RR: Unadjusted risk ratio
aRR: Adjusted risk ratio; Adjusted for age, sex, tobacco use, exercise level, hypertension, diabetes, high blood cholesterol, heart failure, previous history of CHD and BMI
§Includes both annual influenza seasons of 2017 and 2018
DISCUSSION:
To the best of our knowledge, this is the first study in Bangladesh to investigate the prevalence of recent CRI and laboratory confirmed influenza among patients with AMI. The results of this study indicate that a significant proportion of the participants had a history of CRI within a week of AMI onset and that qRT-PCR-confirmed influenza was present, suggesting that recent acute respiratory illness and influenza may potentially contribute to triggering AMI. Notably, the self-reported CRI among AMI patients was prevalent during both influenza and non-influenza seasons, indicating that the circulation of non-influenza respiratory viral pathogens could also potentially play a role in triggering the onset of AMI (34-36). These findings highlight the need for further investigation into the potential association between recent respiratory illnesses and AMI, including the identification of specific viral pathogens and mechanisms underlying this association. About one third of the enrolled AMI patients during the 2017 influenza season reported history of CRI which was much more frequent than that reported during 2018 influenza season. This could be validated in the context of report by the national hospital-based influenza surveillance program in Bangladesh indicating a relatively more severe influenza epidemic and upward surge in severe acute respiratory illness (SARI) admissions occurred in the country during the 2017 compared to the 2018 influenza season (37) suggesting that there could be a potential link between the intensity of the influenza season and the number of AMI cases associated with influenza. The current study did not find a statistically significant association between recent CRI and the severity of myocardial damage, as defined by STEMI or high blood troponin. However, in general, the proportion of STEMI cases over NSTEMI was higher during the influenza season compared to the non-influenza season, and this difference was statistically significant in the unadjusted but not in adjusted regression analysis. It is possible that unrecognized recent influenza infection may play a role in triggering STEMI, however, further research is needed to establish a definitive relationship. Although our findings imply a potential link between recent acute respiratory illnesses and influenza with the occurrence of AMI, it is crucial to acknowledge that our study’s cross-sectional design does not allow us to establish a definitive association between the exposures and the outcome and hence recommend further validation. Further studies, including case-control studies or prospective cohort studies, are needed to confirm an association and elucidate the mechanisms underlying this association. However, these results do highlight the importance of considering the potential impact of respiratory illness, including seasonal influenza, on cardiovascular health among population in Bangladesh. Importance of further research and analysis of the potential benefits of robust infection control measures and influenza vaccination programs for AMI prevention cannot be overstated.
Previous studies have reported a wide range of frequencies of recent respiratory illness among AMI cases, ranging from 2.8% to 60.3% (16). The overall frequency of recent CRI in this study (11.3%) is comparable to previous reports (16), from London (14.3%) and Finland (12.2%), and the frequency of CRI during the 2017 influenza season in this study (36%) is comparable to reports (16), from Sydney, Australia (31.1%), Massachusetts (28%), London (24.3%), Finland (28%) and lastly from Karachi, Pakistan (37) (36.2%). However, it is possible that the differences in reported frequency of acute respiratory illness among AMI cases across studies may be attributed to various factors, such as demographic variations among participants and discrepancies in the criteria used to define recent acute respiratory illness. Most of the previous studies were conducted in high-income countries where the average age of recruited AMI patients was over 60 years old, which was significantly older than the average age of participants in our study (52 years). While Warren et al. (38) defined recent acute respiratory illness for AMI cases as having both respiratory and systemic symptoms with an onset of illness within the past month, our criteria for CRI did not include systemic symptoms. This difference in criteria may be an important limiting factor in appropriately classifying cases with recent acute respiratory illness, as including systemic symptoms in the criteria for recent acute respiratory illness may be crucial in accurately identifying cases that have truly had a recent acute respiratory infection. Our study’s definition for CRI (32, 33) only included respiratory symptoms that developed within a week of AMI in order to minimize recall bias. We believe that major cardiovascular events following an acute respiratory infection are more clinically plausible during this timeframe than a longer period of time after the infection. Despite the lack of systemic symptoms in our case definition, we still believe that our definition for recent respiratory illness in the current study is clinically sensitive enough to increase the likelihood of capturing cases with recent respiratory infection preceding onset of AMI. However, we acknowledge that our definition may have lower specificity, potentially leading to the inclusion of false positive cases for CRI. Another potential contributing factor to the differing frequencies across previous studies is that some were restricted to only influenza seasons while others were conducted during both influenza (30, 38) and non-influenza seasons (39).
