Increased Risk of Acute Myocardial Infarction and Stroke During Hemorrhagic Fever With Renal SyndromeCLINICAL PERSPECTIVE
A Self-Controlled Case Series Study
Background—We recently observed that cardiovascular causes of death are common in patients with hemorrhagic fever with renal syndrome (HFRS), which is caused by hantaviruses. However, it is not known whether HFRS is a risk factor for the acute cardiovascular events of acute myocardial infarction (AMI) and stroke.
Methods and Results—Personal identification numbers from the Swedish HFRS patient database (1997–2012; n=6643) were cross-linked with the National Patient Register from 1987 to 2011. Using the self-controlled case series method, we calculated the incidence rate ratio of AMI/stroke in the 21 days after HFRS against 2 different control periods either excluding (analysis 1) or including (analysis 2) fatal AMI/stroke events. The incidence rate ratios for analyses 1 and 2 for all AMI events were 5.53 (95% confidence interval [CI], 2.6–11.8) and 6.02 (95% CI, 2.95–12.3) and for first AMI events were 3.53 (95% CI, 1.25–9.96) and 4.64 (95% CI, 1.83–11.77). The incidence rate ratios for analyses 1 and 2 for all stroke events were 12.93 (95% CI, 5.62–29.74) and 15.16 (95% CI, 7.21–31.87) and for first stroke events were 14.54 (95% CI, 5.87–36.04) and 17.09 (95% CI, 7.49–38.96). The majority of stroke events occurred in the first week after HFRS. Seasonal effects were not observed, and apart from 1 study, neither sex nor age interacted with the associations observed in this study.
Conclusions—There is a significantly increased risk for AMI and stroke in the immediate time period after HFRS. Therefore, HFRS patients should be carefully monitored during the acute phase of disease to ensure early recognition of symptoms of impending AMI or stroke.
- cerebrovascular disorders
- hemorrhagic fever with renal syndrome
- myocardial infarction
Globally, cardiovascular diseases (CVDs) are the leading cause of death and a major source of disability. The majority of these are caused by ischemic heart disease such as acute myocardial infarction (AMI), stroke, and other cerebrovascular diseases.1 Several factors, including infectious diseases, have been associated with increased risk for CVD.2–6 The host inflammatory response to infection often results in release of proinflammatory cytokines and activation of platelets, leukocytes, and endothelial cells that can activate procoagulant pathways and can inhibit anticoagulant pathways.2,3 This can tip the balance of hemostasis toward a prothrombotic state, laying the foundation for acute cardiovascular events such as AMI and stroke.2,3
Clinical Perspective on p 1302
Viral hemorrhagic fevers are characterized by fever, hemorrhages, and shock. Abnormal coagulation and vascular dysfunction are common observations during viral hemorrhagic fevers,7 which could be risk factors for CVD. Four viral families contain members that cause viral hemorrhagic fevers: Bunyaviridae, Flaviviridae, Filoviridae and Arenaviridae. Pathogenic hantaviruses in the Bunyaviridae family cause hemorrhagic fever with renal syndrome (HFRS) in Eurasia and hantavirus cardiopulmonary syndrome in the Americas.8 HFRS caused by Puumala virus is considered a mild viral hemorrhagic fever7 characterized by thrombocytopenia, enhanced coagulation and fibrinolysis,9 and disseminated intravascular coagulopathy.10
In a highly endemic area for Puumala virus in Northern Sweden, we observed a tendency of HFRS patients to develop acute cardiovascular events. Furthermore, we have recently shown that CVD is a common cause of death during acute HFRS in Sweden.11 Together, these findings suggest a possible association between acute HFRS and CVD. Therefore, we decided to investigate in more detail whether HFRS is a risk factor for CVD. To test this, we compared the incidence of AMI and stroke in the immediate time periods before and after HFRS disease onset using the self-controlled case series (SCCS) method.12
Participants and Databases
HFRS has been a notifiable disease in Sweden since 1997. To diagnose HFRS, a blood sample is obtained from the patient and analyzed by 1 of the 3 HFRS diagnostic laboratories in Sweden. Samples are analyzed for seroconversion by either an indirect immunofluorescence method or ELISA.13 The attending medical doctor and diagnostic laboratory then separately report the HFRS diagnosis using the patient’s personal identification number to the Swedish Institute for Communicable Disease Control. This report is entered into the national HFRS database administered by the Swedish Institute for Communicable Disease Control. All patients diagnosed with HFRS in Sweden from January 1, 1997, until December 31, 2012, were included in the present study.
