Since the emergence of the novel H1N1 influenza 09 (human swine influenza) in North America and Mexico in mid April, the number of confirmed cases has increased to over 55 000 across 105 countries.1 While most individuals will experience a mild clinical illness (coryza, fever, cough and myalgias), there have been 238 reported deaths (0.4%).2 A Centers for Disease Control and Prevention (CDC) report in May provided details of the 30 patients who were hospitalised in California, of whom six required admission to an ICU and four required MV.3

At the time of writing, there have been 3912 confirmed cases of H1N1 influenza 09 in Australia.2 Most reported illness has been mild, but 268 patients (6.9%) have been hospitalised, over 25 (0.6%) have been admitted to ICUs, and five deaths (0.1%) have been reported. Here, we presented the first six cases of H1N1 influenza 09 (confirmed by PCR test from nasopharyngeal swabs) in which patients required admission to intensive care in Australia. Admissions were to three hospitals in the north-western suburbs of Melbourne. These cases highlight the small but significant risk of life-threatening respiratory failure associated with H1N1 influenza 09 infection. All patients experienced a rapid (but reversible) decline in respiratory function, with most requiring complex respiratory support. The age distribution of these patients is consistent with other reports, and lower than that seen in previous influenza seasons.4,5

Four of the six cases we report had risk factors, including asthma, chronic lung disease, smoking, obesity, and pregnancy; these risk factors were similar to those identified in the CDC reports.3 Patients 2 and 4 had no identifiable risk factors.

There are several possible explanations for the acute respiratory failure observed in these patients. Early onset of respiratory failure with widespread pulmonary infiltrates (Patient 3) suggests primary viral pneumonitis, whereas the delayed onset of fever with lobar signs (Patient 2) and pleurisy (Patient 1) suggest secondary bacterial pneumonia. A cytokine-mediated acute lung injury may also explain the late appearance of diffuse pulmonary infiltrates (Patients 4 and 6). The absence of a severe prodrome in Patient 5 suggests an exacerbation of the patient’s COPD.

With the number of cases of H1N1 influenza 09 infection likely to increase, it is anticipated that further cases of severe respiratory failure associated with this influenza will be seen. Based on the cases reported here and other reports, we offer the recommendations shown in Box 2. A high index of suspicion that respiratory failure may ensue is warranted in patients who have risk factors5 or present with tachypnoea (respiratory rate, > 24 breaths/min) and/or hypoxia (Spo2, < 95% on supplemental oxygen), and early referral to hospital is warranted. Youth and prior good health do not preclude the possibility of severe respiratory failure.

The Victorian Department of Human Services currently recommends nasopharyngeal swabs for a PCR test for influenza A in patients admitted to hospital with suspected influenza. Empiric therapy with antiviral agents (oseltamivir or zanamivir)5 should be considered in addition to antibiotic treatment for community-acquired pneumonia pathogens, in consultation with an infectious diseases specialist. Patients with suspected H1N1 infection should be isolated, preferably in a negative pressure isolation room.6,7 Where possible, antiviral filters applied to the expiratory limb of the ventilator circuit may further reduce the risk to health care staff.

Oseltamivir (Tamiflu) and zanamivir (Relenza) reduce viral replication and shedding, and may reduce the risk of more severe illness. Their safety in pregnancy has not been investigated (Category B1 for use in pregnancy), but without treatment there may be a greater risk of premature labour (Patient 6).5,8 Increasing resistance to oseltamivir has been reported in other strains of currently circulating influenza A viruses, but, as yet, not in the H1N1 influenza 09 lineage.9,10

Any patient with respiratory distress or severe hypoxia (requiring an Fio2 of > 0.5) and pulmonary infiltrates on chest x-ray should be referred to an intensive care specialist for further assessment. Mechanical ventilation for these patients is complex, and requires expertise and specialised equipment. We used restrictive tidal volumes (6 mL/kg ideal body weight), high PEEP (15–20 cm H2O), pressure-limited modes of ventilation, alveolar recruitment manoeuvres, inhaled nitric oxide, and restrictive fluid therapy with apparent success. This is consistent with ventilation strategies used by others,11 and in keeping with strategies described by the Acute Respiratory Distress Syndrome (ARDS) Clinical Network.12 Extracorporeal oxygenation therapy has recently been used in other cases (G J D, personal communication). Based on the available data, we would not recommend NIV as the mainstay of respiratory support. The four patients who were given a trial of NIV in this series all required intubation and MV. This is consistent with published data for ARDS and pneumonia.13 NIV temporarily improves oxygenation and reduces the work of breathing, but does not necessarily alter the course of the disease.14 The need for NIV is an indication of severe disease and the likelihood of intubation and MV.

For the most part, H1N1 influenza 09 is a benign disease, but it may lead to severe respiratory complications in a small proportion of patients. In our series, prompt diagnosis and intensive therapy was associated with favourable outcomes.