Serum inhibin A and angiogenic factor levels in pregnancies with previous preeclampsia and/or chronic hypertension: Are they useful markers for prediction of subsequent preeclampsia?
Sibai, B. M., Koch, M., Freire, S., Silva, J. L. P. E., Rudge, M. V. C., Martins-Costa, S., ... Spinnato, J. A. (2008). Serum inhibin A and angiogenic factor levels in pregnancies with previous preeclampsia and/or chronic hypertension: Are they useful markers for prediction of subsequent preeclampsia? American Journal of Obstetrics and Gynecology, 199(3), 268-e1-268.e9. DOI: 10.1016/j.ajog.2008.06.071
Our objective was to determine whether measurement of placenta growth factor (PLGF), inhibin A, or soluble fms-like tyrosine kinase-1 (sFlt-1) at 2 times during pregnancy would usefully predict subsequent preeclampsia (PE) in women at high risk.
We analyzed serum obtained at enrollment (120/7 to 196/7 weeks) and follow-up (24-28 weeks) from 704 patients with previous PE and/or chronic hypertension (CHTN) enrolled in a randomized trial for the prevention of PE. Logistic regression analysis assessed the association of log-transformed markers with subsequent PE; receiver operating characteristic analysis assessed predictive value.
One hundred four developed preeclampsia: 27 at 37 weeks or longer and 77 at less than 37 weeks (9 at less than 27 weeks). None of the markers was associated with PE at 37 weeks or longer. Significant associations were observed between PE at less than 37 weeks and reduced PLGF levels at baseline (P = .022) and follow-up (P < .0001) and elevated inhibin A (P < .0001) and sFlt-1 (P = .0002) levels at follow-up; at 75% specificity, sensitivities ranged from 38% to 52%. Using changes in markers from baseline to follow-up, sensitivities were 52-55%. Associations were observed between baseline markers and PE less than 27 weeks (P ? .0004 for all); sensitivities were 67-89%, but positive predictive values (PPVs) were only 3.4-4.5%.
Inhibin A and circulating angiogenic factors levels obtained at 120/7 to 196/7 weeks have significant associations with onset of PE at less than 27 weeks, as do levels obtained at 24-28 weeks with onset of PE at less than 37 weeks. However, because the corresponding sensitivities and/or PPVs were low, these markers might not be clinically useful to predict PE in women with previous PE and/or CHTN.