Methods

Children who matched the population studied by the Cardiac Synagis Group with haemodynamically significant lesions that would require surgery were identified from the database of the cardiology department of the hospital. Infants under one year of age were identified, rather than under two years as in the study by Feltes et al.,⁴ because from local data, supported by the findings of Feltes et al.,⁴ and the recent recommendation from the British Paediatric Cardiac Association,¹¹ this is the group of patients deemed to be at high risk.

We identified the number of patients under the care of the paediatric cardiologists admitted to Alder Hey with laboratory confirmed respiratory syncytial viral bronchiolitis over the last three seasons from hospital coding records, the database of the cardiology department, and the database of the paediatric intensive care unit.

We obtained the costs per day for an admission to the cardiology ward, and to the paediatric intensive care unit, from the finance department of the Trust, while the costs for inter-hospital transfers were obtained from Unit Costs of Health and Social Care published for 2002.¹² The dosage and cost of palivizumab were obtained from the British National Formulary.¹³

Results

The mean number of patients who meet the criterions for being at high risk is 131 each year. This number includes infants under the care of the paediatric cardiologists at Alder Hey who live outside the Liverpool area, but within the Mersey region or in the North West region.

Over the last three seasons, 39 of these infants deemed to be at high risk have been admitted with bronchiolitis known to be caused by the respiratory syncytial virus. This represents a rate of hospitalisation of 10 per cent, a number which is the same as found in the placebo arm of the study made by the Cardiac Synagis Group.⁴ This does not include infections due to the respiratory syncytial virus acquired in the hospital, which number approximately 10 per year.

Palivizumab is supplied in vials of 50 milligrams or 100 milligrams. A 100 milligram vial costs 706 pounds sterling, equating to 1360 American dollars or 995 euros at the time of writing. By cohorting patients in small groups, it is possible to share vials. Estimating a mean weight of 5 kilograms, and using a monthly dose of 15 milligrams per kilogram for five doses, the cost per patient is 2,650 pounds, equal to 5,100 dollars or 3,700 euros, for the whole season. To treat the 131 patients seen at Alder Hey, therefore, would cost 346,800 pounds sterling, 669,000 American dollars, or 489,000 euros each year.

The Cardiac Synagis Group study showed a reduction of 45 per cent in admissions to hospital.⁴ Applying this reduction to our population would produce an expected reduction in patients hospitalised from 13 to 7 per year. They showed a relative reduction of 56 per cent for length of stay in the hospital wards, and 78 per cent for length of stay in intensive care. This would potentially reduce our total length of stay on the wards from 169 to 76 days, and in intensive care from 93 to 21 days. Admission to our cardiology ward costs 852 pounds sterling, equivalent to 1,640 American dollars or 1,200 euros, each day, and to the intensive care unit 1,550 pounds sterling, or 2,990 American dollars or 2,180 euros. This gives potential savings of 190,800 pounds sterling, equivalent to 368,200 American dollars or 269,000 euros, each year.

There are other potential savings and benefits to the patients and the Trust. The risk of nosocomial infection by the respiratory syncytial virus would be reduced if fewer children were admitted with active infection. Fewer intensive care beds would be occupied with patients having congenital cardiac malformations and infected by the respiratory syncytial virus. As beds in paediatric intensive care remain a relatively scarce resource within the United Kingdom, this should reduce the number of cancelled elective operations and refused admissions. Over the last season at Alder Hey Hospital, 31 elective operations were cancelled, and 58 referrals from within the Mersey region, and 10 from outside the region, were refused because no bed was available.

At present, there are a limited number of services for paediatric retrieval. Those that exist often do not transfer patients from a district general hospital to another paediatric intensive care unit. This means that the staff of the district general hospital has to find a “distant” paediatric intensive care bed, and transfer the patient. This creates a further financial cost, and requires a team from either the accepting intensive care unit, or the referring hospital, to be off-site for several hours. An average inter-hospital transfer takes around three hours. The cost of an ambulance for 58 such transfers is approximately 50,000 pounds, equivalent at the time of writing to 96,500 American dollars, or 70,500 euros. This does not include the costs of the staff at the district general hospital. In addition, there is good evidence to support retrieval by a dedicated retrieval team.¹⁴ If such a reduction in hospitalisation because of infection by the respiratory syncytial virus occurred in our population deemed to be at risk, then this situation could be improved.

