Influenza represents a major periodic threat to health, from a yearly winter excess of mortality due to endemic influenza to the socially devastating pandemics resulting from antigenic shifts. This was evidenced in the H1N1 “Spanish” flu of 1918 [1,2], which resulted in the deaths of between 20 to 50 million people and exceeded the mortality of the First World War [3]. As well as having a major impact in endemic years, pandemic influenza is a constant threat, as demonstrated recently by the occurrence of the high pathogenicity and low infectivity H5N1 (“bird” flu) and the high infectivity but low pathogenicity of H1N1 (“swine” flu).
Throughout its recorded history, influenza has shared a close relationship with salicylates for their analgesic and antipyretic effects, originally used in the form of willow bark extracts and Aspen tea. Following its synthesis in 1853 by Charles Gerhardt, acetylsalicylic acid was industrially produced in 1899 [4] and marketed worldwide as Bayer aspirin. It achieved an almost universal popularity as a result of its antipyretic efficacy in the 1918 flu pandemic. It has been suggested, however, that its extensive use was associated with an increased mortality, potentially due to salicylate toxicity resulting in pulmonary edema caused by the high doses in use at the time [5]. Aspirin remained as the leading analgesic and antipyretic until the development of paracetamol and ibuprofen in the 1950s and 1960s. In 1971, John R. Vane demonstrated how aspirin works [6], describing its anti-inflammatory, analgesic, and antipyretic actions—an achievement for which he achieved the 1982 Nobel Prize for medicine. In the 1970s, researchers discovered how aspirin’s mechanism of action also leads to an antiplatelet effect, and it has since been used successfully in the secondary prevention of ischemic heart disease and cerebrovascular disease [7-10]. More recently, however, following the widespread use of statins in prevention of vascular disease, the additional benefit of aspirin in the primary prevention of cardiovascular disease in low- and moderate-risk patients has been called into question due to the risk of hemorrhagic complications [11].
Although there has been an enormous volume of aspirin used for both influenza and the prevention of cardiovascular events, the impact of aspirin on cardiovascular events in the context of influenza and related increased cardiovascular mortality is poorly understood, perhaps due to the temporal separation in the popularity of aspirin for the two indications. As such, no major guidelines on either influenza [12,13] or the use of aspirin for prevention of cardiovascular events discuss aspirin use in the context of influenza infection [14,15]. Nonetheless, a substantial number of patients at risk for cardiovascular disease have used aspirin for the symptomatic treatment of influenza, while similarly large numbers of patients taking aspirin for secondary prevention have been both vaccinated against and suffered from influenza. Although there have been no trials looking specifically at the effects of aspirin on influenza-related cardiovascular risk, it is possible that modifying the prescription of aspirin according to season or the occurrence of influenza might reduce the impact of influenza on the occurrence of cardiovascular events, both year-on-year and in the context of the massive amplification of disease burden during epidemics and pandemics. This article will consider the evidence in support of this hypothesis.
INFLUENZA AND CARDIOVASCULAR OUTCOMES
Observational epidemiology
The increase in all-cause winter mortality has been extensively investigated since it was first demonstrated in 1841 [16] and shown to be associated with age [17], housing standards [18], and winter temperature [19,20], varying from an excess mortality of 20 to 30% in countries across Europe. There are increases in both respiratory disease and cardiovascular disease, including ischemic heart disease and cerebrovascular disease [19]. There is an estimated increase in both myocardial infarction incidence and mortality of 30 to 40% in multiple populations including Japan [21], Scotland [22], and the United States [23], with a winter increase of 53% in the reporting of cases of myocardial infarction to the US National Registry [24].
The size of the winter excess in all-cause mortality is most closely related to the rate of influenza for that year [25,26], and peaks in all-cause mortality closely follow the temporal patterns of an influenza epidemic. The excess mortality due to influenza has averaged about 15,000 per year in Britain, reaching peaks of 30,000 to 40,000 in larger epidemics, with 65% of that occurring in patients over 65 years of age [25]. Estimates for mortality in the United States are over 50,000 on average per year, with up to 90% occurring in patients over 65 years of age [27]. Some groups have even estimated that as many as 92,000 preventable cardiovascular deaths per year in the United States are due to influenza triggering myocardial infarction [28].
Mortality due to influenza, however, cannot be measured with 100% accuracy. Studies of death certificate reports usually accurately classify primary viral pneumonia or secondary bacterial pneumonia as due to influenza, but only a small proportion of deaths resulting from nonrespiratory causes identify influenza as a contributing factor. Any association between subclinical influenza infection and cardiovascular morbidity, or cases where the certifying physician does not associate the clinical infection with the cardiovascular death, will not be identified. As such, mortality rates due to influenza are historically estimated as excessive during the influenza season—over and above the expected mortality—if the peak in influenza did not occur, estimated from years in which major epidemics did not occur. Using this approach, the number of excess deaths by cause can be analyzed in temporal association with the severity of the influenza epidemic for that year. A basic form of this approach was originally used by Farr in 1847 [29], but was more elegantly used by Collins [30] to measure the contribution of different diseases to overall mortality during multiple epidemics between the pandemic of 1918 and 1929 (Figure 1). In approximately 25 million inhabitants of 35 US cities, the monthly excess death rate for all causes was precisely temporally associated with that from pneumonia and influenza (a standard surrogate for respiratory-related influenza mortality), but there were simultaneous peaks in excess mortality due to all other causes combined. The number of excess deaths due to pneumonia and influenza was similar to the number from all other causes during all epidemics except the pandemics of 1918 and 1920. Subdividing “all other causes” in the smaller epidemics after 1920 showed that 46.4% of nonpneumonia and influenza deaths were certified as due to “organic” heart disease with no increase at all for the next leading cause of death overall—ie, cancer. However, it is important to note that in the two largest epidemics of 1918 and 1920, mortality due to organic heart disease was not significantly increased, constituting only 18.4% of nonpneumonia and influenza excess mortality. This may either reflect the increased coding of deaths as pneumonia and influenza with the increased prevalence of influenza, independent of actual cause, or the shift toward mortality in a young population in large influenza epidemics. It may also be relevant that autopsies were not common and forensic medicine was not developed early in the twentieth century.
 |
Figure 1. The close temporal association between deaths due to pneumonia and influenza during an epidemic, and deaths due to all other causes, as first demonstrated in 1932.
