Article Text

  1. B Contoli1,
  2. F Franco6,
  3. D Di Lallo6,
  4. L Sirianni2,
  5. R Adorisio3,
  6. D Di Carlo4,
  7. F Piersigilli5,
  8. G Seganti5,
  9. M Cuttini1
  1. 1Unit of Epidemiology, Paediatric Hospital Bambino Gesù, Rome, Italy
  2. 2Medical Direction, Paediatric Hospital Bambino Gesù, Rome, Italy
  3. 3Department of Cardiology, Paediatric Hospital Bambino Gesù, Rome, Italy
  4. 4Department of Cardiac Surgery, Paediatric Hospital Bambino Gesù, Rome, Italy
  5. 5Department of Neonatology, Paediatric Hospital Bambino Gesù, Rome, Italy
  6. 6Regional Health Agency, Rome, Italy


Objective To assess survival in a cohort of infants with the primary diagnosis of congenital heart disease admitted to Bambino Gesù Paediatric Hospital (Rome, Italy) in the first six months of life (years 2001–2006). Only residents of the Lazio region were included.

Methods Deterministic record linkage between hospital discharge records and the Regional Death Registry was carried out. Survival and 95% confidence intervals (CI) were estimated according to the Kaplan-Meier method. Log-rank test was used to compare survival curves.

Results 877 infants (55% males) met the inclusion criteria. Median age at first admission to Bambino Gesù hospital was 4 days. The most frequent diagnosis were ventricular septal defect (108, 12,3%); aortic coarctation (89, 10.1%); and tetralogy of Fallot (TOF) (84, 9.6%). 68 patients had transposition of the great arteries (TGA), and 47 hypoplastic left heart syndrome (HLHS). 128 deaths were identified: 112 occurred in the Bambino Gesù hospital; 14 in other hospitals; and 2 at home. The overall survival rate was 90% (CI 88 to 92) at 3 months of age; 88% (CI 86 to 90) at 6 months; and 86% (CI 83 to 88) at 1 year. There were significant differences according to diagnosis: 1 year survival was 94% (CI 87 to 98) for aortic coartation, 88% (CI 78 to 93) for TOF; 76% (CI 64 to 85) for TGA; and 44% (CI 30 to 58) for HLHS.

Conclusion Linkage between different databases can contribute significantly to epidemiological follow-up and prognostic assessment. In this study it allowed us to identify deaths that occurred outside our hospital and obtain valid survival estimates.

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.