Overall, therefore, the additional cost of providing palivizumab for these 131 patients would be 106,000 pounds sterling, equivalent using the current exchange rates to 204,500 American dollars, or 149,500 euros. This amounts to 800 pounds sterling, or 1,540 American dollars, or 1,125 euros, for each patient treated, or alternatively 17,700 pounds sterling, 34,100 American dollars, or 25,000 euros, for each admission avoided.

Discussion

We have studied the potential impact of introducing palivizumab as prophylaxis against infection by the respiratory syncytial virus in the population of children with congenital cardiac disease serviced by a regional centre in the North West of England. We estimate this will produce an additional cost of 106,000 pounds sterling, equivalent with the exchange rates at the time of writing to 204,500 American dollars, or 149,500 euros, each year. This may be an overestimate of the cost, as it does not include patients with respiratory syncytial viral bronchiolitis admitted to other institutions, either their local district general hospitals or to the paediatric units in Manchester, our largest adjacent city. There would be additional savings by reducing these admissions. The effect on length of stay uses the mean values derived from the data of the Cardiac Synagis Study Group.⁴ The mean was used rather than the median, as it was felt that reducing the relatively infrequent but long in-patient stays could have a large impact on bed occupancy and financial costs.

Costs could be reduced by more restrictive criterions for the selection of patients, for example by limiting use to infants under the age of six months, or according to the severity of the congenital cardiac malformation. It is difficult, however, to identify this subset from our data. Feltes et al.⁴ showed a greater reduction in hospitalisation in the three-fifths of the their patients with acyanotic lesions, than in the three-tenths with cyanotic disease. Even their study, however, was not powered for such sub-group analysis. Recent recommendations from the British Paediatric Cardiac Association¹¹ suggest including children with increased pulmonary blood flow, cyanotic disease, pulmonary hypertension, or those with unoperated or partially corrected complex malformations, among other indications.

Other cost-saving regimes have been proposed. For example, current practice in Glasgow is to start prophylaxis with palivizumab when respiratory syncytial viral disease starts to appear in the local community, and to stop the injections when the season has been deemed locally to have ended (personal communication, Dr Coutts). This has reduced the total number of injections given. Patients who undergo corrective surgery during the season may be able to stop the injections post-operatively. Similarly, those who are born and have early corrective surgery before the season began would not need prophylaxis.

There would, of course, be additional costs involved in setting up this service, as well as additional savings and benefits. The exact costs would depend on where the clinics for administration of the prophylaxis were held, that is all at Alder Hey Hospital, or some in the referring hospitals, who would speak to and gain consent from the parents, and who would administer the injections. There would also be costs for the training and education of staff. Patients may have to attend an out-patient clinic monthly during the season. Alternatively, private enterprise could be used as a service to deliver the drugs. The potential reduction in cancelled operations similarly cannot be easily quantified. Nor can the psychological and emotional cost of infants acquiring respiratory syncytial viral bronchiolitis, becoming hospitalised, requiring intensive care, and possibly even needing transfer to a more distant hospital for this care.

The rate of hospitalisation at Alder Hey, at 10 per cent, was similar to that identified an international study,⁴ and to figures reported previously from the United States of America.³ Recent data from Switzerland¹⁵ showed a fourfold lower rate of hospitalisation. This is likely to be a true difference, as these are reliable data from a prospective study at a single site, comparing similar patients over a six year period, and eliminating year to year fluctuation. Small family size, and low rates of attendance for day care in Switzerland, may reduce exposure to the respiratory syncytial virus.¹⁵ Access to healthcare, and the criterions for admission to hospital, may also differ between populations. Local policy-making on the use of palivizumab must consider such variation.

