Objective To investigate the incidence of associated anomalies and document changes and progress in the management and outcome of intrinsic duodenal obstruction (atresia or stenosis) in a large series over a long time period with a view to providing comprehensive data for prenatal counselling.Design A retrospective casenote review.Setting A quaternary referral centre for paediatxic and neonatal surgery.Population l b o hundred and seventy-five infants born with duodenal obstruction between 1951 and 1995.Methods For analysis of management and outcome data, the series was divided into three groups, each admitted over 15 yearperiods: Group A, 1951-1965 Group B, 1966 Group B, -1980; Group C, 1981-1995. Main outcome measures Primary: associated anomalies, complication rates and survival. Secondary: age at diagnosis, duration of hospital stay.Results There were 136 males and 139 females. Seventeen of 30 cases (57%) presenting between 1991 and 1995 were diagnosed prenatally on ultrasound scan. The median age at diagnosis for atresia was 3-5 days (Group A); 2.2 days (Group B) and 1.8 days (Group C). The median age at diagnosis for incomplete obstruction was five days. Associated anomalies included Down's syndrome (n = 82, 30%); Down's plus cardiac malformation (n = 38, 14%); isolated cardiac (n = 64,23%); and gastrointestinal problems (n = 116,42%). Overall complication rates fell from 51% (Group A) to 18% (Group C ) and survival increased from 5 1 % to 95% . Median hospital stay also decreased from 35 days to 18 days.Conclusions These data c o n f m a progressive improvement in the outcome of intrinsic duodenal obstruction over the past 45 years. It is important to note that they only relate to infants born with duodenal atresia and do not take into account possible 'hidden' mortality resulting from spontaneous abortion or termination.