Influenza Virus-Related Critical Illness

Prevention, Diagnosis, Treatment

Eric J. Chow; Joshua D. Doyle; Timothy M. Uyeki


Crit Care. 2019;23(214) 

In This Article

Treatment of Influenza

Treatment of severe influenza presents multiple challenges. The mainstay of therapy for patients with influenza is initiation of antiviral medication as soon as possible after illness onset.[49] Currently available FDA-approved antiviral medications include neuraminidase inhibitors (NAIs) (e.g., oral oseltamivir, inhaled zanamivir, and intravenous peramivir); cap-dependent endonuclease inhibitor (baloxavir marboxil); and adamantanes (e.g., amantadine and rimantadine) (Table 4). NAIs and baloxavir have activity against both influenza A and B viruses. Adamantanes only have activity against influenza A viruses and are not recommended for treatment of influenza due to widespread resistance among currently circulating strains of seasonal influenza A viruses. Notably, FDA-approved antiviral medications for treatment of influenza are approved for early treatment of uncomplicated influenza in outpatients based upon randomized placebo-controlled clinical trials conducted among previously healthy outpatients. Meta-analyses of randomized placebo-controlled clinical trials of early oseltamivir treatment of influenza in pediatric and adult outpatients have reported clinical benefit in reducing duration of illness and risk for some complications associated with influenza.[65,66]

No completed randomized, placebo-controlled trials of antiviral treatment have been conducted in hospitalized influenza patients to establish the efficacy of oseltamivir or other NAIs. A number of observational studies have reported clinical benefit of neuraminidase inhibitors in hospitalized patients, including reduction in duration of hospitalization and risk of death, including in ICU patients.[67–74] Additionally, a systematic review of published reviews/meta-analyses reported survival benefit of NAI treatment in hospitalized patients,[75] although another meta-analysis of observational studies did not.[69] In particular, a large pooled individual patient-level meta-analysis of observational studies from 38 countries identified a 38% reduction in risk of mortality in critically ill adults and those aged ≥ 16 years old when comparing early NAI treatment (< 48 h) with later treatment (> 48 h), and a 69% reduction in mortality risk between influenza patients receiving early NAI treatment and those who did not receive NAIs.[72] The mortality risk reduction of NAI treatment at any time versus no treatment was 28% for critically ill patients aged ≥ 16 years old; while a similar reduction in mortality was identified in critically ill children aged < 16 years, the result was not statistically significant[72] and was likely underpowered because death is less common in hospitalized children with influenza than in adults.

Although studies have shown the greatest clinical benefit when antivirals are started within 2 days of illness onset, some observational studies have shown clinical benefit of neuraminidase inhibitors when started up to 5 days following symptom onset.[15,55,76,77] The large meta-analysis mentioned above also identified a significantly reduced mortality risk reduction (35%) in critically ill patients aged ≥ 16 years old who received NAI treatment > 48 h after symptom onset compared with those who did not.[72] A cohort study of early versus late oseltamivir treatment reported a significant reduction in mortality and median duration of ICU hospitalization in severely ill patients with influenza A(H3N2), but not A(H1N1pdm09) or B virus infection in Greece.[78] One French study reported delays in initiation of oseltamivir treatment prescribed to hospitalized influenza patients and suggested empiric administration of oseltamivir treatment in the emergency department for patients being admitted with lower respiratory tract disease during influenza season.[79] Overall, based upon available observational data to date in hospitalized patients with influenza, including ICU patients, initiation of neuraminidase inhibitor antiviral treatment is recommended as soon as possible for hospitalized patients with suspected or confirmed influenza.

Data on optimal dosing and duration of therapy with neuraminidase inhibitors are limited in critically ill influenza patients. Enterically administered oseltamivir is the preferred treatment for most hospitalized patients, given the lack of data for intravenous peramivir in this population. The use of inhaled zanamivir is not recommended in in critically ill patients due to the lack of data in hospitalized patients and the risk of bronchospasm in patients with underlying lung disease. Studies indicate that oseltamivir administered orally or via oro/naso-gastric tube is well absorbed in critically ill patients and reaches plasma levels comparable to those in ambulatory patients.[80] Similarly, several observational studies indicate that enteric oseltamivir reaches comparable plasma concentrations to non-critically ill patients in those receiving extracorporeal membrane oxygenation (ECMO) and renal replacement therapy,[80–87] although dosing should be reduced in patients with significant renal impairment. There is scant evidence that increased NAI dosing (e.g., twice daily dosing) in critically ill patients provides additional clinical benefit than standard dosing.[80,88–92] Of note, studies also suggest that increased oseltamivir dosing does not provide additional clinical benefit in obese adults, including extreme obesity (BMI > 40).[93,94] Duration of therapy can be difficult to define, as prolonged influenza viral replication and shedding from the both upper and lower respiratory tract can occur in critically ill patients.[95,96] For this reason, it may be beneficial to continue antiviral therapy beyond 5 days, and repeat virologic testing may be beneficial in determining appropriate therapeutic endpoints.[97] Continuing antiviral treatment in critically ill patients until virus is not detectable in the lower respiratory tract may also help reduce the pro-inflammatory dysregulated cytokine response triggered by influenza virus infection and reduce nosocomial influenza virus transmission to healthcare personnel in the ICU. Consultation with a specialist with training in infectious diseases for the potential emergence of antiviral resistant virus infection should be considered for ICU patients with evidence of persistent influenza viral replication after NAI treatment, particularly in severely immunocompromised patients.[49,98]

