Influenza and myocardial infarction

Inflammation & cardiovascular disease: chronic

The complex relationship between infections and cardiovascular events has been examined in the scientific literature for many years. Infections may be implicated at various stages of the pathogenesis of atherosclerosis, from the early development of lesions to destabilization of advanced plaques and myocardial infarction. Chronic atherosclerosis is enhanced by certain infections in low-density lipoprotein (LDL) receptor- or apoE-deficient mice [1], and is reduced by the selective knockout of inflammatory or immune pathway components [2]. In humans, an initial wave of studies reporting strong associations between serological markers of chronic infection (e.g., herpes virus and Chlamydia Pneumonia) or inflammation and cardiovascular disease [3] have been followed by larger-scale work suggesting less strong, but still positive, associations [4–5]. Increased long-term risk of vascular disease among people with noninfective inflammatory disease, such as psoriasis [6] and rheumatoid arthritis [7], is increasingly recognized.

Acute inflammation & cardiovascular disease

In this article, we focus on downstream events, that is, the potential for acute respiratory infections and specifically influenza to trigger acute cardiovascular events such as acute myocardial infarction (AMI). Acute inflammation is associated with profound suppression of endothelium-dependent relaxation [8] – an effect mitigated by aspirin [9]. Although an increased risk of AMI and stroke has been demonstrated in the days following acute infections, the effect is similar for respiratory and urinary tract infections [10]. It is therefore unclear whether generic triggers involving systemic inflammatory processes are to blame or if in fact there might be heightened effects associated with certain infections such as influenza (for which other cardiac complications, such as myocarditis, are well documented). AMI results from a complex interaction between the host’s genetic and lifestyle factors, environmental factors such as the effects of climate and temperature, and other factors such as infectious triggers. Any acute infectious trigger is likely to be working on a background of existing atherosclerosis.

Influenza & AMI

In our recent systematic review, we chose to focus on influenza as it is one of the most common severe respiratory viral infections (with >20% of the population seroconverting in a typical influenza season [11]) and the only one for which effective prophylaxis and treatment is available [12]. Investigating links between influenza and AMI is also highly topical given the recent influenza A H1N1 ‘swine flu pandemic’. A range of observational studies performed in different settings have generally tended to support the hypothesis that acute respiratory infections – and influenza in particular – can trigger AMI [12]. There is also more limited evidence for an adverse effect on cardiovascular death. Almost all of these studies were performed in the context of seasonal (rather than pandemic) influenza. When considering the relationship between influenza infection and AMI, more work is needed to understand the potential impact in a population or the proportion of AMIs that might be affected. Studies describing excess influenza mortality (i.e., deaths occurring during the influenza season above what would be expected at other times of the year) suggest that a relatively high proportion are due to cardiovascular disease (CVD) – approximately 35–50% depending on the definition of CVD death used. Several other studies have used primary care data to examine the relationship between influenza-like illness and cardiovascular outcomes [10,13,14]; these are likely to have substantially underestimated levels of influenza-like illness in the community (as most people with an influenza-like illness do not present to primary care) and thus may underestimate the impact of influenza on cardiovascular events.

Possible impact of pandemic influenza on AMI

Circulating strains of influenza vary year by year. While some symptoms such as fever, cough and sore throat are usually consistent across strains, certain other symptoms might vary between strains. In a Taiwanese study comparing the symptom profiles of children infected with laboratory-confirmed influenza A with those infected with influenza B, the latter were more likely to be diagnosed as gastroenteritis, myositis or upper respiratory infection [15]. It could be hypothesized that certain strains may be more likely to trigger AMI than other strains, although this remains to be investigated. The current pandemic strain, influenza A H1N1, continues to cause mostly mild upper respiratory illness in those affected. It predominantly affects children and young adults, making it difficult to evaluate any effect on AMI risk. In addition, the prompt vaccination campaign in many developed countries targeted at people with chronic diseases, such as CVD, may limit any impact on AMI risk. However, in Southern Hemisphere countries such as Australia that experienced an H1N1 pandemic prior to vaccine being available, while admission rates to intensive care units due to influenza were higher than usual [16], excess mortality has been lower than that seen with seasonal influenza [17]. Worldwide, of those who died in the early phases of the pandemic, around half had reported underlying medical conditions including cardiac conditions (although mechanisms for ascertaining this varied between sources) [18]. However, it is unclear whether there has been any change in hospitalization and death rates from AMI and CVD compared with previous years.

Possible mechanisms

Understanding the mechanism is likely to be fundamental to pin-pointing areas for potential intervention. It is already known from both in vitro and in vivo models that influenza has multiple effects on inflammatory and coagulation pathways. Inflammatory cytokines, such as TNF-α and IL-6, are increased in the context of influenza infection in murine studies [19]. In influenza-infected mice, thrombin generation, fibrin deposition and fibrinolysis are increased compared with noninfected counterparts, suggesting that influenza leads to a prothrombotic state by coagulation activation [20]. In chickens infected with highly pathogenic H5N1, hemostatic abnormalities represented by excess blood clotting and consumptive coagulopathy occurred [21]. Inflammation is also implicated at all stages of the atherosclerotic process from the early binding of leukocytes (to damaged endothelium) to the development of advanced lesions and thrombotic complications [22]. A recent study of inflammatory markers in the context of AMI showed that there were significantly higher levels of cytokines such as IL-6 in patients with ST elevation MI (STEMI) compared with non-ST elevation MI (NSTEMI) patients [23]. Markers of systemic inflammation, such as high-sensitivity C-reactive protein (hsCRP), have been used to predict the risk of vascular disease; in the Scottish Health Survey, very highly increased hsCRP (>10 mg/l) was associated with CVD events (including AMI, coronary artery bypass graft, primary percutaneous coronary intervention, stroke and cardiac failure) [24]. A recent systematic review found that adding data on CRP did not improve risk estimates from the Framingham risk score [25]. The US Preventive Services Task Force concluded that there was insufficient evidence that changes in CRP level lead to the primary prevention of coronary heart disease events [26]. Despite this, it is possible that systemic inflammation triggered by an acute episode of influenza infection might interact with advanced complicated plaques leading to acute myocardial infarction. It is not yet clear whether there is a differential effect on type of AMI (e.g., STEMI vs NSTEMI) or severity.

