Human chorionic gonadotropin value in early pregnancy after in vitro fertilization as a predictor of pregnancy outcome: a systematic review and meta-analysis
Sharon Galperin M.D. , Julian A. Gingold M.D., Ph.D. , Tova Niderberg M.D. , Christine Leinbach Seaton M.D. , Juan Lin Ph.D. , Rachel Schwartz MLS , Staci E. Pollack M.D., M.S.
{"title":"Human chorionic gonadotropin value in early pregnancy after in vitro fertilization as a predictor of pregnancy outcome: a systematic review and meta-analysis","authors":"Sharon Galperin M.D. , Julian A. Gingold M.D., Ph.D. , Tova Niderberg M.D. , Christine Leinbach Seaton M.D. , Juan Lin Ph.D. , Rachel Schwartz MLS , Staci E. Pollack M.D., M.S.","doi":"10.1016/j.xfnr.2024.01.001","DOIUrl":null,"url":null,"abstract":"<div><h3>Objective</h3><p><span>To assess the initial human chorionic gonadotropin (hCG) value as a predictor of live </span>birth<span> (LB) after in vitro fertilization (IVF) therapy.</span></p></div><div><h3>Evidence Review</h3><p><span>A systematic search of PubMed, Embase, and Google Scholar results between January 1, 1985, and January 20, 2022, was conducted. Studies that reported hCG levels for adult females who underwent IVF treatment and had a positive serum pregnancy test were included. Studies must have measured serum hCG levels before 10 weeks gestational age and calculated cutoff hCG level values for predicting pregnancy outcome from receiver operating characteristic (ROC) curves. Exclusion criteria included spontaneous pregnancies<span><span>, intrauterine insemination, </span>gamete intrafallopian transfer, and </span></span>ovulation induction cycles. The outcome assessed was pregnancies resulting in LB. Study quality and risk of bias were assessed via the Newcastle–Ottawa Scale and the QUADAS-2 tool.</p></div><div><h3>Result(s)</h3><p><span>A total of 1,510 abstracts were screened, and 157 full texts were reviewed. Twenty-six retrospective cohort studies were selected, which included 34,220 IVF cycles with positive hCG values. Gestational age on the day of hCG level measurement ranged from 23–33 days, with days 28 and 31 as the most frequently occurring days. Day 28 hCG cutoff values for predicting LB ranged from 49.2–108.0 IU/L, with sensitivities ranging from 71.5%–93.7%. Day 31 hCG cutoff values ranged from 145.00–411.45 IU/L, with sensitivities ranging from 61.2%–96.1%. Summary ROC curves generated from bivariate analyses of sensitivity and specificity showed areas under the curves of 0.845 and 0.718 for days 28 and 31, respectively. Summary ROC curves of day 31 hCG values after fresh and frozen-thawed </span>embryo transfers showed areas under the curve of 0.701 and 0.889, respectively.</p></div><div><h3>Conclusion(s)</h3><p>Day 28 hCG cutoff values may provide more reliable predictions for LB compared with day 31. On day 28, an hCG level of 49.2 IU/L would identify 93.7% of the patients who will go on to have a LB, and an hCG level of 108 IU/L would identify 71.5%. However, significant study heterogeneity complicates the quality of estimated hCG levels, sensitivity, and specificity comparisons.</p></div>","PeriodicalId":73011,"journal":{"name":"F&S reviews","volume":null,"pages":null},"PeriodicalIF":0.0000,"publicationDate":"2024-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"F&S reviews","FirstCategoryId":"1085","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S266657192400001X","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
Abstract
Objective
To assess the initial human chorionic gonadotropin (hCG) value as a predictor of live birth (LB) after in vitro fertilization (IVF) therapy.
Evidence Review
A systematic search of PubMed, Embase, and Google Scholar results between January 1, 1985, and January 20, 2022, was conducted. Studies that reported hCG levels for adult females who underwent IVF treatment and had a positive serum pregnancy test were included. Studies must have measured serum hCG levels before 10 weeks gestational age and calculated cutoff hCG level values for predicting pregnancy outcome from receiver operating characteristic (ROC) curves. Exclusion criteria included spontaneous pregnancies, intrauterine insemination, gamete intrafallopian transfer, and ovulation induction cycles. The outcome assessed was pregnancies resulting in LB. Study quality and risk of bias were assessed via the Newcastle–Ottawa Scale and the QUADAS-2 tool.
Result(s)
A total of 1,510 abstracts were screened, and 157 full texts were reviewed. Twenty-six retrospective cohort studies were selected, which included 34,220 IVF cycles with positive hCG values. Gestational age on the day of hCG level measurement ranged from 23–33 days, with days 28 and 31 as the most frequently occurring days. Day 28 hCG cutoff values for predicting LB ranged from 49.2–108.0 IU/L, with sensitivities ranging from 71.5%–93.7%. Day 31 hCG cutoff values ranged from 145.00–411.45 IU/L, with sensitivities ranging from 61.2%–96.1%. Summary ROC curves generated from bivariate analyses of sensitivity and specificity showed areas under the curves of 0.845 and 0.718 for days 28 and 31, respectively. Summary ROC curves of day 31 hCG values after fresh and frozen-thawed embryo transfers showed areas under the curve of 0.701 and 0.889, respectively.
Conclusion(s)
Day 28 hCG cutoff values may provide more reliable predictions for LB compared with day 31. On day 28, an hCG level of 49.2 IU/L would identify 93.7% of the patients who will go on to have a LB, and an hCG level of 108 IU/L would identify 71.5%. However, significant study heterogeneity complicates the quality of estimated hCG levels, sensitivity, and specificity comparisons.