Diagnosis and morbidity of placenta accreta.

OBJECTIVE To examine the diagnostic precision of ultrasound examination for placenta accreta in women with placenta previa and to compare the morbidity associated with accreta to that of previa alone. METHODS This was a retrospective cohort study of all women with previa with/without accreta examined at the University of California, San Francisco (UCSF) between 2002 and 2008. The sensitivity, specificity, negative predictive value (NPV) and positive predictive value (PPV) of ultrasound examination for the diagnosis of accreta were calculated and compared with results from similar studies in the literature. Univariable analysis was used to compare clinical outcomes. RESULTS The PPV of an ultrasound diagnosis of accreta was 68% and NPV was 98%. Ultrasound had a sensitivity of 89.5%. Compared with previa alone, accreta had an odds ratio (OR) of 89.6 (95% CI, 19.44-412.95) for estimated blood loss > 2 L, an OR of 29.6 (95% CI, 8.20-107.00) for transfusion and an OR of 8.52 (95% CI, 2.58-28.11) for length of hospital stay > 4 days. CONCLUSION Placenta accreta is associated with greater morbidity than is placenta previa alone. Ultrasound examination is a good diagnostic test for accreta in women with placenta previa. This is consistent with most other studies in the literature.