Possible interventions

Vaccine

Public health strategies to prevent influenza could have the added benefit of reducing risks of AMI. Two relatively small randomized trials suggest that influenza vaccination reduces the risk of cardiovascular death and some coronary ischemic events in populations with existing cardiovascular disease [27,28]. Although annual vaccination against seasonal influenza is recommended in many countries for those with chronic medical conditions, including cardiovascular disease and diabetes, there is suboptimal uptake in this group: vaccine coverage was 47.2% in the UK in the winter of 2008–2009 [101] and 38.8% in the USA in 2007–2008 for adults aged 18–64 years with ‘high-risk’ conditions [102]. This represents a potentially major missed opportunity for the prevention of cardiovascular events: the relative risk reduction for CVD death provided by influenza vaccine was approximately 50% in a pooled estimate from these two small trials (although this reduction in risk was not statistically significant when calculated using a random effects model) [12]. By comparison, a prospective meta-analysis of data from 14 randomized trials of statins demonstrated a 19% reduction in coronary mortality per mmol/l reduction in LDL cholesterol [29].

In the recent influenza pandemic, studies have shown that people aged over 60 years (who are most at risk of cardiovascular disease) tend to have some pre-existing immunity to H1N1 [30]. Although the incidence of pandemic influenza may be lower among older people, the risk of complications and deaths is higher. To date, it is unclear whether the 2009 H1N1 pandemic strain increases the risk of AMI. However, encouraging influenza vaccination with pandemic vaccine in those with underlying conditions might be expected to reduce the risk of death from influenza.

Antivirals

In the context of seasonal influenza, antivirals (e.g., oseltamivir, zanamivir and amantadine) are recommended in the UK for people at higher risk of influenza complications (e.g., those with cardiovascular disease) presenting to primary care within 48 h of onset of an influenza-like illness [103]. Antiviral drugs lessen the severity and shorten the duration of influenza illness and thus might be expected to lessen the effects of infection on the risk of AMI. Although limited observational evidence suggests that antiviral drugs might be protective against adverse cardiovascular outcomes in people with seasonal influenza infection [31,32], there is a lack of clear evidence of benefit from prospective trials. For the current influenza pandemic, the WHO recommends a targeted antiviral approach using oseltamivir or zanamivir for those at higher risk of complications [104]. Again, if there is a link between the pandemic strain and MI, then the timely prescription of antivirals may be expected to limit risk (although this remains to be tested).

Other agents

Several agents commonly used for the primary and secondary prevention of acute MI and other acute vascular events also modulate inflammation in addition to their main effects [33]; for example, statins have been shown to reduce hsCRP levels independent of their effect on serum cholesterol [34]. Controversy remains over the likely effectiveness of statins at reducing infection-related mortality. Several cohort studies have suggested that statins may be beneficial for the prevention and treatment of various infections, including one study in which the risk of influenza-related mortality was reduced in patients on moderate doses of statins [35]. It is postulated that the pleiotropic effects of statins, specifically the anti-inflammatory and immunomodulatory effects mediated through decreasing intermediate products of cholesterol synthesis that play a role in intracellular signaling pathways, may be responsible. However, a major criticism of such observational data is the possibility of a ‘healthy user’ effect. Authors of a recent systematic review on statins and infection identify several clinical trials currently underway that may help to illuminate this question [36]. It remains to be seen whether other anti-inflammatory treatment is likely to impact upon AMI triggered by influenza and other respiratory infections.

Conclusion

A consistent association between influenza infection and the risk of myocardial infarction has been demonstrated in a range of settings. This increased risk is likely to contribute substantially to the severe adverse health outcomes attributable to influenza. Vaccinations reduce the risk of both seasonal and swine influenza. Among people with existing vascular disease, there is some evidence that seasonal influenza vaccination reduces the risk of acute cardiac events. However, to date, there is a lack of direct evidence to support such a beneficial effect among other groups such as those with undiagnosed atherosclerotic disease or CVD risk factors. Antiviral drugs reduce the severity and duration of influenza, while they may therefore be expected to reduce any associated risk of AMI, we again lack direct evidence of such a possible benefit. Whether other interventions, such as statins, aspirin or oral anticoagulants have a protective effect against the transient increased risk of AMI during episodes of influenza is not known. Improved knowledge and understanding of the association between influenza infection and AMI is needed to guide the most appropriate intervention strategies in different population groups. For now, the key priority is encouraging vaccination among those most at risk of AMI.

Financial & competing interests disclosure

Charlotte Warren-Gash holds a MRC Clinical Research Training Fellowship. Liam Smeeth is supported by a Wellcome Trust Senior Research Fellowship in Clinical Science. Andrew Hayward is supported by the Camden Primary Care Trust. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.

No writing assistance was utilized in the production of this manuscript.