The National Patient Register is a nationwide database administered by the Swedish National Board of Health and Welfare, consisting of, among others, the Inpatient Register (IPR) and the Outpatient Register.14 The IPR contains information on somatic hospital discharges nationwide and has had complete national coverage since 1987. The Outpatient Register contains information on hospital-based outpatient care diagnoses.14
The personal identification numbers from the HFRS database were cross-linked with the National Patient Register to retrieve information on potential AMI/stroke events. At the time of our request to the Swedish National Board of Health and Welfare, the IPR had registered data only until 2011. All data were anonymized by the Swedish National Board of Health and Welfare before the analyses were performed. This study received ethics approval from the Regional Ethical Review Board, Stockholm, Sweden.
AMI and Stroke Events and Determination of HFRS Index Date
Individuals with International Classification of Disease (ICD; World Health Organization) main or contributing diagnosis codes for AMI/stroke were selected from the HFRS database–cross-linked IPR. The ICD, Ninth Revision (ICD-9) diagnosis codes 410 and 411 and ICD, 10th Revision (ICD-10) diagnosis codes I21 through I22 were used for AMI. The diagnoses codes for stroke were ICD-9 codes 430 to 434 and 436 and ICD-10 codes I60-64. Stroke was further subclassified into hemorrhagic stroke (ICD-10 codes I60–I62) and ischemic stroke (ICD-10 codes I63–I64). The date of an AMI or stroke event is the date of hospitalization found in the IPR for the patient. A first AMI/stroke event is defined as the first time the patient is hospitalized as a result of AMI or stroke in the IPR from 1987 to 2011.
The SCCS method is highly sensitive to the accuracy of dates.12 Therefore, we first selected individuals who had had an AMI/stroke within ±2 years of the date of report to the Swedish Institute for Communicable Disease Control. Then, each individual was reassigned a new date, henceforth called the HFRS index date, based on availability of information in the HFRS database, IPR, and Outpatient Register. The preferred date was the date of disease onset if entered into the HFRS database. If missing, we chose the date of hospitalization/visit to an outpatient clinic with the ICD-10 diagnosis code for HFRS A98.5 in the IPR and the Outpatient Register, respectively. If HFRS was reported to the Swedish Institute for Communicable Disease Control while the patient was hospitalized, the date of hospitalization was used. Finally, the dates of blood sampling, diagnosis, and report were used from the HFRS database in order of preference.
We used the SCCS method,12 a conditional Poisson regression method, to calculate relative incidence rate ratios (IRRs) for AMI and stroke. This method is based on intraperson comparisons of incidence rates of the outcome of interest (AMI or stroke event) after exposure (HFRS) in an observation period subdivided into risk and control periods.12,15 Hence, the method is conditional on participants having an AMI or a stroke event within the observation period relative to the HFRS index date (Figure 1). To avoid immortal time bias,16 which can occur when a control period is included before exposure (here HFRS), we performed 2 separate analyses. The observation period was ±365 days from the HFRS index date in analysis 1 and 365 days from the HFRS index date in analysis 2. Individuals who died within 365 days of their AMI or stroke event were excluded from analysis 1 but not analysis 2. The incubation period for HFRS is normally 3 weeks,8 but the exact date of infection is normally not possible to determine for infected individuals. Because we do not know the effect of hantavirus infection during the incubation period of HFRS on AMI and stroke incidence rates, we included a separate risk period (“buffer”) ranging from 30 days to 1 day before the HRFS index date in analysis 1 (Figure 1). The risk periods for both analyses were defined as days 1 to 21 and days 22 to 90 after the HFRS index date. Furthermore, both analyses had days 91 to 365 after the HFRS index date as a control period. Analysis 1 also had days 365 to 31 before the HFRS index date as a control period (Figure 1).
A seasonal pattern has been observed for HFRS and AMI in Sweden; both are more common during the colder months of the year.17,18 Therefore, we performed separate analyses in which we adjusted for the potential confounder seasonality by dividing the year into 2 parts: colder months (October through March) and warmer months (April through September).
We analyzed whether the median age and sex differed between the original HFRS cohort and the study groups for AMI/stroke events using the nonparametric Mann-Whitney U test for age differences and the χ2 test for sex distribution.