Due to cross-sectional study design, we were unable to show a direct association between recent CRI and AMI. However, several previous case-control studies showed an association of recent respiratory illness with AMI. A case-control study based on large general practice database in Europe showed risk of AMI incidence twofold within 7 days after respiratory infection (39, 40). Another longitudinal population-based cohort study performed in United States indicated risk of acute cardiovascular events including AMI, stroke and death highest during the first month of hospitalization for pneumonia (39, 40). In general AMI cases could be as much as twice more likely than controls to report history of recent respiratory illness occurring within 7 days of AMI onset and the strength of this association is lesser for respiratory illness occurring >7 days of onset of AMI and fell over time (30, 38, 39). There is high double burden of both acute respiratory infections as well as acute cardiovascular events (26) in Bangladesh and despite evidence in other countries there is no data for Bangladesh about prevalence of recent respiratory illness preceding onset of AMI.
We reported a low frequency of influenza positivity in AMI cases. In the previous studies, the frequency of influenza detection by real-time PCR, paired serum influenza antibodies and single baseline influenza antibody titer among AMI patients ranged from 14% to 86.3%(16). The detection rates may considerably vary due to study specific laboratory methods applied, pattern as well as intensity of influenza strains during study period, and study conducted during both or either influenza and/or non-influenza seasons. The most confirmatory standard test method to diagnose influenza is RT-PCR test of respiratory swabs as recommended by WHO. Most of the previous studies identified low numbers of influenza by PCR alone (30, 38, 41). Nevertheless, investigators considered using baseline serology (41) and analysis of paired serums (30) for IgG or baseline serology for IgA (38) to report additional influenza with or without PCR. The WHO recommends swabbing patients within 10 days of onset of respiratory symptoms to increase the likelihood of detecting influenza RNA by PCR before diminution of viral shedding (42). There is still limited clarity on the exact timeline of onset of AMI after influenza infection, hence PCR test will likely have lower sensitivity if viral shedding diminishes before swabbing AMI patients. However, to maximize sensitivity to detect viral shedding, all participants in our study were swabbed within 72 hours of the onset of AMI. Serological analysis of convalescent serum in addition to PCR could have enhanced the sensitivity to detect more influenza positive AMI patients which is limitation of the current study. Moreover, due to administrative delays, we enrolled and tested only a minimal number of AMI cases during the peak influenza months May-September in 2017 influenza season when sequentially A(H1N1)pdm09, A/H3, and influenza B were the predominant strains circulating nationally (37).
During the current study, all cases of rRT-PCR confirmed influenza among AMI patients were identified only within the influenza seasons. The influenza subtypes that were identified fully corresponded to the month-specific circulating influenza strains identified through the national influenza surveillance scheme in the country (37) signifying typical influenza strains were also circulating among AMI patients in Bangladesh during the 2017 and 2018 influenza seasons. Interestingly, our study found a higher frequency of real-time PCR confirmed influenza in AMI patients compared to similar studies conducted in high income countries, where only 0/70 and 1/275 AMI cases tested positive for influenza nucleic acid (30, 38). It is possible that the population in Bangladesh has a higher susceptibility to influenza-associated AMI due to factors such as low vaccination rates, or high prevalence of cardiovascular co-morbidities. Nevertheless, our study design was cross-sectional and did not have a control group, therefore not specifically designed to investigate the association between laboratory confirmed influenza and AMI. However, very few previous case-control studies were able to reveal a direct association between laboratory confirmed influenza and AMI (41), perhaps due to the fact that influenza may be less common in the particular age group where AMI usually occurs. Conversely, more case-control studies have reported significant effectiveness of influenza vaccine against AMI (16, 30) which could be an indirect evidence of influenza’s association with AMI.
