To estimate, in maternal red blood cell alloimmunization, the diagnostic value of fetal ultrasonography and Doppler blood flow velocity in the evaluation and prediction of fetal anemia.
Literature from 1970 to 2000 was identified using general bibliographic databases (MEDLINE and EMBASE), the Cochrane Library and relevant specialist register of the Cochrane Collaboration, and by checking reference lists of known primary and review articles. Studies were selected if the accuracy of the fetal ultrasound parameters or Doppler studies of blood flow in the fetal vessels was estimated compared with a reference standard (fetal hemoglobin). The diagnostic tests evaluated were ultrasound measurement of the fetal spleen perimeter and Doppler studies of blood velocity estimates in the umbilical vein, ductus venosus, middle cerebral artery, thoracic aorta, and umbilical vessel combined with the thoracic aorta. Study selection, quality assessment, and data abstraction were performed independently and in duplicate. Data from the selected studies were abstracted as 2 × 2 tables comparing the diagnostic test result with the reference standard. Diagnostic accuracy was expressed as likelihood ratios.
The review included eight primary studies with 362 pregnancies affected by red cell alloimmunization. Prospective patient recruitment and complete population details were reported in half of the selected studies (four of eight). Only one study reported masking the diagnostic test results to clinicians. The diagnostic test performance varied widely according to the type of the test evaluated and the cutoff level used to define fetal anemia, which varied from study to study. The diagnostic test study of highest methodological quality reported a positive likelihood ratio of 8.45 (95% confidence interval 4.69, 15.56) and negative likelihood ratio of 0.02 (95% confidence interval 0.001, 0.25) for maximum middle cerebral artery Doppler velocity.
The literature reporting noninvasive techniques to predict fetal anemia is methodologically poor and a standard approach to the evaluation of these techniques is lacking. A recommendation for practice cannot be generated without further rigorous research.