To investigate whether sex and age act as effect modifiers on the potential association between HFRS and AMI/stroke, we included interactions between sex and exposure and between age and exposure in the analysis. Likelihood ratio tests were then carried out to test the hypothesis that no interaction terms should be included in the model.12
In Sweden, the median age for AMI and stroke onset is 70 and 75 years, respectively.19,20 Therefore, individuals in the AMI study groups were divided into groups of ≤70 and >70 years of age. The individuals in the stroke study groups were divided into groups of ≤75 and >75 years of age.
Data were analyzed with SPSS (version 20) and Stata (version 12.0 SE).
A total of 6643 individuals were diagnosed with HFRS in Sweden from 1997 to 2012. Their median age was 52 years (interquartile range, 39–62 years); the male-to-female ratio was 1.48:1 (Table 1). The distribution of the various dates used for determining the HFRS index date is shown in Table I in the online-only Data Supplement for the HFRS cohort and in Table II in the online-only Data Supplement for the AMI/stroke study groups.
Acute Myocardial Infarction
In total, 320 of the HFRS-diagnosed individuals had been hospitalized for AMI at 1 or more time points during 1987 to 2011 (Figure 2). The study group for analysis 1 consisted of 51 individuals with 55 AMI events within ±365 days of the HFRS index date; of these 55 AMI events, 41 were first events (75%) and 14 were recurrent AMI events (25%; Figure 2 and Table 1). The study group for analysis 2 consisted of 37 individuals with 40 AMI events within 365 days of HFRS; of these 40 AMI events, 27 (68%) were first events and 13 (32%) were recurrent AMI events (Figure 2 and Table 1). The median age for individuals with AMI events was significantly higher than for the original HFRS cohort. There was a higher proportion of male to female patients in analysis 1 compared with the original HFRS cohort but not in analysis 2 (Table 1).
In analysis 1, the IRR in the first 3 weeks after HFRS was 5.53 (95% confidence interval [CI], 2.6–11.8) for all AMI events and 3.53 (95% CI, 1.25–9.96) for first AMI events (Table 2). The IRR did not differ significantly from the control periods in risk period 2 (22–90 days after HFRS onset). In analysis 2, the IRR in the first 3 weeks after HFRS was 6.02 (95% CI, 2.95–12.3) for all AMI events and 4.64 (95% CI, 1.83–11.77) for first AMI events (Table 2). As in analysis 1, the IRR for the risk period 22 to 90 days after HFRS did not differ significantly from that of the control periods. A histogram of AMI events in relation to HFRS is displayed in Figure I in the online-only Data Supplement.
No statistically significant confounding effects were observed for seasonality on the correlation between HFRS and AMI, although a borderline effect was observed for season in analyses 1 and 2 of first AMI events (Table III in the online-only Data Supplement). The season-adjusted IRR in analyses 1 and 2 for first AMI events in the time period 1 to 21 days after HFRS was 2.96 (95% CI, 1.03 – 8.47) and 3.8 (95% CI, 1.46–9.9), respectively. Apart from the study group for all AMI events in analysis 2, no modifying effects of age at HFRS >70 years and sex were found on the association between HFRS and AMI (Tables IV and V in the online-only Data Supplement). The model containing the interaction term HFRS×age was borderline significant for analysis 2, which meant that the IRRs for AMI were likely different for individuals >70 and <70 years of age (Table IV in the online-only Data Supplement). We therefore performed a follow-up analysis in which the IRRs for analysis 2 of all AMI events were recalculated separately for the 2 age groups. For individuals >70 years of age (n=14; events=16), the IRR for an AMI in the first 3 weeks after HFRS was 15.33 (95% CI, 5.15–45.62); for individuals <70 years of age (n=23; events=24), the IRR for an AMI in the first 3 weeks after HFRS was 2.92 (95% CI, 0.99–8.63).
Of all HFRS-diagnosed patients, a total of 258 individuals were hospitalized for stroke at 1 or more time points during 1987 to 2011 (Figure 2). The study group for analysis 1 consisted of 26 individuals who had 28 stroke events ±365 days of HFRS; of these, 22 (79%) were a first stroke event and 6 (21%) were recurrent stroke events (Table 1). Of the 28 stroke events, 22 (78.6%) were ischemic and 6 (21.4%) were hemorrhagic. For the first stroke events, 18 (81.8%) were ischemic and 4 (18.2%) were hemorrhagic.
The study group for analysis 2 consisted of 30 individuals who had 31 stroke events within 365 days of HFRS (Table 1); 25 (81%) were a first event and 6 (19%) were recurrent stroke events. Of the 31 stroke events, 22 (71%) were ischemic and 9 (29%) were hemorrhagic. For the first stroke events, 18 (72%) were ischemic and 7 (28%) were hemorrhagic.
