Thromboembolic Events in Patients with Influenza: A Scoping Review

Introduction: Influenza is an acute respiratory infection that usually causes a short-term and self-limiting illness. However, in high-risk populations, this can lead to several complications, with an increase in mortality. Aside from the well-known extrapulmonary complications, several studies have investigated the relationship between influenza and acute cardio and cerebrovascular events. Reviews of the thromboembolic complications associated with influenza are lacking. Objectives: the study aims to conduct a scoping review to analyze the epidemiological and clinical characteristics of patients suffering from influenza and thromboembolic complications. Materials and methods: A computerized search of historical published cases using PubMed and the terms “influenza” or “flu” and “thrombosis”, “embolism”, “thromboembolism”, “stroke”, or “infarct” for the last twenty-five years was conducted. Only articles reporting detailed data on patients with thromboembolic complications of laboratory-confirmed influenza were considered eligible for inclusion in the scoping review. Results: Fifty-eight cases with laboratory documented influenza A or B and a related intravascular thrombosis were retrieved. Their characteristics were analyzed along with those of a patient who motivated our search. The localizations of thromboembolic events were pulmonary embolism 21/58 (36.2%), DVT 12/58 (20.6%), DVT and pulmonary embolism 3/58 (5.1%), acute ischemic stroke 11/58 (18.9%), arterial thrombosis 4/58 (6.8%), and acute myocardial infarction 5/58 (8.6%). Discussion: Our findings are important in clarifying which thromboembolic complications are more frequent in adults and children with influenza. Symptoms of pulmonary embolism and influenza can be very similar, so a careful clinical evaluation is required for proper patient management, possible instrumental deepening, and appropriate pharmacological interventions, especially for patients with respiratory failure.


Introduction
Influenza is an acute respiratory infection caused by influenza viruses (A or B) which are spread from person to person primarily through droplets and during the winter season. Annually, influenza infects approximately 10% to 20% of the world's population causing mostly a non-complicated, self-limiting disease [1]. Influenza outbreaks occur in two distinct patterns: endemic (seasonal) and pandemic. Pandemics are typically decades apart and are linked to the influenza A virus strain with novel hemagglutinin molecule forms [2,3].
The virulence of influenza A viruses depends on antigenic drift or reassortment between human and animal influenza A viruses that allows the virus to evade host immunity [4]. In high-risk populations, influenza can cause several complications, with an increase in mortality. Obesity, pregnancy, immunosuppression, chronic pulmonary diseases, a computed tomography pulmonary artery examination was performed on the seventh day of hospitalization, revealing opacification defects in the left main branch, segmental branches of both the left upper and lower lobes, and the tributary branches of the posterior segment of the right lower lobe. On echocardiography, the right ventricle appeared normal.
In light of the above finding, fondaparinux 7.5 mg once daily was immediately administered. The patient improved and no longer required oxygen and was switched to an oral regimen containing dabigatran 150 mg twice daily.
A computed tomography pulmonary artery examination was repeated a week later, revealing nearly complete resolution of both the embolic and parenchymal images. The patient was discharged with no complications.

Literature Review
A computerized search of historical published cases was conducted using PubMed. The search was conducted for the period including the last twenty-five years, using the following string terms: "influenza" [Title/Abstract] OR "flu" [Title/Abstract] AND "thrombosis" [Title/Abstract] OR "embolism" [Title/Abstract] OR "thromboembolism" [Title/Abstract] OR "stroke" [Title/Abstract] OR "infarct" [Title/Abstract]. Furthermore, all references were hand-searched for additional relevant articles, and a citation tracker was used to identify any additional relevant literature.
The most recent search was executed in May 2022, and the search results were case reports and case series.
Articles published in languages other than English, French, or Italian were not considered. The selected articles were separately reviewed by two independent authors (A.C. and C.I.).
If an article reported detailed data on patients with thromboembolic complications of a laboratory-documented influenza, it was considered eligible for inclusion in the systematic review. We also included children and people under the age of 18.
For each patient included in the study, the following parameters were taken into account: sex, age, site of thrombosis, risk factors and/or co-morbidities, other complications different from thromboembolism, and outcomes.
The guidelines for Preferred Reporting Items for Systematic Review and Meta-Analysis (Moher et al. PRISMA) [10] and the PRISMA Extension for Scoping Reviews (PRISMA-Scr): Checklist and Explanation [11] were followed.