(Reproduced with permission from Collins SD. Excess mortality from causes other than influenza and pneumonia during influenza epidemics. Pub Health Rep 1932;47(46):2159-2179) |
The gap between the excess all-cause mortality due to influenza and that due to pneumonia and influenza during seasonal influenza or epidemics has since been replicated in multiple cohorts. In the Hong Kong/68 influenza A2 epidemics of 1968 and 1969 and 1969 and 1970, Miller et al [31] estimated there were an excess 31,100 and 46,900 deaths, respectively, in Britain, but only 12,300 (39.5%) and 32,600 (69.5%) of these were certified as due to “influenza, bronchitis, or pneumonia.” In the epidemic of 1976, there were 2.5 times the number of excess deaths attributed to other causes and during the 1990 epidemic there were 10 times the number. Only 7700 of 26,080 (29.5%) deaths in the 1990 epidemic were attributed to influenza or pneumonia [32]. In the United States, the all-cause excess annual mortality has been estimated at a mean of 51,203 between 1990 and 1999, but only 8097 (15.8%) were estimated due to primary influenza or pneumonia. There were significantly higher rates in those aged >65 years, and therefore an overall increase with time in an aging population [27]. In an extensive analysis of the underlying cause of the increase in influenza-associated mortality, Reichert et al [26] found that ischemic heart disease was the predominant cause, followed by cerebrovascular and respiratory disease.
The gap between all-cause mortality and pneumonia and influenza excess in “hard” mortality data is replicated in surrogates of morbidity, with an average of 133,900 (30,757 to 271,529) influenza-associated admissions per influenza season in the United States between 1979 and 2001 defined as due to pneumonia and influenza. Of those due to respiratory plus circulatory admissions, there were 294,128 (86,494 to 544,909) [33], replicating previous estimates from earlier periods [34,35], and again demonstrating significantly increased mortality and morbidity in those aged >65 years.
These estimates in epidemiological studies use approximated data of the “normal” mortality in low influenza years to calculate statistical models in defining the excess mortality. They, therefore, are subject to confusion by covarying rates in other conditions (eg, respiratory syncytial virus) and winter temperatures, although these are adjusted for as far as can be achieved. In addition, they use direct coding of deaths or admissions by medical staff to determine the proportion due to respiratory disease or otherwise. Thus, significant ascertainment errors occur, both in erroneously coding influenza-associated respiratory deaths as noninfluenza causes such as heart failure, and in coding cardiovascular deaths as due to influenza itself. This error rate is demonstrated by an increased preference of coders to cite influenza as the primary cause later in an epidemic than at the start [36] and the increased proportion of deaths ascribed to pneumonia and influenza in the larger epidemics as compared to the smaller ones [30]. However, this is unlikely to explain the discrepancy fully. In addition, the relative accuracy of these methods is demonstrated by the temporal relationship within 2 weeks between increased influenza rates and a high autopsy-proven rate of ischemic heart disease and myocardial infarction. In a large series of deaths in St.Petersburg, Russia, the odds for acute myocardial infarction or chronic ischemic heart disease at autopsy were 1.3 (95% CI 1.08-1.56) for influenza epidemic weeks versus off-season weeks [37].
Direct temporal relationships have been shown between nonspecific respiratory illness and cardiovascular events. A case-control analysis of 1922 cases of myocardial infarction nested within the UK General Practice Research Database demonstrated a significantly high rate of respiratory tract infection in cases for up to 15 days prior to myocardial infarction, but not over a longer lag period, with an overall relative risk of 2.7 (1.6-4.7). This research was susceptible to misclassification of symptoms, since early cardiac ischemia was misdiagnosed as an upper respiratory tract infection (URTI) [38]. However, there was no increased risk in urinary tract infection. This association was also shown in a small prospective series of patients with myocardial infarction from the United States, although this was susceptible to ascertainment error, particularly in excluding controls with respiratory disease [39]. This was again noted in a prospective series of Finnish farmers presenting with symptoms of upper respiratory tract infection who then developed a myocardial infarction [40].
The impact of vaccination
Stronger evidence for a causal relationship between influenza and cardiovascular events is demonstrated by the effect of influenza antiviral treatment or vaccination. In a meta-analysis of 20 observational studies, influenza vaccination was associated with a 68% lower rate of all-cause mortality and a 56% lower rate of pneumonia than it was without vaccination [41]. A number of studies have suggested that a significant proportion of this low mortality rate may be related to reduced rates of cardiovascular events. This is especially true in high-risk populations, such that influenza vaccination is now routinely recommended for patients at high risk for coronary ischemia [42,43]. In a retrospective cohort studied during the influenza season, Nichol et al [44] demonstrated a 19% lower risk of hospitalization for cardiac disease associated with influenza vaccination, 16 to 23% lower risk of cerebrovascular disease, 29 to 32% lower risk of pneumonia, and 48 to 50% lower mortality. These lower rates occurred even in the face of higher comorbidities and greater age in the vaccinated group. Similarly, retrospective case-control studies reported lower rates of patients with myocardial infarction having previously been vaccinated (odds ratio [OR] = 0.33, 95%CI 0.18-0.82) [45] and lower rates in patients with primary cardiac arrest (OR = 0.51, 0.33-0.79) [46] or stroke (OR = 0.50, 0.26-0.94) [47]. These results, by their nature, do not discriminate between protection or case selection. However, as “high-risk” patients with multiple comorbidities including cardiovascular disease are more likely to be vaccinated, the opposite association might be expected. Other confounding variables were present in these studies, which could potentially result in the effect demonstrated, such as differential uptake of vaccination by educational status or current employment. In addition, vaccination was not found to be associated with a low rate of recurrent coronary events in another prospective cohort study, although this small study was also prone to bias due to misclassification [48]. Finally, a large retrospective cohort study demonstrated that prescription of oseltamivir for influenza versus no treatment was associated with a 28% lower rate of stroke, although this was heavily confounded by a higher premorbid history of stroke, hypertension, and advanced age in the control group, which may not have been fully adjusted for in the analysis [49].
The strongest evidence for the impact of influenza vaccination on outcome is provided for by the open, randomized, controlled Flu Vaccination Acute Coronary Syndromes (FLUVACS) trial [50], which recruited 301 patients post-myocardial infarction or coronary angioplasty. Over 1 year of follow-up, there was a significantly reduced cardiovascular mortality in the vaccinated group (6 vs 17%, p = .002, hazard ratio [HR] = 0.34, 0.17-0.71), although this patient group is clearly high-risk and extrapolation of effects to low-risk groups is not necessarily reliable.