The median age for those diagnosed with stroke was significantly higher than the median age of the original HFRS cohort. However, the sex distribution for all study groups with stroke did not differ significantly from that of the HFRS cohort (Table 1).
In analysis 1, the IRR for the first 3 weeks after HFRS was 12.93 (95% CI, 5.62–29.74) for all stroke events and 14.54 (95% CI, 5.87–36.04) for first stroke events (Table 3). In analysis 2, the IRR for the first 3 weeks after HFRS was 15.16 (95% CI, 7.21–31.87) for all stroke events and 17.09 (95% CI, 7.49–38.96) for first stroke events (Table 3). The IRRs for the second risk period (after 3 weeks following HFRS onset) did not differ significantly from the control periods in either analysis (Table 3). There were no stroke events in the third week; therefore, we calculated the IRRs for the first and second weeks after HFRS. Interestingly, in analysis 1, the IRRs in the first week after HFRS were 33.94 (95% CI, 14.18–81.28) and 37.41 (95% CI, 14.38–97.35) for all and first stroke events, respectively. There was no significant difference from the control periods for the IRRs in the second week. Furthermore, the IRRs for analysis 2 were 36.4 (95% CI, 16.6–79.78) and 41.08 (95% CI, 17.75–95.09) for all and first stroke events in the first week after HFRS, respectively. The IRRs for all and first stroke events in the second week after HFRS were 9.1 (95% CI, 2.6–31.96) and 12.32 (95% CI, 3.48–43.66), respectively. A histogram of stroke events in relation to time after HFRS is shown in Figure II in the online-only Data Supplement.
There were no significant seasonal effects on stroke events (Table III in the online-only Data Supplement). No effects on the association between HFRS and stroke by age at HFRS above and below 75 years and sex could be observed in analyses 1 and 2 (Tables IV and V in the online-only Data Supplement).
Our results show that HFRS is linked to a transient but significant increased risk for AMI and stroke, suggesting a role for hantavirus infection in triggering these events. This finding was demonstrated by 2 separate analyses taking into account the impact of fatal AMI/stroke events using the SCCS method. The SCCS method is sensitive to timing and dates12; therefore, it is crucial the HFRS index dates are close to disease onset. The least accurate date, that is, the date of report to the Swedish Institute for Communicable Disease Control, was used in only ≈5% of the events for the AMI study groups, and this was not even an issue in the stroke study groups. Thus, the majority of dates fall within the first risk period, enabling a good temporal distinction between AMI/stroke events occurring in the risk and control periods.
A critical assumption for the SCCS method is that the occurrence of an event (AMI or stroke) does not alter the likelihood of exposure (HFRS).21 This is called the immortal time bias and could potentially inflate the IRR for AMI and stroke.16 In essence, we retrospectively select for patients who have been diagnosed with HFRS and AMI/stroke. Individuals who develop fatal AMI/stroke events before a theoretical Puumala virus infection will obviously not figure into our cohort. Therefore, the incidence rates of AMI/stroke in the year before HFRS in our study could be underestimated. We circumvent this issue by omitting individuals who died within 365 days of their AMI/stroke events in analysis 1. Hence, we avoid introducing the immortal time bias in our studies.
One advantage of SCCS is that multiplicative effects of fixed confounders factor out; therefore, the baseline risk for cardiovascular events can be considered to be removed when shorter periods of time are analyzed.12 However, it is possible that fixed effects can still act as effect modifiers.12 For instance, although seroprevalence for Puumala virus is represented equally in both sexes, the disease HFRS is overrepresented in middle-aged men.17,22 Therefore, sex and age group might be factors that determine who is more likely to be diagnosed with HFRS. In our study, we investigated whether there was a significant difference in the IRRs for AMI or stroke after HFRS between male and female patients (sex) and age groups. Therefore, we included a separate test to identify whether an interaction term should be included in our model for age group and HFRS or sex and HFRS. No significant modifying effects were found for sex and HFRS or age and HFRS on the association between HFRS and AMI/stroke, apart from analysis 2 of all AMI events for which age had a significant interaction. When calculated separately, patients in the age group >70 years had a 15-fold higher risk of developing AMI in the first 3 weeks after HFRS. It is possible that older age might increase the risk for AMI after HFRS. This finding is also in line with previous observations that mortality after HFRS increased with age.23
Furthermore, we compared the age and sex distributions between participants of the study groups and those in the original complete HFRS cohort and found that those who had had AMI and stroke were significantly older. In general, there was a trend toward a higher proportion of male patients in the study groups compared with the original HFRS cohort, although it was only statistically significant in AMI analysis 1.
