Background The

overall worldwide birth prevalence of gastroschisis has risen from 0.29 per 10 000 births in 1974 to 1.66 per 10 000 births in 1998.1 This increase has been observed since the 1970s but began at the end of the 1980s in several areas.1-5 There is a wide regional variation in the birth prevalence of gastroschisis, with the incidence in Japan being 0.467 per 10 000 births for 1996- 1997,2 compared with 4.4 per 10 000 births in South East England in 1995.3 The incidence of the macroscopically similar exomphalos (omphalocele) however, has remained stable or decreased.4 The causation of gastroschisis remains uncertain, although the interruption of the fetal omphalomesenteric arterial sup-ply has been suggested as the pathophysiology.6 Teenage mothers have been shown to be at significantly higher risk of having a child with gastroschisis, suggesting a lifestyle expo-sure to an environmental teratogen. Maternal use of recre-ational drugs, including cocaine, marijuana, amphetamines and alcohol, as well as maternal smoking, increases the risk of gastroschisis.4 However, no significant changes in the rates of maternal drug use, tobacco smoking or alcohol abuse, or of conception while taking the oral contraceptive pill, have been found to correlate with the sudden and rapid increase in the incidence of gastroschisis.3 The causation of gastroschisis is probably multifactorial.4 The sporadic incidence and low familial recurrence rate make a genetic cause unlikely7 although an interaction between a specific teratogen and an environmental or genetic predisposi-tion has been suggested. The rapid rate of increase in inci-dence and the wide regional variation suggest an environmental or nutritional cause.1,3 A teratogen with a high exposure rate and low efficacy is likely, as no unusual poten-tial teratogens have been identified in case history or case control studies.7 A link to influenza infection during the first trimester of pregnancy, with an unusually high rate of chil-dren born with gastroschisis in the first quarter of the year in England, has been proposed.3 Method Data from the ward admissions registers of the paediatric sur-gical wards of Pretoria Academic Hospital (PAH) and Kalafong Hospital (KH), and from the discharge records and the weekly statistics sheets and audits thereof of the Paediatri