Between 50-100 million people died during the 1918 pandemic. Death followed from aggressive secondary bronchopneumonia, influenza-related lung disease and associated cyanosis and cardiac collapse (12). During the 1918 pandemic, there was an unexplained excess influenza mortality in persons 20–40 years of age. This mortality increase may have been due to limited native immunity and / or a vigorous immune response directed against the virus in healthy young persons (12). Today the mortality of the 1918 pandemic would almost certainly be reduced because of the availability of ICUs, antibiotics, and antiviral medications – innovations of the mid-twentieth century; the cost will be a dramatic increase in critical care admissions and length of stay, assuming that this capacity is available (13). Long-stay ICU patients have significantly higher critical care and hospital mortality rates compared to short-stay patients, occupy a disproportionate number of critical care bed-days (14,15), and consume even greater resources (16,17).

Influenza is usually transmitted in aerosols by coughing or sneezing, but can be transmitted through bird droppings, saliva, nasal secretions, faeces and blood, or contact with contaminated surfaces. Influenza spreads globally by seasonal epidemics, resulting in hundreds of thousands of deaths annually, and when a new strain of the virus arises from genetic reassortments of flu viruses from human, avian or swine hosts, more virulent strains can result in millions of deaths from pandemic spread. In the 20th century there were 3 pandemics (1918, 1957 and 1968) which killed tens of millions of people. The 2009 H1N1is derived from a triple recombination of human, avian and swine influenza viruses (19). Most older adults have substantial immunity to H1N1 variants that have circulated during the past century; the early epidemiology of this virus indicates that there may be partial protection from multiple previous influenza infections, despite this being a newly identified virus.

Among critically ill Canadian children, the median age was 5.0 years (range 1 month to 17 years), 54.4% were female and the mean PRISM III score was 9. One or more chronic co-morbid illnesses were observed in 70.2% of patients: lung disease (44%); neurological diseases (19%); immune suppression or immunodeficiency (16%); history of prematurity (9%) and congenital heart disease (7%). Two patients were pregnant adolescents and Aboriginal Canadians comprised 25% of the children. Mechanical ventilation was used in 68% of children admitted to ICU and the median duration of ventilation was 6 days (range 0 to 67). Four children died.

Early laboratory diagnosis of 2009 H1N1 infection ideally requires Reverse Transcriptase-Polymerase Chain Reaction (RT-PCR) methodology. Immunofluorescent techniques and enzyme-linked immunoassays of clinical specimens may lack diagnostic sensitivity (26,27). Viral cultures require up to 1 week for processing. While RT-PCR is the preferred definitive diagnostic technique and has very high sensitivity under ideal circumstances, several Canadian and international outbreak sites have reported imperfect sensitivity of standard nasopharyngeal swab RT-PCR testing in severe cases. These reports have been coupled with others suggesting that tracheal aspirates in intubated patients may increase the diagnostic yield, as may repeat testing within 48-72 hours.

One consistent observation among those who have become critically ill is that of rapidly progressive pneumonitis following presentation and a requirement for mechanical ventilation within 24 hours. Patients may display a relative insensitivity to usual measures of oxygenation assistance with positive end-expiratory pressure (PEEP). Controlled ventilation, with attention to a lung protective strategy (41), in combination with appropriate sedation and judicious use of neuromuscular blockade has been commonly used. The severity of hypoxemia and need for patient-ventilator synchrony, coupled with disease in young previously healthy patients, may translate to a greater than usual amount of sedating medications being used. Avoidance of volume overload (and judicious diuresis) may also be associated with reduced duration of ventilation and length of stay in ICU for most patients with ARDS and this strategy should be attempted for patients with H1N1 (42). Other means of oxygenation assistance that have not been shown to improve mortality in other forms of ARDS, but may improve oxygenation for individual patients have included prone positioning (which can be difficult to perform due to the severity of illness coupled with obesity in many patients), and inhaled nitric oxide (43,44). HFOV is currently being evaluated as a rescue therapy for patients with severe ARDS in randomized controlled trials in many jurisdictions (45) and has been frequently used among patients with H1N1-related refractory hypoxemia. ECMO is an uncommonly used modality of oxygenation assistance in non-transplantation associated acute lung injury in adults. However, a recent study has shown that transfer to a centre experienced with the use of ECMO is associated with improved disability-free survival among patients with ARDS and severe hypoxia (46). Clinicians in Australia similarly recommend consideration of ECMO for recalcitrant H1N1(47) but Canadian experience with ECMO for H1N1-related ARDS is minimal. If patients are unable to be supported in their current environment and transfer to another center with additional modalities (e.g. HFO, ECMO) is considered, consultation with centres experienced with such modalities in order to determine whether patients are likely to benefit, should occur before the patient is too unstable for transfer.

Viral shedding has been shown to be prolonged in hospitalized patients with seasonal influenza (Lee CID) and in 2009 H1N1 infections(58,59). In one study, approximately one-third of patients continued to shed live virus at least 1-week after symptom onset (59). In another, 47% of patients continued to shed virus for 7 days or more; 8% for greater than 10 days following symptom onset, despite therapy with oseltamivir (58). Thus, durations of neuraminidase inhibitor therapy longer than 5 days may be appropriate and many jurisdictions are currently recommending therapy for 10 days for H1N1-related critical illness.

We advocate that these adjunctive therapies ideally be used in the context of a clinical trial.

With respect to infection prevention and control related to mode of ventilatory assistance, there is no overwhelming evidence to promote or deny particular modes. During the SARS outbreak, which was associated with a much greater degree of nosocomial spread than seasonal influenza, there was concern that non-invasive and high frequency oscillatory ventilation may promote excess aerosolization of viral laden particles and place surrounding patients and staff at risk. Limited evidence suggests that the process of endotracheal intubation, especially in an uncontrolled setting, may be associated with increased risk of acquiring infection; however, this risk is mitigated extensively if adequate personal protective equipment is worn (77,78). Patients with H1N1-related ARDS are unlikely to have a quickly reversible lung injury, and prior studies suggest that non-invasive ventilation is unlikely to provide definitive ventilation support and obviate the need for invasive support for these patients (79). This, coupled with the potential risk of increased aerosolization, leads us to recommend against prolonged non-invasive ventilation. Although HFOV may also generate substantial aerosol, this concern may be mitigated by rapid piston shut-down with disconnections and loss of pressure in the circuit. Most HFO circuits can also be equipped with bacterial viral filters and a scavenger system to the exhalation port that further mitigates risk.