of which are that prospective clinical studies are generally not programmed within armed conflicts, and the complex matrix of information sources extending from remote locations where battlefield trauma care begins to their stateside ongoing chronic care. Accepting these factors, the present work bears the attendant uncertainties arising from the nature of retrospective administrative data exploration. Caution is warranted in interpreting the study’s limited and imprecise associations because of a high risk of confounding. Quite simply, the ability to characterize competing cardiovascular risk variables, both medical and social, both before and after the traumatic injury, is limited. Although one may rationally believe that an index of trauma severity should be relatively independent of such confounding, even small inaccuracies resulting from the crude nature of the dataset could have impact on the interpretation because of the small number of outcomes leading to wide confidence intervals for all the findings, with most approaching 1. Surveillance bias, the concept that those most severely injured had more healthcare interactions in their rehabilitation phase, remains a significant risk to these data, particularly as diagnoses may be laden on subjects for detailed consideration of their disability considerations within the Veterans Affairs system. The crude nature of utilizing coding data for outcomes is well understood, and at best insert a level of noise within the data (for example, CAD encompassed a total of 58 International Classification of Diseases-9 codes). Few quantitative biological variables contributed to the analysis, and among them a single, nonstandardized data point on heart rate and blood pressure seem to be a crude estimate. In comparison with precursor conditions such as diabetes mellitus and hypertension, the heightened incidence for CAD observed over a short period of follow-up (<2.8 years) in such a young cohort is surprising and leads one to question whether inaccuracies in coding could be at play. In a search for consistency within the data, a notably absent association was between burns and subsequent outcomes, which may be simply a result of limited power, however in questioning the potential mechanisms (eg, chronic inflammation) for any relationship between trauma and cardiovascular outcomes such a finding would have been internally consistent. Lastly, a true, concurrent, and matched control population was not included in the analysis. Thus, the precise nature and magnitude of the relationship between trauma severity and subsequent cardiovascular risk and outcomes requires further confirmation and definition.
Assuming the relationships are true, what mechanisms might be at play? The authors explore these extensively, and at present they largely represent speculation. Whether mediated through stress and its known relationship to CAD, or inflammation and its adverse effects on endothelial function and coagulation, or other mechanisms is unknown. Opportunities to unravel these relationships might be possible in this cohort either through linking other diagnoses (eg, mental health disorders such as depression and anxiety) or serum biomarkers to observed outcomes. Continued surveillance of this cohort in a more prospective fashion may also prove fruitful to further characterize their health status and adjudicate ongoing outcomes.
In sum, the present study creates a hypothesis fully worthy of an answer. The present study leaves us beneath the threshold for a refined understanding of the long-term cardiovascular consequences of battlefield trauma that would be needed for informed discussions and decisions around veteran and military healthcare policy. However, the matter is consequential to our military and its people, and thus filling in the puzzle pieces to create a more full understanding is a matter of some urgency. If borne out as true and quantitatively meaningful, such a finding could alter decision making around disability status for individuals, and resources provided to military and veterans health organizations to most effectively care for their populations. Beyond efforts to continue to characterize and follow the present cohort, extending these analyses to battle injuries beneath the threshold of requiring intensive care unit care would enlarge the cohort and strengthen the findings, if not aid in defining an injury threshold that leads to long-term consequences. Considering that the present cohort represents approximately 5% of battle injuries, the total burden of cardiovascular risk to the population may be considerable, particularly if a widened range of relevant outcomes (eg, stroke) are included. Collaborating with civilian organizations to study whether similar relationships are found in noncombat traumatic injuries (such as motor vehicle collisions) could provide confirmatory support. Lastly, a more careful understanding between trauma severity and precursors to clinical CAD (such as subclinical atherosclerosis) would be mechanistically important and could be examined prospectively among service members who served in these conflicts.
The stakes for the question of battlefield injuries and risk for chronic health conditions are real, with consequences for the people who deserve an understanding of the personal cost of their injuries and sacrifices, and our institutions charged with caring for them. By extension, if found true, other forms of trauma may be also considered as a risk marker for future cardiovascular disease, similar to the manner in which chronic inflammatory conditions such as rheumatoid arthritis contribute to cardiovascular disease risk. We should pledge nothing short of a full effort to unravel this issue. Those who have served our country so bravely in the past, and those who will serve it in the future, deserve nothing less.
Dr Taylor is a retired officer from the United States Army Medical Corps. He discloses consulting or speaking relationships with Amgen, Lilly, and Sanofi.
The opinions expressed in this article are not necessarily those of the editors or of the American Heart Association.
- © 2015 American Heart Association, Inc.