The costs of using palivizumab are high, but are comparable with other new biologic therapies. Etanercept is a monoclonal soluble tumour necrosis factor receptor fusion protein, and is used for treatment of refractory juvenile arthritis. It costs approximately 10,000 pounds sterling, or 19,300 American dollars or 14,100 euros, for each patient in each year. There are currently 35 children in the Mersey and North West regions receiving etanercept, at a cost of 350,000 pounds sterling, equivalent to 675,500 American dollars or 493,500 euros each year. It is unrealistic to expect costs of biologic therapies such as palivizumab to fall, at least in the short- to medium-term, until the pharmaceutical companies have recovered the costs of research and development.

Safety concerns have been raised regarding palivizumab. In particular, Moore et al.¹⁶ reported that, over a three year period, palivizumab was associated with 113 deaths in children under two years of age, more than any other drug. There was no suggestion of any causal relationship, and surveillance since its introduction has shown that the mortality rate among infants who received palivizumab is not higher than for those who did not.¹⁷ Among these children co-morbidity was high.¹⁸ Two-thirds had at least one congenital anomaly, over two-fifths had multiple anomalies, and around one-third was extremely premature, being born at less than 28 weeks gestation. A cause of death was reported for two-thirds of cases, with almost two-fifths dying from their congenital anomalies, and almost one quarter from respiratory infections.¹⁸ Side-effects are rare and generally mild. In the study made by the Cardiac Synagis Group, there were no serious adverse events judged to be related to the study drug.⁸

Much is made of the cost of palivizumab in recent publications. A letter from Onuzo¹⁹ raised several issues. Questions relating to the validity of the recommendations were addressed in the reply from the original authors, but both sets of correspondents called for further evaluation of cost-benefit, as did Handforth et al.²⁰ in an editorial also published last year in the British Medical Journal. Using a number of 22 needed to treat culled from the date of Feltes et al.,⁴ Onuzo¹⁹ predicts a cost of 90,000 pounds sterling, equivalent to 173,700 American dollars or 127,000 euros, to prevent one admission to hospital. This did not take into account any of the potential cost-saving measures we have discussed, nor weigh this against the potential financial, organisational or psychological savings.

Yount and Mahle²¹ have explored the costs of using palivizumab for a hypothetical cohort of patients in the United States of America, and conclude that the cost of palivizumab for the United States could exceed 30 million American dollars, equivalent to 15.5 million pounds sterling or 21.9 million euros with the exchange rates extant at the time of writing. There are a number of differences between their analysis and ours. They have included time taken for health professionals to administer the injections, and for families to attend for injections and accompany their children when hospitalised. The staff costs and distances travelled are higher than would be incurred in the United Kingdom²² and other, potentially quicker, cheaper and/or more acceptable²³ methods of administration are not considered. Yount and Mahle²¹ also consider many important long term outcomes, and make estimates of quality-adjusted life years saved. They do not calculate other financial or non-financial costs of hospitalisation, such as nosocomial infection, lack of intensive care beds leading to transfer to distant units, cancellation of surgery, or the emotional implications of hospitalisation. Handforth et al.²⁰ also question the value of palivizumab in infants with cardiac disease in view of its high cost. They also discuss the problems with the development of a vaccine. It is forty years since the formalin inactivated vaccine against respiratory syncytial virus was first used, and they suggest that routine vaccination is unlikely this decade. There are, however, currently no other obvious therapies that have the potential to reduce admissions to hospital and intensive care during the winter months when beds are at their most scarce.

In conclusion, we believe that the additional financial cost of providing prophylaxis with palivizumab against bronchiolitis due to the respiratory syncytial virus to all patients with congenital cardiac disease under the age of one year of age with haemodynamically significant lesions, and cared for at the Royal Liverpool Chidlren's Hospital, would be 106,000 pounds sterling, 204,500 American dollars, or 149,500 euros per year. This equates to 800 pounds sterling, or 1,540 American dollars, or 1,125 euros for each patient treated, or 17,700 pounds sterling, 34,100 American dollars, or 25,000 euros for each admission avoided. Whether or not such a cost is acceptable is currently decided by individual Trusts responsible for managing the care of children with congenital cardiac disease in the United Kingdom. It is now accepted that rationing is inevitable in the National Health Service of the United Kingdom.²⁴ Such decisions must be acceptable economically, sociologically and legally, and must also be open to and acceptable to consumers.