In a recent report, Naghavi and coworkers 1 determined the frequency of antecedent influenza vaccination in 218 of their patients with coronary heart disease during the months of October 1997 through March 1998. Among 109 patients who experienced an acute myocardial infarction (AMI) in that period, the percentage who had been vaccinated in that current season was 47% versus 71% among 109 patients with coronary heart disease without AMI (odds ratio 0.33, 95% confidence interval 0.13 to 0.82, p 0.017). They suggested that vaccination against influenza might reduce the risk of AMI. The public health implications, if their observation is correct, are tremendous. Although the idea that vaccination could beneficially impact on AMI is novel and seemingly unfounded, there is a surprising amount of observational evidence to support the hypothesis. Disruption of the so-called “vulnerable plaque” with resultant intravascular thrombosis often leads to unstable coronary syndromes and most ischemic strokes. Plaque disruption and intraluminal thrombosis is not a random event. Purported triggers of plaque disruption have been reported in nearly 50% of patients with AMI. Plaque disruption is most likely caused by surges in sympathetic activity with a sudden increase in blood pressure, pulse rate, heart contraction, and coronary blood flow. Thrombosis then occurs on plaques when the systemic thrombotic tendency is high, as with platelet hyperaggregability or hypercoagulability, or thrombus may abnormally persist because of impaired fibrinolysis. It was first observed in the 1920s that acute vascular events occur more frequently in winter months. Since then, similar observations have been made in several temperate but geographically diverse areas. In 300,000 deaths from the Canadian Mortality Database, the incidence of fatal AMI and ischemic stroke was increased 19% and 20%, respectively, in January compared with the trough in September. A total of 259,891 cases of AMI from the Second National Registry of Myocardial Infarction, which occurred between July 1, 1994, and July 31, 1996, were analyzed by season. The incidence of AMI was 53% greater in winter than in summer, with case fatality rates following a similar seasonal pattern. Conversely, in a subtropical region with no seasonal temperature extremes, no seasonal variation of the incidence of AMI was observed. The winter peaks in AMI mortality have been correlated with temperature and shown to be greater among those with less personal protection from temperature extremes. Whereas no mechanism has been elucidated, many physiologic and biochemical changes associated with plaque disruption and intravascular thrombosis are known to correlate with temperature or season. Sympathetic tone, blood pressure, myocardial oxygen consumption, red blood cell count, hematocrit, platelet count, blood viscosity, and white blood cell count are all increased in winter. Cholesterol, antitrypsin, thromboglobulin, plasma cortisol, and C-reactive protein also increase in winter. Components of the clotting system show winter elevations including fibrinogen, factor VII activity, and platelet factor 4, while the fibinolytic system concurrently reaches a seasonal trough. Could the seasonal changes observed in both AMI incidence and relevant biochemical factors be due to seasonal infections? Systemic infection produces changes similar to those observed in cold weather. The link between infection and inflammation on the one hand and the immune response and coagulation on the other hand is phylogenetically ancient. Systemic infection results in elevations in blood pressure, white blood cell count, total and low-density lipoprotein cholesterol, tumor necrosis factor, interleukin-1, C-reactive protein, and fibrinogen. These changes certainly could either precipitate plaque disruption or predispose to intravascular thrombosis. In 1981, Pesonen and Siitonen were the first to call attention to an association between respiratory infections and AMI. This was followed in the United States by a matched case-control study, which showed an odds ratio of 2.1 for antecedent upper respiratory symptoms among patients with AMI. In a nested case-control study of 1,922 cases and 7,649 matched controls, significantly more cases (2.8%) than controls (0.9%) recalled having an acute respiratory tract infection in the 10 days before their index AMI (odds ratio 3.6, 95% confidence interval 2.2 to 5.7). In a database of 2,264 persons with AMI, 19% reported a flu-like illness or infection in the week preceding their AMI. To avoid recall bias, a nested case-control study of 3,172 male farmers used office records of visits to confirm that those with recurrent or chronic upper respiratory infections were at increased From the Division of Cardiovascular Disease, Department of Internal Medicine, University of Kansas School of Medicine, Kansas City, Kansas. Manuscript received October 9, 2001; revised manuscript received and accepted November 21, 2001. Address for reprints: David Meyers, MD, MPH, Division of Cardiovascular Disease, Kansas University Medical Center, 3901 Rainbow Boulevard, Kansas City, Kansas 66160-7378. E-mail: dmeyers@ kumc.edu.
Read full abstract