Number of oocytes retrieved and resulting pregnancy - Risk factors for ovarian hyperstimulation syndrome

OBJECTIVE: To test whether the risk of developing clinically significant ov arian hyperstimulation syndrome (OHSS) is related to the number of oocytes retrieved by puncture and the resulting pregnancy and to determine the most suitable cutoff limit of the number of oocytes in predicting OHSS. STUDY DESIGN: The study included 973 patients who underwent ovarian stimula tion for in vitro fertilization or intracytoplasmic sperm injection. Using the classification of Schenker and Weinstein, we identified patients who de veloped moderate and severe OHSS. By multiple logistic regression we establ ished the risk of moderate or severe OHSS development in relation to the nu mber of oocytes retrieved and the resulting pregnancy. A receiver operator characteristic curve was constructed to describe the relation between sensi tivity and the false positive rate for the number of oocytes retrieved in t he prediction of OHSS. RESULTS: We identified 35 (3.6%) patients who developed OHSS, 8 (0.8%) seve re and 27 (2.8%) moderate. The risk of developing OHSS increased with the n umber of oocytes retrieved (odds ratio=1.14) and with pregnancy (odds ratio =1.14). The most suitable limit for predicting OHSS was 10 oocytes, with 81 .9% specificity and 48.6% sensitivity. The risk of OHSS development in cycl es with less than or equal to 10 oocytes and no conception was 1.31% and, w ith conception, 5.12%. The risk of OHSS development in cycles with >10 oocy tes and no conception was 4.43% and with conception, 15.93%. CONCLUSION: The risk of OHSS development increases with the number of oocyt es retrieved and with pregnancy. The most suitable limit for predicting OHS S is 10 oocytes; however, due to low sensitivity, it is also necessary to c onsider other factors when establishing the increased risk of OHSS.