Overall, there is strong evidence of seasonal variation in morbidity and mortality due to cardiovascular disease, which is multifactorial in nature, but the severity of which is strongly correlated with the rate of influenza for that year (Panel 1). The excess number of deaths related to influenza is not adequately explained by respiratory complications, and the large nonrespiratory excess deaths are predominantly due to cardiovascular disease. It is unclear from current evidence to what extent these represent de novo cardiac events in mild-to-moderate risk patients versus high-risk patients, many of whom have already experienced events, and to what extent this represents primary cardiac disease in the absence of any clinical influenza versus cardiac complications secondary to a clinical episode of influenza. A causal relationship is suggested by the temporal relationship between acute respiratory infections and cardiovascular events and by the reduced incidence of cardiovascular events following influenza vaccination, especially for myocardial infarction, stroke, and cardiac failure. It is this increase in cardiovascular mortality due to influenza that is potentially treatable or preventable by aspirin.
 |
PANEL 1. Summary points about the association between influenza and cardiovascular disease |
THE PATHOPHYSIOLOGY OF INFLUENZA AND ASPIRIN
Influenza and cardiovascular disease
The role of inflammation in the pathogenesis and destabilization of atherosclerotic plaques has been extensively debated, from the identification of a primarily inflammatory response in the earliest manifestation of atherosclerosis, the “fatty streak” [51], to the roles of both the innate and adaptive immune systems in the generation and destabilization of atherosclerotic plaques [52]. The latter is partly through the interaction between inflammatory cells and lipid constituents of the blood such as low-density lipoprotein-cholesterol, potentially through its oxidation or glycation [53]. Infectious agents such as herpes viruses [54] and Chlamydia pneumoniae [55] have been identified within atherosclerotic plaques, although proposed causal links have not been confirmed by the development of atherosclerosis in animal transfer studies, and antibiotic treatment has not reduced atherosclerotic disease in randomized controlled trials [56,57]. The clinical relevance of inflammation has been suggested by the increased risk of ischemic heart disease and stroke seen in multiple cohorts of patients with raised biomarkers of inflammation, particularly high-sensitivity C-reactive protein (hsCRP), independent of other major risk factors. In addition, the reduction in cardiovascular risk due to aspirin has been shown to be proportional to the hsCRP level [58]. Finally, in the large randomized, controlled Justification for the Use of statin in Prevention: an Intervention Trial Evaluating Rosuvastatin (JUPITER), patients with a raised hsCRP, independent of hyperlidipemia, significantly benefited from treatment with rosuvastatin [59].
Influenza could increase the risk of cardiovascular events through three potential mechanisms:
- Through a direct effect of the virus on the induction or destabilization of atherosclerosis or via prothrombotic effects
- Through the generation of a potent inflammatory response destabilizing existing cardiovascular disease, whether atherosclerotic, hypertensive, prothrombotic, or via another mediator of cardiovascular events
- Through increasing the physiological demand during an influenza infection, therefore increasing myocardial or cerebrovascular demand beyond the capacity of a patient’s physiological reserve, in the presence of underlying cardiovascular disease
Influenza virus has recently been demonstrated in atherosclerotic vessel walls with a systemic inflammatory response following intranasal infection of mice [60] and has been found in atherosclerotic lesions from the aortas of patients undergoing coronary bypass [61]. Beyond these limited findings, there has been no clear pathological demonstration of a direct action of influenza virus on the development or destabilization of atherosclerotic plaques.
The pathogenicity of influenza is related to the severity of the inflammatory response to infection. The H5N1 outbreak of bird flu was associated with a very high mortality, although low infectivity, due to the generation of a cytokine “storm” [62,63]. Influenza infection has been shown to increase the trafficking of macrophages into arterial walls [64], and to cause smooth muscle proliferation and T-cell infiltration in atherosclerotic plaques [65]. In addition, T-cells with a greater response to influenza-derived antigens have been reported to be more common in symptomatic carotid atherosclerotic plaques [66]. These reasons all demonstrate a possible link between inflammation and atherosclerosis in the context of influenza. Given the excessive cytokine storm seen in some influenza infection, the severity of this association may be enhanced beyond that seen in other infections [67]. Whether the relationship is due to the destabilization of preexisting cardiovascular disease or the development of atherosclerosis is unclear. The former hypothesis would be suggested by the temporal relationship between influenza infection and the occurrence of cardiovascular events, as described in epidemiological studies, but it has also been proposed that influenza has a more chronic role in the pathogenesis of atherosclerosis. One hypothesis proposes that the 1918 to 1919 influenza pandemic explains the rise and fall of atherosclerotic disease in the twentieth century, which probably occurred independently of the advent of medical interventions [68]. This hypothesis proposes that the greater impact of the virus in certain subpopulations resulted in a selected group of survivors, many of whom survived infection. Therefore, this population was different in composition relative to a nonselected population and also may have undergone physiological changes due to influenza infection, which resulted in the development of atherosclerotic disease and its complications in later life. Indirect evidence for this hypothesis is found in one report stating a greater likelihood of patients with myocardial infarction have influenza A antibodies compared to controls [69], although the cases and controls differed in key demographic respects, including age.
Influenza has also been shown to be associated with activated thrombotic states, with effects on both the endothelium and on the constituents of the blood. There is increased platelet aggregation and morphological changes with decreased platelet numbers [70], suggesting increased consumption, associated with consumption of clotting factors and the production of degradation products [71], and a procoagulant endothelium [72].
Possible mechanisms of action for aspirin
The pharmacological actions of aspirin are determined by two compounds—the reactive acetate moiety and the metabolite salicylate. These two compounds contribute to the multiple cyclooxygenase (COX)-dependent and COX-independent pharmacological actions of aspirin. COX-1 is the dominant isoform in platelets—it is responsible for the generation of thromboxane A2. COX-2 is involved in the formation of prostacyclins and prostaglandins such as PGE2 and PGI2.
