- Estelle Naumburg, paediatrician ()a,
- Rino Bellocco, biostatisticianb,
- Sven Cnattingius, professorb,
- Per Hall, associate professorb,
- Anders Ekbom, professorb
- a Department of Women's and Children's Health, Section for Paediatrics, Uppsala University, S-751 85 Uppsala, Sweden
- b Department of Medical Epidemiology, Box 281, S-171 77 Stockholm, Sweden
- Correspondence to: E Naumburg
- Accepted 1 November 1999
Obstetric ultrasound examination is part of routine antenatal care and is regarded as safe for both the fetus and the mother. In vitro, however, ultrasound has been shown to cause membrane changes that could affect embryogenesis and late prenatal and postnatal development.1 Studies have also shown an association between exposure to ultrasound and an increased frequency of non-righthandness, indicating that fetal development may be affected by the ultrasonic waves.2
Concerns over a possible association between exposure to ultrasound in utero and an increased risk of childhood malignancies have not been substantiated, but previous studies have been hampered by low statistical power or based on interviews with the parents done retrospectively, or both.3–5
To assess the impact of ultrasound and the risks of childhood lymphatic and myeloid leukaemia, we performed a nationwide population based case-control study using prospectively assembled data on prenatal exposure to ultrasound
Subjects, methods, and results
The cases in this study comprised all children born and diagnosed as having leukaemia between 1973 and 1989 and reported to the nationwide Swedish registers of birth, cancer, and causes of death—in all, 752 cases. One control was randomly selected for each child with leukaemia from the Swedish Birth Registry and matched by sex and year and month of birth. The study was restricted to cases and controls without Down's syndrome (n=731), and medical records of 652 (89%) matched case-control pairs could be retrieved (578 cases with lymphatic leukaemia and 74 with myeloid leukaemia).
Altogether, 361 (48%) of the children with leukaemia had developed it before the age of 4, and 21 children were born in twin pregnancies. Information on exposure was extracted from antenatal, obstetric, and other standardised medical records by one of us (EN), who was blind to whether the child was a case or control. Conditional logistic regression was performed to study the association between prenatal exposure to ultrasound and childhood leukaemia (lymphatic and myeloid leukaemia). Maximum likelihood methods were used to estimate the odds ratio and 95% confidence intervals.
In all, 200 children with lymphatic leukaemia and 214 controls had been exposed prenatally to ultrasound (odds ratio 0.85; 95% confidence interval 0.62 to 1.17) (table). The risk of lymphatic leukaemia was not influenced by either the number of ultrasound examinations or when the examination was performed (data not shown). Twenty nine of the children with myeloid leukaemia and 27 of the controls had been exposed to ultrasound prenatally (odds ratio 1.0; 0.42 to 2.40) (table). The risk of myeloid leukemia was not influenced by the number of ultrasound examinations (table). A slightly higher, but not significant, risk was seen for those examined during the second trimester (odds ratio 1.42; 0.88 to 2.29). Adjustments for potential confounding, such as maternal age, high birth weight, and twin pregnancies, did not alter the results (data not shown).
We could not detect any association between exposure to ultrasound during pregnancy and lymphatic or myeloid leukaemia, and the results of the study are therefore reassuring. The strengths of the study are its size, the exclusion of children with Down's syndrome, and the use of prospectively assembled exposure data. Ultrasound examination was gradually introduced in Sweden during the study period, and the proportion of exposed fetuses (36%) is therefore appropriate; any possible underestimation of exposure should be similar in both cases and controls.
We conclude that single or repeated intrauterine exposure to ultrasound, early or late in the pregnancy, does not influence the risk of subsequent development of lymphatic or myeloid childhood leukaemia.
Contributors: EN coordinated the study and assembled the data. SC and AE were responsible for initiating the study and the original study design. RB was responsible for the statisitical work. PH contributed to the interpretation of the data. The paper was written jointly by all authors, with EN as lead author. AE is guarantor for the paper.
Funding This study was supported by grant 3520-B94-01XAB from the Swedish Cancer Fund and grant SSI P 959.96 from the Swedish Radiation Protection Agency.
Competing interests None declared