NEONATAL SOCIETY ABSTRACTS
Odds ratios for respiratory syncytial virus (RSV) rehospitalisation in the first 2 years of life in premature infants with and without neonatal respiratory illness
Presented at the Neonatal Society 2002 Summer Meeting (programme).
Thwaites R1, Piercy J2, Ryan J2, James P3
1 Child Health, St. Mary's Hospital, Portsmouth, PO3 6AD, UK
2 Mapi Values, Macclesfield, SK10 5JB, UK
3 CHKS, Arden Rd, Alcester, B49 6HN, UK
Hypothesis: Premature birth and neonatal respiratory illness are both associated with higher rate and duration of RSV rehospitalisation in infancy.
Introduction: RSV infection causes significant problems for ex-premature infants particularly those with Bronchopulmonary Dysplasia (1). We used the CHKS database, which covers 55% of the UK, to quantify the odds ratios (OR) and lengths of stay (LOS) for RSV hospitalisations in term and preterm infants with and without neonatal respiratory problems. Analysis was restricted to hospitals with complete data for the whole time period (equivalent to about 30% of UK births).
Methods: Admissions of children less than 2 years old between Jan 1998 and Dec 2001 were analysed. Data was extracted on episodes with RSV specific ICD-10 codes as discharge diagnosis (n=14,728). Where possible such admissions were linked back to a birth admission record (n=8,597). All births where the baby was discharged alive (n=694,503) for the same 4 year period were assessed for 'risk factors' using ICD-10 codes for gestation and neonatal respiratory illnesses.
OR RSV admission: Gps 1 v 2: p=0.001, 3 v 4: p=0.001; 1, 2, 3, 4 and 5 v 6 all p<0.001
LOS: Gps 1 v 2: p=0.576, 3 v 4: p=0.536, 1 v 6: p=0.08. 2, 3, 4 and 5 v 6: all p<0.001
Discussion: Absolute rates of RSV admission shown are significant underestimates as only 58% could be linked back to a birth record, furthermore we only include those babies with confirmed RSV, coded as such by CHKS. Adjusting for unlinked records (assuming these occurred at the same rate in each group) confirmed RSV admission rates for groups 1-6 would rise to 18.9, 4.8, 5.8, 4.6, 3.6 & 1.9% respectively.
The term group 5 with neonatal respiratory illness and later RSV merits further investigation and may provide new insights into RSV pathogenesis.
Acknowledgements: This study was supported by an educational grant from Abbott Laboratories, Maidenhead, UK
1. Groothuis JR, Gutierrez KM & Lauer BA. Pediatrics 1988;82:199-203