At low doses, aspirin inhibits COX-1 (and thromboxane A2 formation). Inhibition of COX-2 by aspirin can be observed at high doses, which are not used in cardiovascular disease prevention. Aspirin exhibits an affinity for COX-1 approximately 160 times greater than it does for COX-2. Aspirin could specifically reduce the impact of influenza at a mechanistic level by the direct inhibition of the action or propagation of influenza, by specifically impairing the immune response to influenza and therefore reducing the inflammatory milieu that occurs during infection, or by counteracting its prothrombotic effects. Influenza has been shown to require activation of IkappaB kinase/NF-kappa B (IKK/NF-kB) for its efficient replication and propagation [73,74], and this is inhibited by high serum concentrations of acetylsalicylic acid [75]. Furthermore, the pronounced cytokine response to influenza has been related to targets of aspirin inhibition, both COX-2 mediated mechanisms, particularly in the context of the H5N1-mediated cytokine storm [76], and by the IKK/NF-kB pathway [77]. The coadministration of an antiviral and selective COX-2 inhibitors resulted in prolonged survival in H5N1-infected mice compared to antiviral treated mice alone, with lower concentrations of pulmonary inflammatory cytokines and T-cell lymphopenia [78], although the reduced rate of pulmonary viral clearance in some mice indicates the careful balance between suppressing an excessive immune response and oversuppressing an appropriate one.
Aspirin (via Cox and non-COX mediated pathways) may act by nonspecifically reducing the overall inflammatory reaction to influenza and so reducing the inflammatory milieu and prothrombotic state. Evidence for the impact of aspirin on the inflammation-dependent response of the vascular endothelium is seen in venous plethysmography studies [79], indicating a specific function for aspirin in preventing endothelial dysfunction independent of the causative agent. Similarly, aspirin has a potent antiplatelet effect, which functions against activated platelets, independent of the specific stimulus. In reducing the inflammatory response, aspirin will also reduce the physiological demands upon the body, thereby reducing demand-induced myocardial or cerebrovascular ischemia during any systemic stress, including influenza.
Aspirin and influenza vaccination
Aspirin may have a direct impact on the propagation or pathogenicity of influenza, but just as there are concerns that it may reduce the inflammatory response and therefore affect clearance of the virus, it is possible that concomitant treatment with aspirin might reduce the efficacy of influenza vaccination. Limited evidence for this comes from in vitro studies demonstrating that nonsteroidal anti-inflammatory drugs (NSAIDs) impair the capacity of peripheral blood mononuclear cells to generate an antibody response [80] and that aspirin can induce immune tolerance, reducing the responsiveness of antigen-presenting cells by decreasing the expression of costimulatory molecules [81]. However, this theoretical evidence has not been tested in clinical studies, and the only randomized controlled trial of the effect of aspirin on the immune response to influenza vaccination demonstrated that in 281 adults >65 years of age, a greater proportion of patients given aspirin generated a more than fourfold gamma G immunoglobulin (IgG) response to influenza [82] than did patients receiving placebo. However, this one small randomized controlled trial was too small to demonstrate a difference in either influenza or cardiovascular outcomes, and therefore further research is required to answer this question.
ASPIRIN DURING INFLUENZA EPIDEMICS
Aspirin gained international popularity due to its efficacy in reducing the inflammatory response to influenza infection in the 1918 pandemic. However, no accurate evidence exists documenting the relationship between the use of aspirin in this pandemic, or later epidemics, and outcome. Moreover, aspirin doses of several grams per day were given routinely at that time. It is notable that the 1918 pandemic did not show the large excess of cardiovascular deaths seen in seasonal influenza and other epidemics with the majority of deaths coded as pneumonia and influenza. Therefore, any apparent reduction in cardiovascular death due to aspirin would be unlikely to have been observed. However, in its long history, no active trials have been performed to assess the impact of aspirin on outcome in influenza. Indeed, only a small number of studies have assessed its impact on upper respiratory tract symptoms [83,84], fever [85], or pain itself [86], although no studies have demonstrated a significant risk of harm or the prolongation of the illness [87] at current therapeutic doses.
A number of large randomized trials have been carried out in various patient groups looking at the efficacy of aspirin on the reduction of cardiovascular events, and many of these have been running during influenza epidemics. Unfortunately, no trial has identified influenza as a specific outcome of interest, but given the size of these trials there may well be sufficient information to correlate differences in cardiovascular events during epidemics with the use of aspirin.
As described earlier, there is a seasonal increase in the incidence and mortality due to cardiovascular events, which appears to be due in large part to the impact of seasonal influenza and influenza epidemics. The mechanism of this increase is not completely understood and therefore neither the specific nor nonspecific (and COX and non-COX-mediated) actions of aspirin to reduce this risk can be assumed. Although aspirin may have a specific effect on the propagation of influenza and an added benefit in the reduction of events by the inhibition of the cytokine response, inflammatory milieu, and prothrombotic state seen in influenza, the lack of clinical evidence means that we cannot infer a specific role for aspirin in its direct treatment.
We know that aspirin reduces the rate of cardiovascular events in studies of both primary and secondary prevention, although in the former the evidence is unclear as to the level of risk of hemorrhage versus benefit of current best medical treatment [11]. Given that there is a relatively consistent relative risk reduction in ischemic cardiovascular events due to aspirin, the absolute risk reduction is likely to be mainly dependent upon the absolute population risk. We have shown that the absolute risk of cardiovascular events is greater during an influenza epidemic, with no known increased relative risk of hemorrhage. Therefore, if the relative risk reduction of aspirin is maintained, then the absolute risk reduction due to aspirin is likely to be greater. For an individual patient, this implies that his or her estimated cardiovascular risk upon which treatment decisions are made is not a static entity but changes with season and particularly with the occurrence and severity of an influenza epidemic. It can thus be argued that the indication for treatment with aspirin should change as well. Current guidelines in the United States [14], Britain [15], and Europe [88] propose that patients with a >20% cardiovascular disease risk over the next 10 years, or 10% coronary heart disease risk over 10 years, should be actively considered for treatment with low-dose aspirin. These guidelines do not consider the recent meta-analysis questioning the benefit of aspirin in primary prevention, although this was limited by relatively small numbers of patients at high cardiovascular risk. In fact, in the moderate-risk group (5-year chronic heart disease risk of 5 to 10%) aspirin in addition to other drugs would still have a greater absolute risk reduction than the increase in hemorrhage rate (0.8 vs 0.4%), while the risks were balanced in the high-risk group [11]. However, the relatively small number of outcomes makes these estimates difficult to judge accurately. Assuming a balance of risk and harm in the moderate-risk group for year-round treatment with aspirin, the potential reduction in risk by winter treatment alone may shift the balance in favor of treatment (Panel 2). This assumes no parallel increase in hemorrhagic risk, which may not be justified, since gastrointestinal hemorrhage shares many of the same risk factors as cardiovascular disease. It also assumes that any benefit seen by the addition of aspirin will not be lost due to a rebound effect of aspirin cessation in the summer resulting in more cardiovascular events, although there is a reported 3-fold risk of cardiac ischemia in moderate- to high-risk patients after aspirin cessation [89].
