The limitation of referral level fetal ultrasound examination in the detection of spina bifida in Western Australia, 1990-1991.

Objective: To ascertain babies born with spina bifida that was not detected by prenatal ultrasound examination performed after 16 weeks' gestational age at Western Australian referral centres, 1990-1991.

Design: A retrospective study of the antenatal ultrasound details of those infants born with spina bifida in Western Australia during the 24-month period, 1990 and 1991. Data were collected by interviewing parents, clinically assessing affected individuals and reviewing genetic, clinical and investigative records, and from the Birth Defects Registry of Western Australia.

Setting: Western Australia, which has a relatively high spina bifida birth prevalence of 1 in 1000, has centralised neonatal medical, surgical and genetic services, and a Birth Defects Registry. This enabled us to ascertain all Western Australian neonates with spina bifida for the purposes of this study. There is no universal maternal serum alpha-fetoprotein (MSAFP) screening program and the performance of ultrasonography at referral level is of variable quality.

Participants: Newborns with spina bifida and their parents.

Main Outcome Measures: Ultrasound screening for spina bifida was deemed to have failed when referral to a specialist imaging centre for the specific purpose of detecting anatomical abnormality after 16 weeks' gestational age gave a falsely negative result.

Results: Of the 47 infants born with spina bifida in 1990 and 1991, ultrasound screening at more than 16 weeks' gestational age was documented and was falsely negative in 14. Six of the 14 had a relevant family or medical history for the condition. Five of the lesions were covered and eight of the patients still survive.

Conclusion: Ad-hoc fetal ultrasound examination via existing referral centres had obvious limitations in detecting spina bifida in a population at low risk. MSAFP screening has a well documented role in detecting neural tube defects, as eight to 10 of the 14 lesions missed by the referral ultrasonography would have been ascertained in a program of this nature. The study indicated that adequate pre-screening clinical histories were not sought, thus limiting the antenatal testing options offered to at-risk couples. This study emphasised the importance of a statewide review of the specificity and sensitivity of the anatomical fetal ultrasound examination, in view of the expansion of this procedure and its variable quality depending on operator experience and equipment quality.