Results
Forty-one articles reporting 8086 cases of respiratory infection-related intravascular thrombosis (arterial and venous) were found. Eleven articles with a total of 8029 cases were excluded. One was excluded because there was no description of the characteristics

Results
Forty-one articles reporting 8086 cases of respiratory infection-related intravascular thrombosis (arterial and venous) were found. Eleven articles with a total of 8029 cases were excluded. One was excluded because there was no description of the characteristics of the population studied, and the cause of the stroke was determined to be secondary to hypotension [12]; the other 10 articles looked at patients who had a stroke or a myocardial infarction but whose characteristics were not described, and the respiratory tract infections included viruses other than influenza viruses [13][14][15][16][17][18][19][20][21][22]. Finally, 30 full-text articles containing 57 cases of patients with laboratory-documented influenza A or B and a related intravascular thrombosis (arterial and venous) diagnosed through instrumental evaluations were considered [9,. Table 1 analytically shows the epidemiological and clinical characteristics, complications, and outcomes of the study population. Eight articles reported cases of influenza-related intravascular thrombosis in children (mean age 6.2 years), including a 16-year-old boy [43][44][45][46][47][48][49][50][51]. Thirty-six (62%) were males, and twenty-two (37.9%) were females. There were seven females and one male among the children under the age of 14. The median age was 38 years old and was calculated from fifty-seven patients because the age was not reported in one case. Adults had an average age of 42.2 years. Documented influenza A infections were found in 55/58 (94.8%) of cases. Positive PCR biomolecular test results for influenza A were found in 50/58 of the patients, while the diagnostic method was not reported in 5/58 of the patients [27,38,42,45,48]. Influenza B was found in 3/58 of the patients. In one case, the method was not reported, and in the other two, a positive antigen on a nasal swab was identified [36,38,41]. Except for eight cases, all influenza A cases were H1N1. Two cases were serologically identified as influenza A H3N2, while the remaining six were not [24].
The eight children were not immunocompromised. Five of them tested positive for H1N1 influenza A and three for non-typed influenza A. One of them had a basilar artery infarction that progressed to a locked-in syndrome [44]; another child with a protein S deficiency, caused by a heterozygous K196E mutation, suffered from a complete thrombosis of the superior mesenteric vein, necessitating a resection of the ischemic small intestine [45]. A patient with a low protein S activity (45%-normal value 57-125%) and positivity for HLA A29 presented with bilateral ischemic maculopathy with encephalitis progression [48]. Breker DA et al. [49] reported retinal and geniculate infarction, causing vision loss in a previously healthy thirteen-year-old girl, whereas Calzedda R et al. [46] described a case of a two-year-old Caucasian female admitted to the emergency department with an acute onset of fever, dry cough, left hemiparesis, speech difficulties, and persistent weakness associated with an ischemic stroke on MRI that was confirmed by magnetic resonance angiography, as well as the presence of H1N1 RNA copies (>500 copies/mL) and elevated IL 1-β and IL 6 levels in CSF samples. Bell et al. and Honorat R et al. both described an onset of fever and seizures, and a magnetic resonance angiography confirmed left middle cerebral artery infarction [43,47]. Finally, Javedani PP et al. described an acute infarction involving the right frontal, parietal, temporal, and occipital lobes, as well as hyper densities suggestive of thrombosed cortical veins in a four-month-old female who presented with fever, decreased oral intake, and limp appearance after antibiotic administration [50].
A 16-year-old boy was immunocompromised due to cerebral palsy; he tested positive for H1N1 and developed a deep venous thrombosis in his left leg [50]. Six of the young patients received antiviral therapy [44][45][46][47][48][49], with data missing in only two cases [43,50]. All pediatric patients survived, albeit with complications. Only one patient recovered completely [47], and in only one case, the data was not reported [50].
Other complications occurred in 30/58 (51.7%) of the cases. A total of 23/58 (39.6%) of the patients were admitted to intensive care units and underwent mechanical ventilation, and 5/58 (8.6%) underwent non-invasive mechanical ventilation.