 |
PANEL 2. Summary points about the potential effects of aspirin on influenza and influenza-associated cardiovascular disease |
Given the lack of trials, any estimate of the benefit that might result from seasonal treatment with aspirin is likely to be flawed. However, it is possible to crudely estimate the number of patients who might be eligible. Given a conservative estimate of a relative risk of 1.3 for the occurrence of a myocardial infarction during the winter [21,24] or for the occurrence of ischemic heart disease during an influenza epidemic [37], then patients with a current 10-year absolute cardiovascular disease risk of approximately 15 to 20% or an 8.7 to 10% 10-year risk of coronary heart disease would potentially exceed the threshold for treatment in the winter months, assuming independence of the increased seasonal risk of myocardial infarction from other cardiovascular disease risk factors. According to recent estimates of the prevalence of these levels of cardiovascular disease in the British population, approximately 10% of patients standardized by age fall into this risk category, without prior diabetes or cardiovascular disease [90]. An alternative view would suggest that patients in the 20 to 26% risk strata who are currently advised to take aspirin are at relatively lower risk during the summer months, because the risk calculations include the winter months during which the risk is elevated. As such, a significant proportion of patients may not need to take aspirin during the summer, and so have a lower risk of gastrointestinal hemorrhage.
Such crude estimations as these are not intended to provide an accurate measure of the potential impact of such a policy, and do not have the capacity to assess the health impact of treatment as we do not know the proportion of excess cardiovascular deaths or morbidity due to influenza which occur in each strata of risk. However, they indicate the scale upon which seasonal modification in treatment might be applicable, which affects 10 to 20% of the population.
IMPLICATIONS
There are no published data on the sustained relative efficacy of aspirin in the reduction of cardiovascular events during an influenza epidemic. Therefore, it cannot be reliably concluded that treatment should be altered at these times without increasing the overall risk to patients, even though it is most likely that the relative efficacy is either maintained or greater at these times of physiological stress. However, the question is readily accessible to investigation through either reviewing previous randomized trials, or through carrying out new randomized controlled trials.
The simplest strategy is the seasonal modification of aspirin use, wherein providing aspirin in the winter may reduce cardiovascular events to a greater extent than it increases hemorrhagic risk. In the summer, however, the dose can be withheld or reduced since the risk-benefit ratio is less favorable. Thus, aspirin can either be started in the winter in moderate-risk individuals or increased in high-risk patients already receiving aspirin, although there is a lack of evidence for a true dose-response relationship. This proposal could be managed by the incorporation of season in treatment-dependent cardiovascular risk algorithms, indicating the shift from moderate to high risk in the winter.
Given the efficacy of influenza vaccination, it could be argued that this is a preferred approach to reducing the cardiovascular burden of influenza epidemics. However, this strategy alone is unlikely to provide a complete solution:
- Vaccination is not 100% effective [41].
- Future epidemics may result in a significant degree of morbidity and mortality before a vaccine has been developed, as was seen with the recent H1N1 swine flu outbreak.
- There are covarying increases in other inflammatory winter diseases, which may increase cardiovascular risk independently of influenza, such as respiratory syncytial virus.
- The winter rise in cardiovascular mortality is also dependent upon noninfectious variables, which would also be addressed by seasonal prescription.
A second strategy would depend on the early identification of influenza epidemic years, a process that is already in place for the release of antiviral drugs in the National Health Service (UK). During these seasons, patients at moderate risk could again start aspirin treatment for the duration of the influenza outbreak. This would enhance the specificity for appropriate use of aspirin in this manner and increase the benefit-risk ratio, but would not reduce the excess cardiovascular morbidity and mortality in endemic years when these do not pass the incidence threshold for treatment.
Third, patients with moderate cardiovascular risk could be prescribed aspirin at the time of developing influenza following presentation to their physician, with cotreatment with antivirals when indicated for that season. Again, this would increase the specificity of treatment to reduce the risk of cardiovascular events without substantially increasing the hemorrhagic risk, but again would miss a greater number of preventable cardiovascular events.
Finally, given the limited evidence that there is an increased response to vaccination due to cotreatment with aspirin, this might be amenable to assessment by the review of prospective cohort studies, but could be more reliably addressed by a randomized controlled trial.
The first of these strategies could be readily assessed by a randomized controlled trial, which would move moderate-risk patients into the high-risk category by prescribing either aspirin or placebo during winter, and nothing in summer. Similarly, high-risk patients on low-dose aspirin in the summer could be randomized to high doses of aspirin during the winter versus no change in dose. A trial of combined aspirin and antivirals versus antivirals plus placebo during an outbreak in moderate-risk patients would also be feasible. Finally, moderate-risk patients could be randomized to placebo, aspirin, vaccination, or both in a factorial design, incorporating outcome measures for both influenza and cardiovascular outcomes, with a subgroup undergoing serological assessment.
CONCLUSIONS
Seasonal and epidemic influenza is associated with an increased risk of cardiovascular events resulting in significant morbidity and mortality (Panel 1). This association is likely to reflect the inflammatory environment and prothrombotic state during an influenza infection, but viral-specific mechanisms may exist. Evidence of a specific action of aspirin on influenza in man or its impact on cardiovascular disease is inadequate, but its relative efficacy is likely to be either unchanged or potentially greater during an influenza season (Panel 2). Therefore, as patients are at greater risk during the winter season, the absolute risk reduction with aspirin is likely to be increased, supporting its use during either the winter months or influenza epidemics in moderate-risk patients not already on aspirin or increased doses for high-risk patients. Ideally new randomized controlled trials looking at seasonal treatment are required to answer these questions and establish the balance of risk and benefit.