Analyses in the current study to assess association of recent CRI with severity of myocardial damage, when restricted to only influenza seasons, showed frequencies of recent CRI generally higher among STEMI than NSTEMI which was statistically not significant. Besides, it is noteworthy that although not statistically significant, we observed a strong positive trend in the association between CRI and high-troponin among AMI patients, and in some cases likelihood of development of AMI with high-troponin level was up to 80% higher among patients with CRI compared to those without, for example during 2018 influenza season. Furthermore, all of the identified influenza strains were exclusive to STEMI cases, emphasizing the need to consider these findings in patient evaluations. Both STEMI and high troponin level are related to severe myocardial damage. We assume that the analyses restricted to only influenza seasons were underpowered to find statistically significant results due to the small sample size within the subgroups related to myocardial damage. Nevertheless, there is previous evidence that influenza infection may increase the risk while the influenza vaccine is effective against large size infarcts, high troponin or CK-MB levels in patients with AMI (43). An upward trend of association between CRI and STEMI or high-troponin observed in the current study should be interpreted with caution as this may indicate a possible link between seasonality of acute respiratory infections including influenza and AMI events of greater myocardial damage among high risk unvaccinated individuals, which could be further explored through robust analytical studies conducted across multiple seasons. We believe, the magnitude and direction of such associations would depend considerably on the intensity and pattern of circulating seasonal influenza strains embedded within climatic factors (35, 44) and perhaps additionally and importantly, on clinically unrecognized respiratory viral infections (11, 13, 34). For example, five out of seven influenza positive cases in the current study did not report recent CRI which may imply that link between influenza and AMI may be more complex than our current understanding. This may also suggest that there may be other mechanisms by which influenza increases the risk of AMI, even in the absence of an acute respiratory illness. For example, influenza may cause changes in the immune system, blood clotting, or cardiac function increasing the risk of AMI. Alternatively, it may be that the individuals in the study who tested positive for influenza but did not report recent acute respiratory illness had underlying conditions predisposing them to AMI, and the influenza infection simply acted as a trigger (45). Accordingly, further analysis in the current study showed higher prevalence of STEMI among participants during influenza season than during non-influenza season. The univariate analysis showed there was a significant 9% increase in the risk of STEMI during influenza than during non-influenza season suggesting that acute respiratory illnesses may increase the risk of STEMI during influenza season.
The underlying pathophysiology of STEMI is complete blockage of the coronary artery by atherothrombosis causing transmural cardiac myonecrosis. This is primarily driven by an acute end stage of a chronic inflammatory atherosclerotic lesion characterized by abrupt rupturing of the de-stabilized atherosclerotic plaque due to short term exposure of certain triggering factors that may differ from the number of known cardiovascular risk factors (46). Such triggers of plaque rupture can include respiratory viral infections including influenza along with smoking, excessive alcohol, hypertension, heavy physical exertion or any kind of stressful events (47, 48). One study showed respiratory viral infections can precipitate both STEMI and NSTEMI and was positively associated with risk of mortality among NSTEMI, but not among STEMI (49). Nevertheless, patients after STEMI have a higher in-hospital mortality rate and worse short-term outcome while NSTEMI patients have poorer long-term prognosis (50). Blood troponin level are well correlated to the extent of infarction in both STEMI and NSTEMI but more impressive in STEMI (51, 52). In the current study, age at onset of STEMI was significantly lower than NSTEMI, suggesting that these different subtypes of AMI may have different risk factors and underlying mechanisms. Further research is needed to explore the relationship between influenza and different subtypes of AMI, as well as any factors influencing the association between influenza and STEMI. Preventing early onset STEMI is crucial in Bangladesh, where the age of onset for AMI is much earlier than in high-income countries. A simple yet effective measure to combat this issue could be the administration of the influenza vaccine. This could not only help curb the early onset of STEMI, but also significantly lower in-hospital mortality among young individuals in Bangladesh.
The immune system plays a critical role in both the pathophysiology of AMI and the physiological mechanisms behind the protection offered by the influenza vaccine against AMI. This interplay between the immune system and AMI highlights the vital importance of understanding the intricacies of this relationship in order to effectively prevent and treat AMI. Dominant pro-inflammatory over the anti-inflammatory component of the immune system may favor sudden atherosclerosis progression leading to acute cardiovascular events like AMI (53). Influenza virus can induce significant acute changes in pro-inflammatory cytokine levels in blood and pro-inflammatory as well as prothrombotic effects in atherosclerotic plaques which can trigger AMI onset through plaque destabilization(54). However, there are inter-individual differences in the intensity of an rapid pro-inflammatory response which may explain the difference in the level of risk of AMI among individuals in response to an acute stimulus such as influenza (55). Investigations continue globally to understand the relationship between influenza and AMI, with the goal of using influenza vaccination to prevent AMI in high-risk individuals. Several observational studies (16, 30, 56), small scale (57, 58) and large scale (22) randomized clinical trials reported protective efficacy of influenza vaccine against adverse cardiovascular events including hospitalization or death due to AMI. Animal study showed influenza vaccine stabilized atherosclerotic plaque through promoting anti-inflammatory atheroprotective immune response (59) implying possibly a greater protection against underlying pathophysiology of onset of STEMI than NSTEMI. Influenza vaccine has been shown to blunt pro-inflammatory and enhance anti-inflammatory mediators after coronary artery bypass surgery (60).