Preventing respiratory syncitial virus bronchiolitis

Except in very high risk infants there is no cost effective prophylatic agent

Winter in the United Kingdom—wet, cold, miserable, and, yet again, the season for respiratory syncitial virus (RSV) bronchiolitis. About 3% of each year's birth cohort are admitted with bronchiolitis every winter in Europe, Australasia, and North America (20 000 infants in the UK, of whom 600 need ventilation1). Traditionally certain groups of infants are considered to be at high risk of developing more severe RSV bronchiolitis. These high risk groups include infants born prematurely (insufficient transfer of maternal RSV IgG) and those with chronic lung disease of prematurity, other underlying cardiorespiratory disease, or immunodeficiency. However the great majority of infants admitted are previously normal babies. The treatment of RSV bronchiolitis has had a chequered history, and, despite initial enthusiasm, it is now widely accepted that bronchodilators, steroids, and ribavirin have no overall significant benefit.2 This therapeutic nihilism makes paediatricians uneasy, and if we have no treatment, then surely prevention must be the answer.

Pooled hyperimmune RSV intravenous immunoglobulin (RSV IVIG, Respigam) was licensed by the Food and Drug Admininstration in 1996 after the PREVENT study.3 Monthly prophylaxis over the RSV season with RSV IVIG led to an overall reduction of 41% in admissions for RSV bronchiolitis in high risk groups. However, RSV IVIG required regular intravenous infusions of a high volume and protein load from pooled donors, with the risk of transmission of blood born pathogens. A Cochrane review of RSV IVIG is available.4

Palivizumab (Synagis) is a recombinant humanised mouse monoclonal antibody to the RSV F protein. It is a neutralising antibody that prevents RSV fusing with the cell membrane and can be given intramuscularly. The IMpact study was a multicentre randomised double blind placebo controlled trial of palivizumab. Infants born premature (<36 weeks' gestation) or with chronic lung disease of prematurity were randomised to receive either five monthly injections of placebo (n=500) or palivizumab (n=1002) over the RSV season. The primary end point was admission with RSV disease. The study showed a relative reduction in RSV related admissions of 55% (10.6% placebo, 4.8% palivizumab, p=0.0004).5 Adverse events were the same in both study arms. The study was not powered to detect reductions in mortality. There was no significant reduction in prolonged admission (>14 days) or the number of days spent on a ventilator between the two groups.

Palivizumab is safe and certainly works, so should we use it? It has been licensed in the US, and the American Academy of Pediatrics suggests that palivizumab should be considered for infants either born prematurely or treated for chronic lung disease within six months of the RSV season.6 Unfortunately, palivizumab is also very expensive.

The IMpact trial was not designed as a pharmacoeconomic study. When introducing a new preventive therapy clinicians need to consider not only the existing morbidity and mortality of the disease but also the efficacy and cost effectiveness of the prophylactic agent. We have recently summarised the incidence of readmission due to RSV disease noted in observational studies from North America and the UK.7 Broadly similar readmission rates for RSV bronchiolitis were noted, of about 6-8 % for infants born <32 weeks' gestation and 12-17% for infants with chronic lung disease. Even in these high risk groups, mortality from RSV bronchiolitis is now extremely low, 0.13% in the IMpact study.

Several cost effectiveness studies have been performed. In the IMpact study the absolute risk reduction for the whole study group was 5.8%, giving a number needed to treat—that is, to prevent one hospital admission—of 17.2, with an expenditure of £25 500 (95%confidence interval £16 500 to £49 500) to prevent one hospital admission.8 This type of analysis has been criticised, mainly because the admission rate among the placebo treated controls in the IMpact study was lower than previously noted. However, the broad agreement of the recent observational studies suggests that the number needed to treat calculations are reasonable, and possibly an underestimate. Other cost effectiveness studies have given similar results.9^–12

Although these analyses do not take into consideration the increased incidence in wheezing during childhood after RSV bronchiolitis, it is unlikely that these extra costs will be significant. The only group of infants in whom the cost of admission was similar to the cost of palivizumab was those with severe chronic lung disease.11 It is in this very small group of infants at the highest risk where more data on the morbidity and mortality of RSV bronchiolitis, and the role of palivizumab, are needed. Either a prospective study or detailed postmarketing surveillance of infants who do and do not receive prophylaxis is required.

In the meantime palivizumab has no role to play in the vast majority of infants admitted with RSV bronchiolitis. Prevention may in future be achieved with the development of a safe, effective vaccine. Bronchiolitis represents an abnormal host immune response to RSV infection.13 Further characterisation of this response and its genetic basis could lead to identification and targeting of infants most at risk of developing severe RSV bronchiolitis.