For patients who cannot tolerate or absorb enteric oseltamivir due to gastric stasis, malabsorption, or other gastrointestinal processes, intravenous peramivir may be an alternative;[99,100] however, studies have not identified an advantage for intravenous peramivir in comparison with enteric oseltamivir.[101] Notably, a randomized trial conducted in three influenza seasons found similar clinical outcomes between IV peramivir and enteric oseltamivir in hospitalized adult influenza patients;[102] a separate trial did not identify significant additional clinical benefit of peramivir in combination with standard-of-care therapy (which often included an NAI).[103] A more recent, multicenter randomized controlled trial also found similar clinical benefit between enteric oseltamivir and intravenous peramivir in hospitalized influenza patients.[104]

In 2018, a novel antiviral agent, baloxavir marboxil, was FDA-approved for early treatment of uncomplicated influenza in outpatients aged ≥ 12 years old. Baloxavir acts via inhibition of the influenza virus cap-dependent endonuclease, a different mechanism than the neuraminidase inhibitors, and can treat NAI-resistant influenza virus infections. Randomized controlled trials of single-dose oral baloxavir showed similar clinical benefit to 5 days of twice-daily oral oseltamivir.[105] However, because these studies were limited to patients with uncomplicated influenza, the role of baloxavir monotherapy or in combination with an NAI for treatment of hospitalized influenza patients is unclear. Specifically, optimal dosing, duration of therapy, and appropriate endpoints have yet to be determined for baloxavir treatment of hospitalized influenza patients. In the outpatient RCT, patients treated with single-dose baloxavir showed significant reduction in influenza viral levels in the upper respiratory tract at 24 h compared with those receiving placebo or oral oseltamivir.[105] However, it is unknown whether this reduction in influenza viral shedding correlates with reduced transmissibility. A potential concern for the use of baloxavir in critically ill patients is the rapid development of resistance observed during the outpatient clinical trials.[106] A trial to assess the efficacy and safety of baloxavir in combination with oseltamivir versus oseltamivir monotherapy in hospitalized influenza patients is currently enrolling participants.[107]

There are no completed randomized clinical trials of adjunctive corticosteroid treatment in influenza patients. A trial of corticosteroid therapy was planned during the 2009 H1N1 pandemic, but was halted due to limited number of enrolees.[108] One observational study in China during the 2009 H1N1 pandemic reported that administration of parenteral glucocorticoids within 72 h of illness onset tripled the risk of developing critical illness or death from influenza A(H1N1)pdm09 virus infection.[109] A re-analysis of prospectively collected data on 1846 influenza patients admitted with primary influenza pneumonia to 148 ICUs in Spain during 2009–2014 using propensity scoring matching reported that corticosteroid use was significantly associated with ICU mortality.[110] Meta-analyses of observational studies have concluded that that corticosteroid treatment of hospitalized influenza patients does not result in better outcomes and may be associated with adverse outcomes including increased mortality.[111–113] Similarly, a retrospective observational study conducted on critically ill children during the 2009 H1N1 pandemic found that high-dose (equivalent to 2 mg/kg per day of methylprednisolone) corticosteroid treatment was associated with mortality in the ICU, although a causative relationship was not determined.[30] A selection of individual observational studies in critically ill children and adults have also reported potential association between corticosteroid treatment and adverse influenza outcomes.[30,114,115] A recent Cochrane review of available observational studies suggested increased mortality when adjunctive corticosteroid therapy is used for influenza patients; however, the available evidence was of low quality and the authors suggest interpreting these results with caution.[116]

Multiple studies have reported that corticosteroid treatment is associated with prolonged influenza viral shedding in hospitalized patients,[117–119] including in sporadic human infections with avian influenza A(H7N9) virus in China,[120] and increased rates of secondary bacterial and fungal co-infections,[121,122] which may lead to adverse clinical outcomes. However, there is some evidence to suggest that the increased risk attributed to corticosteroid treatment is a result of bias in observational studies. A large, retrospective study of critically ill adults in Canada found an increased risk of mortality in patients receiving corticosteroids; however, after adjusting for time-dependent differences between groups, no significant differences in mortality were observed with corticosteroid treatment.[123] Moreover, potential differences between low-dose and medium-/high-dose corticosteroid treatment are not well understood. One observational study of hospitalized patients with viral pneumonia due to avian influenza A(H7N9) virus infection in China reported that high-dose, but not low or moderate-dose corticosteroids, was associated with increased 30-day and 60-day mortality.[124] Currently, on the basis of available observational data to date, adjunctive corticosteroid treatment is not recommended for children or adults hospitalized with influenza, including critically ill patients, unless clinically indicated for another reason, such as treatment of asthma or COPD exacerbation or septic shock.[49] Further studies are required to understand the clinical benefit or harms associated with corticosteroid treatment of critically ill influenza patients.

Although neuraminidase inhibitors (oseltamivir) are currently recommended for antiviral treatment of influenza in hospitalized patients based on observational studies, including in critically ill patients, there are a number of novel strategies and products for treating influenza in various stages of development. One approach under investigation is triple-combination antiviral drug (TCAD) therapy, which combines amantadine, ribavirin, and oseltamivir for treatment of influenza in critically ill and high-risk patients. Unfortunately, studies to date have not shown a benefit of TCAD over oseltamivir monotherapy.[125–127] Several novel antiviral compounds are in various stages of investigation, including small-molecule polymerase inhibitors such as pimodivir[128] and favipiravir.[129] A number of monoclonal and polyclonal antibodies, targeted against a variety of influenza viral proteins, are also in development.[130–133] Similarly, convalescent plasma has shown potential benefit in the treatment of severe influenza, and further trials are underway.[134–136] Another area of intense interest is the modification of the host antiviral response to influenza virus infection. There are ongoing preclinical and clinical studies of a variety of other immunomodulatory agents for treatment of influenza, including celecoxib,[137] statins, etanercept, pioglitazone, azithromycin,[138] and interferons.[139]