Influenza 2010–2011: Lessons from the 2009 pandemic

Antiviral therapy efficacious only if started early in ambulatory adults and children

Several large prospective studies in ambulatory adult and pediatric patients have clearly shown that antiviral therapy can reduce the duration of symptomatic illness due to influenza by up to 2 days if started within 48 hours of symptom onset.^47,52 In fact, the earlier these drugs are started, the better the clinical outcome. ⁴⁷ Further, starting antiviral therapy early is associated with lower rates of hospitalization, death, and complications requiring antibiotics.⁴⁷ Recent data from Japan also suggest that such early therapy may be partially responsible for the low death rate in that country during the recent pandemic.²⁶

Given the evidence of efficacy, antiviral drugs should be considered in all patients with risk factors for severe disease. Antiviral drugs are also appropriate in patients without specific risk factors because of the risk of progression to severe disease in these patients, especially in the context of pandemic H1N1 influenza.^38,53,54 Further, therapy is associated with symptomatic improvement and reduced infectious complications even in patients without risk factors for severe disease.⁵⁵ Such early therapy may also have a positive impact on secondary infections among contacts.⁵⁵

Antiviral therapy recommended in hospitalized patients with influenza

Clinical studies of the treatment of hospitalized influenza patients are limited, with few prospectively conducted studies. Because of differences in clinical course and viral kinetics in hospitalized patients and emerging data in these patients, the ambulatory treatment data and paradigms likely do not apply to hospitalized adults.^{29,43,44,56–59}

To date, only four prospective, randomized clinical trials have been completed in hospitalized patients with severe influenza, and only one has been published.^60–63 These studies indicate that combination therapy, higher doses, and intravenous therapy may have a role in this unique population.^60–63

Several large observational cohort studies suggested that clinical and virologic outcomes were better in hospitalized patients who received antiviral treatment.^{4,29,31,33,42,56–58,64,65}

For seasonal influenza, antiviral drugs accelerate the decline in viral load, shorten the duration of viral shedding,²⁹ and reduce hospital length of stay⁶⁶ and risk of death.^33,57,67 Their impact appears to be greatest if they are started early, but efficacy was still observed if they were started up to 4 days after illness onset, as viral replication continues longer in hospitalized patients. The benefit may be greater in immunocompromised patients, preventing progression to pneumonia and improving survival.^46,68

In pandemic H1N1 influenza, data suggested that timely antiviral treatment was associated with enhanced viral clearance and improved survival in hospitalized patients. Unfortunately, many patients had a delay before starting antiviral therapy.^4,29,64

Higher-dose oral therapy has been advocated for severely ill patients, although evidence is lacking at the moment. A recently completed study in Southeast Asia shows that prospective studies in adults are needed to document a benefit of such higher-dose therapies before they are widely accepted as standard practice.^4,63 This study found that clinical and virologic outcomes in severely ill patients were no better with oseltamivir in higher doses than in standard doses.⁶³ Whether this study can be generalized to US populations is not clear, since viral dynamics differ by virus type, clinical care (especially referral patterns and timing) may be different in Southeast Asia, and children predominated in this study.

Ongoing studies will, we hope, demonstrate if intravenous therapy (eg, peramivir, zanamivir) is better than oral therapy for such patients. This is especially important, since oral therapy may result in adequate blood levels in many patients.⁵¹

In the United States, many patients with febrile respiratory illnesses were hospitalized and started on antibacterial drugs, but antiviral drugs were not given or initiation of these drugs was delayed.⁶⁴ Influenza should be suspected as a cause of fever or respiratory symptoms, including pneumonia, in any hospitalized patient when influenza is circulating in the community. Antiviral therapy should be started empirically and should not be delayed while awaiting test results.⁶⁴ Further, much like with bacterial pneumonia, testing may be erroneously negative or unavailable until progression has occurred. Therefore, antiviral therapy should be initiated early in any patient in whom influenza is included in the differential diagnosis.

Should a longer course of therapy be considered? Prolonged viral shedding has clearly been documented in patients infected with the pandemic 2009 A (H1N1) virus, and in hospitalized or immunocompromised adults with seasonal influenza.^{29,44,46,68–71} Given the current information and the lack of prospective studies comparing 5 days vs a longer course of therapy, 10 days of therapy has been suggested for patients with severe pandemic H1N1 infection requiring hospitalization (particularly if they are treated with corticosteroids or require intensive care) or who are immunosuppressed. ^4,72 Longer therapy may be necessary and should be guided by virologic monitoring, optimally of the lower respiratory tract if easily accessible.