In Sweden, an increased AMI incidence has been reported for the colder months compared with the warmer months,18 and a seasonal trend could perhaps be present in the analysis of first AMI events. The season-adjusted IRR for the analysis of first AMI events in the first 3 weeks after HFRS was still significantly increased. There were no seasonal effects in the other study groups.
Our study was based on data obtained from the National Patient Register. It has been reported that the validity for diagnosis codes relating to AMI and stroke was 100% and 98.6%, respectively, indicating very good quality of the data.14
Previous studies have linked acute infections with an increased risk for AMI and stroke.2–6 One of these studies was based on the UK General Practice Research Database using SCCS, the same method used here.5 In that study, the IRR for a first AMI event in the first 3 days after disease onset of systemic respiratory and urinary tract infections was increased 5- and 3-fold, respectively.5 This is similar to the finding in our study that the IRR for first AMI events is increased ≈4-fold in the first 3 weeks after HFRS compared with the control periods. Furthermore, Smeeth et al5 show that the risk for a first stroke event was significantly increased ≈3-fold the first week after both upper respiratory and urinary tract infections. It is noteworthy that our IRRs are high for a first stroke event after HFRS, ≈40-fold higher than in the control periods, indicating that HFRS is a strong risk factor for the development of acute cerebrovascular events. To the best of our knowledge, no prior studies have uncovered an association between an acute infection and stroke of the magnitude observed here.
The possibility that HFRS is a trigger for acute cardiovascular events is in line with previous studies focusing on hantaviral infection and pathogenesis. HFRS patients have abnormal coagulation with elevated thrombosis and fibrinolysis9 and disseminated intravascular coagulopathy.10 Additionally, hantavirus infection seems to have a negative impact on heart functions; cardiac disturbances have been observed in more than half of the HFRS patients.24 Furthermore, the central nervous system was also shown to be affected by hantavirus infection, with reports of cerebral hemorrhage, the presence of Puumala virus RNA, and increased levels of the monocyte/macrophage activation marker neopterin in cerebrospinal fluid.25–28 HFRS also causes intravascular damage29 and dysfunctional platelets.30 Endothelial cells are infected with hantaviruses with subsequent activation.31–34 Consequently, hantavirus infection induces an enhanced state of inflammation and coagulation, which might contribute to the increased risk for AMI/stroke during acute HFRS reported here. The enhanced coagulopathy in patients observed in prior studies10 could also explain why the majority of stroke cases in our study were ischemic rather than hemorrhagic, as could possibly be expected from a hantaviral hemorrhagic fever.
Case reports suggest an association between the dengue virus and stroke,35 but to the best of our knowledge, this issue has not been studied in a cohort of dengue virus–infected patients. Furthermore, the dengue virus was also shown to cause cardiac disease in a recent study.36 We do not know whether our results could apply to other viral hemorrhagic fevers, even though coagulation disorders are more frequent.37,38 It remains to be investigated whether viral hemorrhagic fever in general is associated with increased risk of AMI/stroke.
We identified an increased risk for AMI and stroke in the first weeks after HFRS, but a limitation of our study was the small number of events and the low precision of the estimated IRR. To determine the IRR more precisely, more cases would have to be included. However, the HFRS database contains information on all diagnosed HFRS cases in Sweden since 1997 (n=6643), which we have included in our study. Therefore, it would be impossible to increase the size of our cohort. As for other infectious diseases, there is a considerable proportion of unrecognized HFRS cases,22 and whether our findings also apply to them is unknown. Because fewer female patients are diagnosed with HFRS, there is a risk of selection bias with regard to sex.23 Finally, the majority of exposure dates were based on the date of disease onset. It is possible that the date of disease onset was inaccurate in some cases as a result of patient recall bias and error in data entry, although this probably would introduce an error of a few days, and the risk period is measured in weeks.
HFRS is associated with a transient but significantly increased risk for the acute cardiovascular events AMI and stroke, suggesting that hantavirus can trigger severe cardiovascular events in infected patients.
We gratefully acknowledge Professor Paddy Farrington of The Open University for his valuable input and advice regarding the SCCS method.