REFERENCES
1. Reid AH, Fanning TG, Hultin JV, Taubenberger JK. Origin and evolution of the 1918 “Spanish” influenza virus hemagglutinin gene. Proc Natl Acad Sci USA 1999;96:1651-1656. [Medline]
2. Reid AH, Fanning TG, Janczewski TA, Taubenberger JK. Characterization of the 1918 “Spanish” influenza virus neuraminidase gene. Proc Natl Acad Sci USA 2000;97:6785-6790. [Medline]
3. Crosby AW. America’s forgotten pandemic: the influenza of 1918. Cambridge, United Kingdom: Cambridge University Press; 1988.
4. Vane JR. The fight against rheumatism: from willow bark to COX-1 sparing drugs. J Physiol Pharmacol 2000;51:573-586. [Medline]
5. Starko KM. Salicylates and pandemic influenza mortality, 1918-1919, pharmacology, pathology and historic evidence. Clin Infect Dis 2009;49(9):1405-1410. [Medline]
6. Vane JR. Inhibition of prostaglandin synthesis as a mechanism of action for aspirin-like drugs. Nat New Biol 1971;231(25):232-235. [Medline]
7. Aspirin Myocardial Infarction Study (AMIS) Research Group. AMIS: the aspirin myocardial infarction study: final results. Circulation 1980;62(Suppl V):79-84. [Medline]
8. Coronary Drug Project (CDP) Research Group. Aspirin in coronary heart disease. Circulation 1980;62(Suppl V):59-62. [Medline]
9. Farrell B, Godwin J, Richards S, Warlow C. The United Kingdom transient ischaemic attack (UK-TIA) aspirin trial: final results. J Neurol Neurosurg Psychiatr 1991;54:1044-1054. [Medline]
10. Antithrombotic Trialists’ Collaboration. Collaborative metaanalysis of randomised trials of antiplatelet therapy for prevention of death, myocardial infarction, and stroke in high risk patients. BMJ 2002;324:71-86. [Medline]
11. Antithrombotic Trialists’ (ATT) Collaboration. Aspirin in the primary and secondary prevention of vascular disease: collaborative metaanalysis of individual participant data from randomised trials. Lancet 2009;373(9678):1849-1860. [Medline]
12. Harper SA, Bradley JS, Englund JA, et al. Seasonal influenza in adults and children—diagnosis, treatment, chemoprophylaxis, and institutional outbreak management: clinical practice guidelines of the Infectious Diseases Society of America. Clin Infect Dis 2009;48(8):1003-1032. [Medline]
13. Pandemic Flu: Clinical management of patients with an influenzalike illness during an influenza pandemic. Thorax 2007;62(Suppl i): i1-i13. [pdf]
14. Pearson TA, Blair SN, Daniels SR, et al. AHA guidelines for primary prevention of cardiovascular disease and stroke: 2002 update. Circulation 2002;106:388-391. [Medline]
15. Wood D, Wray R, Poulter N, et al. JBS2: Joint British Societies’ guidelines on prevention of cardiovascular disease in clinical practice. Heart 2005;92(Suppl V):v1-v52.
16. General Register Office. The Third Annual Report of Births, Marriages and Deaths. London: HMSO; 1841:pp.102–109.
17. Sheth T, Nair C, Muller J, Yusuf S. Increased winter mortality from acute myocardial infarction and stroke: The effect of age. J Am Coll Cardiol 1999;33(7):1916-1919. [Medline]
18. Healy JD. Excess winter mortality in Europe: a cross country analysis identifying key risk factors. J Epidemiol Community Health 2003;57:784-789. [Medline]
19. Eurowinter Group. Cold exposure and winter mortality from ischaemic heart disease, cerebrovascular disease, respiratory disease, and all causes in warm and cold regions of Europe. Lancet 1997;349:1341-1346. [Medline]
20. Laake K, Sverre JM. Winter excess mortality: A comparison between Norway and England plus Wales. Age Ageing 1996;25:343-348. [Medline]
21. Rumana N, Kita Y, Turin TC, et al. Seasonal pattern of incidence and case fatality of acute myocardial infarction in a Japanese population (from the Takashima AMI Registry, 1988 to 2003). Am J Cardiol 2008;102:1307-1311. [Medline]
22. Gemmell I, Mcloone P, Boddy FA, Dickinson GJ, Watt GC. Seasonal variation in mortality in Scotland. Int J Epid 2000;29:274-279. [Medline]
23. Mbanu I, Wellenius GA, Mittleman MA, Peeples L, Stallings LA, Kales SN. Seasonality and coronary heart disease deaths in United States firefighters. Chronobiol Int 2007;24(4):715-726. [Medline]
24. Spencer FA, Goldberg RJ, Becker RC, Gore JM. Seasonal distribution of acute myocardial infarction in the Second National Registry of Myocardial Infarction. J Am Coll Cardiol 1998;31(6):1226-1233. [Medline]
25. Clifford RE, Smith JW, Tillett HE, Wherry PJ. Excess mortality associated with influenza in England and Wales. Int J Epid 1977;6(2):115-128. [Medline]
26. Reichert TA, Simonsen L, Sharma A, Pardo SA, Fedson DS, Miller MA. Influenza and the winter increase in mortality in the United States, 1959-1999. Am J Epid 2004;160(5):492-502. [Medline]
27. Thompson WW, Shay DK, Weintraub E, et al. Mortality associated with influenza and respiratory syncytial virus in the United States. JAMA 2003;289(2):179-186. [Medline]
28. Madjid M, Naghavi M, Litovsky S, Casscells SW. Influenza and cardiovascular disease: A new opportunity for prevention and the need for further studies. Circulation 2003;108:2730-2736. [Medline]
29. Farr W. Tenth Annual Report of the Registrar General. London: HMSO, pxlii; 1847.
30. Collins SD. Excess mortality from causes other than influenza and pneumonia during influenza epidemics. Pub Health Rep 1932;47(46):2159-2179.