Discussion
Our patient had pulmonary embolism while on thromboprophylaxis for severe influenzarelated pneumonia, which prompted the literature review. In 33/58 (56.8%) of the patients, no significant comorbidity or modifiable risk factors for thromboembolic events were reported. Except for a certain prevalence of the male sex (M/F 2.3:1) and of obese and hypertensive patients (17% and 19% of the total population, respectively), it was not possible to identify specific risk factors for the development of thromboembolic events during influenza. However, our study was not designed for such purposes. VTE incidence varies by gender across the lifespan. In younger adulthood, women have a slightly higher annual incidence of VTE than men, but after midlife, men have a faster increase in VTE incidence than women [52]. According to the literature, twenty of our patients over the age of 40 were male, while seven were female. Surprisingly, most of the patients in the study were under the age of 50 (35/58, 60.3%), despite old age being one of the most common risk factors for venous thromboembolism [53][54][55].
Many studies in the literature indicate that Black Americans are at a higher risk of VTE than are white Americans, and Asian populations have, by far, the lowest incidence of VTE. The reasons for these potential racial differences remain unknown, with risk factors such as obesity, diabetes, and elevated factor VIII being more common in Black people, and genetic polymorphisms such as factor V Leiden and the prothrombin gene 20210A mutation being more common in white people. However, data on the incidence of and risk factors for VTE outside of Europe and North America, as well as in populations with non-European ancestry, are limited, and some of these differences may be due to a lack of surveillance for VTE, a lack of suspicion in "low risk populations", or a lack of access to medical care, as well as different regional scenarios [56]. Unfortunately, the ethnicity of the patients was not indicated in many studies, so we cannot draw conclusions about this condition.
The most common onset symptom in our series was respiratory insufficiency (18/58 of the patients, 31.3%). Most patients had an H1N1 infection, which could indicate that this strain has a greater ability to induce thromboembolic phenomena.
Except for one case of complete superior mesenteric vein thrombosis, no pulmonary embolic events or deep-vein thrombosis were reported in the pediatric population. Stroke was the only other complication reported in this last population. In adults, on the other hand, venous thromboembolic events were the most frequently reported. This could underlie different mechanisms and risk factors for thromboembolic events in adults and children. In the current study, 36.2% of the cases, including our patient, had "de novo" pulmonary thromboembolism with no evidence of underlying DVT. This finding contrasts with the incidence of PE "de novo" reported in the literature among non-influenza patients, which ranged from 0% to 22.6% [57,58]. This suggests that local inflammation of the lung parenchyma may theoretically contribute to endothelial activation and subsequent induction of a hypercoagulable state in the surrounding microenvironment, similar to the endothelial response preceding the development of DVT in the extremities, with the potential to progress to local "de novo" thrombosis of the pulmonary vessels. This may be related to "immunothrombosis". Moreover, during influenza virus infection, the extrinsic coagulation pathway was stimulated with a reduced generation of key inhibitors of coagulation and fibrinolysis, namely activated protein C and plasminogen-activator inhibitor type-1 [59]. However, influenza was not identified as an independent risk factor in a nested case-control study of patients suspected of having pulmonary embolism (adjusted OR 0.22, 95% CI 0.03-1.72) [1,8,59].
Data on thromboembolic risk in pregnancy are conflicting, but pregnancy and the post-partum period are known to be hypercoagulable states. In developed countries, pulmonary embolism is an important cause of maternal mortality [60][61][62]. Acute myocardial infarction is uncommon in pregnant women, especially those under the age of 30 [63]. Only three patients in our series were pregnant, two of them under 30 years old, and they had a deep venous thrombosis, an extensive cerebrovascular accident, and an acute myocardial infarction.
Screening for congenital hypercoagulable states was not performed or reported in most of the cases included in our case, so we cannot rule out the possibility that some events were caused by factors other than influenza virus infection. SARS-CoV-2 and influenza viruses share many similarities, including transmission routes and clinical presentations. The influenza virus, like SARS-CoV-2, has extensive effects on inflammatory and coagulation pathways and may be a cause of cardiovascular disease [59,[64][65][66]. A significant component of susceptibility is attributed to host genetics, such as host frailty, in addition to the underlying hyperinflammatory syndrome, which is the most implicated in cardiovascular complications in both viruses [59,67,68]. Adult influenza is typically a self-limiting disease; thus, it is frequently not reported to public health authorities, resulting in underreported data [66]. Regardless, various epidemiological studies have suggested a link between cardiovascular mortality and influenza, but unlike for SARS-CoV-2 infections, there have been no studies evaluating the progression and prognostic value of cardiac markers for influenza. Although neutrophil-driven immunothrombosis is a key component of severe COVID-19, innate immunity and neutrophil extracellular trap formation (NETosis) have also been linked to influenza. Immunothrombotic vessel occlusion and NETosis, on the other hand, have been found to be significantly higher in severe SARS-CoV-2 pneumonia compared to influenza pneumonia [69]. Moreover, it should be noted that over 48% of patients had severe disease requiring invasive and non-invasive mechanical ventilation, and this could be linked to the development of thrombosis.
In conclusion, our findings are important in clarifying that thrombosis is a complication seen with influenza in adults and children. Thrombotic events can occur despite thromboprophylaxis in hospitalized patients. Symptoms of pulmonary embolism and influenza can be very similar, so a careful clinical evaluation is required for proper patient management, possible instrumental deepening with pulmonary CT angiography, and appropriate pharmacological interventions, especially for patients with respiratory failure. To better understand this phenomenon, it could be interesting to perform a molecular swab to search for influenza viruses in all cases of thromboembolic events occurring in the winter period. Furthermore, because cerebral ischemic events, myocardial infarction, and deep venous thrombosis are time-dependent pathologies, it is critical to maintain high clinical vigilance in influenza infections and promptly request brain CT scans, electrocardiograms, or arterial or venous Doppler echo-color examinations to rule them out or proceed with early treatment. This is particularly important in patients with complicated influenza in the absence of rapid clinical resolution or in cases of worsening respiratory function or the appearance of symptoms suggestive of organ damage, seemingly unrelated to influenza.
Author Contributions: A.C. conceived the work; R.R. searched and selected the articles; A.C. and C.I. (Claudia Imburgia) have separately reviewed the selected articles; R.R. and C.I. (Chiara Iaria) critically evaluated the selected articles, collected and analyzed the data, and wrote the article; S.B. and M.T. contributed to the management of the clinical case presented and its description; A.C. supervised and coordinated the work and revised the manuscript. All authors contributed to the final version of the paper. All authors have read and agreed to the published version of the manuscript.
Funding: There are no funding sources to declare.

Informed Consent Statement:
The clinical case patient has given his written permission for the data to be published.

Data Availability Statement:
All data used and/or analyzed during this study are included in this published article.

Conflicts of Interest:
The authors declare no conflict of interest.