Sources of Funding
This study was funded by the Swedish Heart-Lung Foundation, County Councils of Northern Sweden, County Council of Västerbotten, Medical Faculty of Umeå University, Swedish Research Council, Karolinska Institutet, Swedish Foundation for Strategic Research, and Mats Klebergs foundation.
The online-only Data Supplement is available with this article at http://circ.ahajournals.org/lookup/suppl/doi:10.1161/CIRCULATIONAHA.113.001870/-/DC1.
- Received December 3, 2012.
- Accepted December 20, 2013.
- © 2014 American Heart Association, Inc.
- 1.↵Global Atlas on Cardiovascular Disease Prevention and Control. Geneva, Switzerland: World Health Organization;2011.
- Warren-Gash C,
- Hayward AC,
- Hemingway H,
- Denaxas S,
- Thomas SL,
- Timmis AD,
- Whitaker H,
- Smeeth L
- Jonsson CB,
- Figueiredo LT,
- Vapalahti O
- 18.↵Swedish National Board of Health and Welfare. The Swedish AMI Statistics. http://www.socialstyrelsen.se/statistik/statistikefteramne/hjartinfarkter. Accessed August 8, 2012.
- 19.↵Swedeheart. Annual Report Swedeheart 2010. http://www.ucr.uu.se/swedeheart/index.php/arsrapporter. Accessed June 19, 2012.
- 20.↵Västerbottens County Council. The Swedish Stroke Register, Annual Report 2010. http://www.riks-stroke.org/index.php?content=analyser. Accessed June 19, 2012.
- Whitaker HJ,
- Hocine MN,
- Farrington CP
- Makela S,
- Kokkonen L,
- Ala-Houhala I,
- Groundstroem K,
- Harmoinen A,
- Huhtala H,
- Hurme M,
- Paakkala A,
- Porsti I,
- Virtanen V,
- Vaheri A,
- Mustonen J
- Alexeyev OA,
- Morozov VG
- Hautala T,
- Mähönen SM,
- Sironen T,
- Hautala N,
- Pääkkö E,
- Karttunen A,
- Salmela PI,
- Ilonen J,
- Vainio O,
- Glumoff V,
- Rytky S,
- Plyusnin A,
- Vaheri A,
- Vapalahti O,
- Kauma H
- Klingström J,
- Hardestam J,
- Stoltz M,
- Zuber B,
- Lundkvist A,
- Linder S,
- Ahlm C
- Cosgriff TM,
- Lee HW,
- See AF,
- Parrish DB,
- Moon JS,
- Kim DJ,
- Lewis RM
- Shrivastava-Ranjan P,
- Rollin PE,
- Spiropoulou CF
- Björkström NK,
- Lindgren T,
- Stoltz M,
- Fauriat C,
- Braun M,
- Evander M,
- Michaëlsson J,
- Malmberg KJ,
- Klingström J,
- Ahlm C,
- Ljunggren HG
- Miranda CH,
- Borges Mde C,
- Matsuno AK,
- Vilar FC,
- Gali LG,
- Volpe GJ,
- Schmidt A,
- Pazin-Filho A,
- Silva FM,
- Castro-Jorge LA,
- Oliveira MF,
- Saggioro F,
- Martines RB,
- Fonseca BA
During the first year after hantavirus infection, a higher proportion of the causes of death are cardiovascular. Hantavirus infections activate endothelial cells, causing increased platelet binding and activation. Furthermore, approximately one-fourth of patients diagnosed with hemorrhagic fever with renal syndrome (HFRS), a hantaviral disease, fulfill the criteria for disseminated intravascular coagulopathy. These are imminent portents indicating that cardiovascular sequelae may occur during hantaviral disease. From a cohort of 6643 HFRS patients, we identified individuals with either acute myocardial infarction (AMI) or stroke within the year before and after HFRS disease onset. By using the individuals as their own controls in the self-controlled case series method and by applying 2 different types of analysis, we calculated the incidence rate ratio for AMI and stroke in the first 3 weeks after HFRS. Both analyses showed an increased risk of AMI (≈4- and 6-fold for first and all AMI events, respectively) and stroke (≈16- and 14-fold for first and all stroke events, respectively) in the first 3 weeks after HFRS. The finding that HFRS is a risk factor for AMI and stroke suggests that HFRS patients should be carefully monitored during the course of infection to ensure early recognition of symptoms of impending stroke or AMI and to initiate adequate measures if required. Further studies are needed to investigate the potential benefits of anticoagulants or other treatments to prevent cardiovascular complications during HFRS.