31. Miller DL, Pereira MS, Clarke M. Epidemiology of the Hong Kong/68 variant of influenza A2 in Britain. Br Med J 1971;1:475-479. [Medline]
32. Curwen M, Dunnell K, Ashley J. Hidden influenza deaths. BMJ 1990;300:896.
33. Thompson WW, Shay DK, Weintraub E, et al. Influenza-associated hospitalisations in the United States. JAMA 2004;292(11):1333-1340. [Medline]
34. Simonsen L, Fukuda K, Schonberger LB, Cox NJ. The impact of influenza epidemics on hospitalisations. J Inf Dis 2000;181:831-837. [Medline]
35. Barker WH. Excess pneumonia and influenza associated hospitalization during influenza epidemics in the United States, 1970-78. Am J Pub Health 1986;76(7):761-765. [Medline]
36. Fleming DM, Crombie DL, Norbury CA, Cross KW. Observations on the influenza epidemic of November/December 1989. Br J Gen Pract 1990;40:495-497. [Medline]
37. Madjid M, Miller CC, Zarubaev VV, et al. Influenza epidemics and acute respiratory disease activity are associated with a surge in autopsy-confirmed coronary heart disease death: results from 8 years of autopsies in 34,892 subjects. Eur Heart J 2007;28:1205-1210. [Medline]
38. Meier CR, Jick SS, Derby LE, Vasilakis C, Jick H. Acute respiratorytract infections and risk of first-time acute myocardial infarction. Lancet 1998;351:1467-1471. [Medline]
39. Spodick DH, Flessas AP, Johnson MM. Association of acute respiratory symptoms with onset of acute myocardial infarction: prospective investigation of 150 consecutive patients and matched control patients. Am J Cardiol 1984;53:481-482. [Medline]
40. Penttinen J, Valonen P. The risk of myocardial infarction among Finnish Farmers seeking medical care for an infection. Am J Pub Health 1996;86(10):1440-1442. [Medline]
41. Gross PA, Hermogenes AW, Sacks HS, Lau J, Levandowski RA. The efficacy of influenza vaccine in elderly persons: A meta-analysis and review of the literature. Ann Intern Med 1995;123:518-527. [Medline]
42. Gohlke H, Albus C, Gysan DB, Hahmann HW, Mathes P. Cardiovascular prevention in clinical practice (ESC and German Guidelines 2007). Herz 2009;34:4-14. [Medline]
43. Davis MM, Taubert K, Benin AL, et al. Influenza vaccination as secondary prevention for cardiovascular disease: A science advisory from the American Heart Association/American College of Cardiology. Circulation 2006;114:1549-1553. [Medline]
44. Nichol KL, Nordin J, Mullooly J, Lask R, Fillbrandt K, Iwane M. Influenza vaccination and reduction in hospitalisations for cardiac disease and stroke among the elderly. N Engl J Med 2003;248(14):1322-1332. [Medline]
45. Naghavi M, Barlas Z, Siadaty S, Naguib S, Madjid M, Casscells W. Association of influenza vaccination and reduced risk of recurrent myocardial infarction. Circulation 2000;102:3039-3045. [Medline]
46. Siscovick DS, Raghunathan E, Lin D, et al. Influenza vaccination and the risk of primary cardiac arrest. Am J Epidemiol 2000;152:674-677. [Medline]
47. Lavallee P, Perchaud V, Gautier-Berrtrand M, Grabli D, Amarenco P. Association between influenza vaccination and risk of brain infarction. Stroke 2002;33:513-518. [Medline]
48. Jackson LA, Yu O, Heckbert SR, et al. Influenza is not associated with a reduction in the risk of coronary events. Am J Epidemiol 2002;156:634-640. [Medline]
49. Madjid M, Curkendall S, Blumentals WA. The influence of oseltamivir treatment of the risk of stroke after influenza infection. Cardiology 2009;113:98-107. [Medline]
50. Gurfinkel EP, Leon de la Fuente R, Mendiz O, Mautner B. Flu vaccination in acute coronary syndromes and planned percutaneous coronary interventions (FLUVACS) Study. Eur Heart J 2004;25(1):25-31. [Medline]
51. Stary HC, Chandler AB, Glagov S, et al. A definition of initial, fatty streak, and intermediate lesions of atherosclerosis: a report from the Committee on Vascular Lesions of the Council on Arteriosclerosis, American Heart Association. Circulation 1994;89:2462-2478. [Medline]
52. Libby P, Ridker PM, Hansson GK. Inflammation in atherosclerosis: from pathophysiology to practice. J Am Coll Cardiol 2009;54(23):2129-2138. [Medline]
53. Griendling KK, Alexander RW. Oxidative stress and cardiovascular disease. Circulation 1997;96:3264-3265. [Medline]
54. Hendrix MG, Salimans MM, van Bowen CP, Bruggeman CA. High prevalence of latently presence cytomegalovirus in arterial walls of patients suffering from grade III atherosclerosis. Am J Pathol 1990;136:23-28. [Medline]
55. Jackson LA, Campbell LA, Schmidt RA, et al. Specificity of detection of chlamydia pneumonia in cardiovascular atheroma; evaluation of the innocent bystander hypothesis. Am J Pathol 1997;150:1785-1790. [Medline]
56. Muhlstein JB, Anderson JL, Carlquist JF, et al. Randomized secondary prevention trial of azithromycin in patients with coronary artery disease: primary clinical results of the ACADEMIC study. Circulation 2000;102:1755-1760. [Medline]
57. O’Connor CM, Dunne MW, Pfeffer MA, et al. Azithromycin for the secondary prevention of coronary heart disease events: the WIZARD study. A randomized controlled trial. JAMA 2003;290:1459-1466. [Medline]
58. Ridker PM, Cushman M, Stampfer MJ, Tracy RP, Hennekens CH. Inflammation, aspirin and the risk of cardiovascular disease in apparently healthy men. N Engl J Med 1997;336(14):973-979. [Medline]
59. Ridker PM, Danielson E, Fonseca FA, et al. Reduction in Creactive protein and LDL cholesterol and cardiovascular event rates after initiation of rosuvastatin: a prospective study of the JUPITER trial. Lancet 2009;373:1175-1182. [Medline]
60. Haidari M, Wyde PR, Litovsky S, et al. Influenza virus directly infects, inflames, and resides in the arteries of atherosclerotic and normal mice. Atherosclerosis 2010;208(1):90-96. [Medline]
61. Gurevich VS, Pleskov VM, Levaya MV, et al. Influenza virus infection in progressing atherosclerosis. Kardiologia 2002;42:21-24. [Medline]