For hospitalized patients with seasonal influenza virus infection, the optimal duration of treatment has not been established, but a prolonged course seems reasonable for immunocompromised patients.^46,54

For patients who do not have a clinical response or who have a relapsing or prolonged virologic course, isolates should be assessed for emergence of resistance.^54,73

Antiviral resistance

Sporadic cases of resistance to neuraminidase inhibitors were recognized in the 2009 influenza A (H1N1) and avian H5N1 infections, typically in viruses with the H275Y mutation.^4,75 Risk factors for the emergence of resistance are high viral load and prolonged shedding, as is common in children and immunocompromised patients, and exposure to low drug concentrations, such as during the course of prophylactic antiviral therapy.^{45,74,76–78} Clinical evidence suggests that strains with the H275Y mutation are transmissible, can cause disease similar to that of wild type virus, and are resistant to oseltamivir but remain susceptible to zanamivir.^{45,74–76,79}

Tests for resistance are not widely available. When testing is considered, robust testing methods that can detect resistance to a wide range of mutations, not just H275Y, should be used.⁷⁴ If resistance is considered, the patient should be managed in collaboration with a specialist in infectious disease.

Since resistance may be recognized midseason, national health authorities monitor data on resistance and update it for clinicians regularly (see www.cdc.gov/flu/ and www.who.int/csr/disease/influenza/en/).

LESSONS LEARNED AND FUTURE DIRECTIONS

We were very fortunate that the recent pandemic was relatively mild compared with earlier pandemics. Nonetheless, it has provided a number of useful lessons to guide clinical care of patients with influenza and to focus future research efforts.

Vaccination. Both seasonal and pandemic influenza vaccines are safe and offer effective protection. Unfortunately, a vaccine against a pandemic virus is not likely to be available during the first wave of a pandemic. Improved surveillance may identify a potential pandemic threat sooner and allow earlier preparation of vaccines. Novel strategies, such as adjuvants, cell culture instead of eggs, and a wider array of rapidly growing seed strains may allow for faster responses to future pandemics.⁸⁰

Since the overall impact of vaccination may be limited by low vaccination rates in the community and in health care professionals, strategies to improve their vaccination uptake and the benefits of universal vs targeted vaccination warrant further study. The critical role of vaccination is unquestioned, and many groups are now calling for mandatory influenza vaccination of health care workers, with rare exceptions.^81–85 Further, current guidelines recommend influenza vaccination for all people without contraindications 6 months of age and older.⁶

Infection control remains an important intervention in the control of influenza. While there continues to be some disagreement about the relative contribution of aerosols in the transmission of influenza, recent data suggest that N95 respirators offer little advantage over properly worn surgical masks for seasonal influenza.^86,87 Nonetheless, infectious aerosols may be generated during certain clinical procedures, such as resuscitation, intubation, bronchoscopy, sputum suction, high-flow oxygen therapy, and bilevel positive airway pressure ventilation, and most experts would recommend the use of N95 respirators in addition to standard precautions.⁸⁸

Antiviral drugs will continue to play a significant role in the management of influenza, given the inherent limitations of vaccines. Expanded, early use of these agents, particularly in high-risk patients and those requiring hospitalization, may result in improved clinical outcomes. If influenza is suspected in such individuals, antiviral drugs should be started immediately and discontinued only if active infection is ruled out or an alternative diagnosis is established, such as respiratory syncytial virus infection. Since humans are not colonized with influenza, broad empiric use of anti-influenza antiviral drugs is unlikely a major contributor to the emergence of resistance.

The optimal duration and route of delivery of antiviral drugs need to be clarified through prospective controlled studies.

The current pandemic also highlights the need for better antiviral therapies for seriously ill patients. Novel antiviral drugs should be developed to allow for the use of antiviral combinations. Such combinations may reduce the emergence of resistance, as is the case with other viral infections in which resistance emerges quickly with monotherapy, and would improve the ease of selecting therapy if strains of various susceptibility patterns are circulating. The optimal role of antibody-based therapies warrants further study.^89,90

Testing. Since rapid antigen assays have limited sensitivity and since samples obtained from the upper tract may be negative in patients with pneumonia, robust molecular testing strategies are preferred. Sampling of the lower airways is critical to rule out influenza in patients with pneumonia with negative upper tract samples.

The pathogenesis of influenza also needs more study. It is now recognized that both uncontrolled viral replication and hyperactivated cytokine and chemokine responses contribute to disease manifestation of severe influenza infection, and that the degree of severity varies with different viruses (eg, pandemic H1N1 vs highly pathogenic avian H5N1).⁹¹ Understanding the relative effect of antiviral and anti-inflammatory interventions on clinical outcomes may allow more tailored therapy depending on the pathogenesis of future pandemics.

Animal hosts. The current pandemic clearly shows the importance of influenza viruses within animals. Efforts to improve our surveillance of viral disease in a wide range of animal species and studies to understand the pathogenesis and antigenic changes of influenza in animal hosts are critical.⁹²