62. Chowell G, Bertozzi SM, Colchero MA, et al. Severe respiratory disease concurrent with the circulation of H1N1 influenza. N Engl J Med 2009;361(7):674-679. [Medline]
63. Cheung CY, Poon LL, Lau AS, et al. Induction of proinflammatory cytokines in human macrophages by influenza A (H5N1) viruses: a mechanism for the unusual severity of human disease? Lancet 2002;360:1831-1837. [Medline]
64. Van Lenten BJ, Wagner AC, Anantharamaiah GM, et al. Influenza infection promotes macrophage traffic into arteries of mice that is prevented b D-4F, an apolipoprotein A-I mimetic peptide. Circulation 2002;106:1127-1132. [Medline]
65. Naghavi M, Wyde P, Litovsky S, et al. Influenza infection exerts prominent inflammatory and thrombotic effects on the atherosclerotic plaques of apolipoprotein E-deficient mice. Circulation 2003;107:762-768. [Medline]
66. Keller TT, van der Meer JJ, Teeling P, et al. Selective expansion of influenza A virus-specific T cells in symptomatic human carotid artery atherosclerotic plaques. Stroke 2008;39:174-179. [Medline]
67. Bouwman JJ, Visseren FL, Bosch MC, Bouter KP, Diepersloot RJ. Procoagulant and inflammatory response of virus-infected monocytes. Eur J Clin Invest 2002;32:759-766. [Medline]
68. Azambuja MI. Spanish flu and early 20th-century expansion of a coronary heart disease-probe subpopulation. Tex Heart Inst J 2004;31:14-21. [Medline]
69. Guan XR, Li X, Xin XM, et al. Influenza virus infection and risk of acute myocardial infarction. Inflammation 2008;31(4):266-272. [Medline]
70. Terada H, Baldini M, Ebbe S, Madoff MA. Interaction of influenza virus with blood platelets. Blood 1966;28:213-228. [Medline]
71. Davison AM, Thomson D, Robson JS. Intravascular coagulation complicating influenza A virus infection. Br Med J 1973;1:654-655. [Medline]
72. Visseren FL, Bouwman JJ, Bouter KP, Dipersloot RJ, de Groot PH, Erkelens DW. Procoagulant activity of endothelial cells after infection with respiratory viruses. Thromb Haemost 2000;84:319-324. [Medline]
73. Wurzer WJ, Ehrhardt C, Pleschka S, et al. NF-kappaB-dependent induction of tumor necrosis factor-related apoptosis-inducing ligand (TRAIL) and Fas/FasL is crucial for efficient influenza virus propagation. J Biol Chem 2004;279:30931-30937. [Medline]
74. Nimmerjahn F, Dudziak D, Dirmeier U, et al. Active NF-kappaB signalling is a prerequisite for influenza virus infection. J Gen Virol 2004;85:2347-2356. [Medline]
75. Mazur I, Wurzer WJ, Ehrhardt C, et al. Acetylsalicylic acid (ASA) blocks influenza virus propagation via its NF-kB inhibiting activity. Cell Micro 2007;9(7):1683-1694. [Medline]
76. Lee SM, Cheung CY, Nicholls JM, et al. Hyperinduction of cyclooxygenase-2-mediated proinflammatory cascade: a mechanism for the pathogenesis of avian influenza H5N1 infection. J Infect Dis 2008;198(4):525-535. [Medline]
77. Bernasconi D, Amici C, La Frazia S, Ianaro A, Santoro MG. The IkB Kinase is a key factor in triggering influenza A virus-induced inflammatory cytokine production in airway epithelial cells. J Biol Chem 2005;280(25):24127-24134. [Medline]
78. Zheng BJ, Chan KW, Lin YP, et al. Delayed antiviral plus immunomodulator treatment still reduces mortality in mice infected by high inoculums of influenza A/H5N1 virus. PNAS 2008;105(23):8091-8096. [Medline]
79. Kharbanda RK, Walton BW, Allen M, et al. Prevention of inflammation-induced endothelial dysfunction: A novel vasculoprotective action of aspirin. Circulation 2002;105:2600-2604. [Medline]
80. Bancos S, Bernard MP, Topham DJ, Phipps RP. Ibuprofen and other widely used non-steroida anti-inflammatory drugs inhibit antibody production in human cells. Cell Immunol 2009;258(1):18-28. [Medline]
81. Buckland M, Lombardi G. Aspirin and the induction of tolerance by dendritic cells. Handb Exp Pharmacol 2009;188:197-213. [Medline]
82. Hsia J, Tang T, Parrott M, Rogalla K. Augmentation of the immune response to influenza vaccine by acetylsalicylic acid: a clinical trial in a geriatric population. Methods Find Exp Clin Pharmacol 1994;16(9):677-683. [Medline]
83. Eccles R, Loose I, Jawad M, Nyman L. Effects of acetylsalicylic acid on sore throat pain and other pain symptoms associated with acute upper respiratory tract infections. Pain Medicine 2003;4:118-124. [Medline]
84. Schachtel BP, Fillingim JM, Beiter DJ, Lane AC, Schwartz LA. Rating scales for analgesics in sore throat. Clin Pharm Ther 1984;36:151-156. [Medline]
85. Bachert C, Chuchalin AG, Eisebett R, Netayzhenko VZ, Voelker M. Aspirin compared with acetaminophen in the treatment of fever and other symptoms of upper respiratory tract infection in adults: a multicenter, randomized double-blind, double-dummy, placebo-controlled, parallel-group, single-dose, 6-h dose-ranging study. Clin Ther 2005;27:993-1003. [Medline]
86. Edwards JE, Oldman A, Smith L, et al. Single dose aspirin for acute pain. Cochrane Database Syst Rev 2000;(2):CD002067. [Medline]
87. Eccles R. Efficacy and safety of over-the-counter analgesics in the treatment of common cold and flu. J Clin Pharm Ther 2006;31:309-319. [Medline]
88. De Backer G, Ambrosioni E, Borch-Johnsen K, et al. European guidelines on cardiovascular disease prevention in clinical practice. Third Joint Task Force of European and other Societies on Cardiovascular Disease Prevention in Clinical Practice. Eur Heart J 2003;24:1601-1610. [Medline]
89. Biondi-Zoccai GG, Lotrionte M, Agostoni P, et al. A systematic review and meta-analysis on the hazards of discontinuing or not adhering to aspirin among 50,279 patients at risk for coronary artery disease. Eur Heart J 2006;24:2667-2674. [Medline]
90. Neil HA, Perera R, Armitage JM, Farmer AJ, Mant D, Durrington PN. Estimated 10-year cardiovascular risk in a British population: results of a national screening project. Int J Clin Pract 2008;62(9):1322-1331. [Medline]