Intrahepatic cholestasis of pregnancy is a pregnancy-specific liver disorder, possibly associated with an increased risk of severe fetal adverse events. Total serum bile acids (TSBA) concentration, alone or in combination with serum aminotransferases, have been the most often used biomarkers for the diagnosis of intrahepatic cholestasis of pregnancy in clinical practice. Serum bile acid profile, composed of primary or secondary, conjugated or non-conjugated bile acids, may provide more specific disease information.To assess and compare, independently or in combination, the diagnostic accuracy of total serum bile acids or serum bile acids profile, or both, for the diagnosis of intrahepatic cholestasis of pregnancy in pregnant women, presenting with pruritus. To define the optimal cut-off values for components of serum bile acid profile; to investigate possible sources of heterogeneity.We searched the Cochrane Hepato-Biliary Group Controlled Trials Register, the Cochrane Hepato-Biliary Group Diagnostic Test Accuracy Studies Register, the Cochrane Library, MEDLINE Ovid, Embase Ovid, Science Citation Index Expanded, Conference Proceedings Citation Index - Science, BIOSIS, CINAHL, two Chinese databases (CKNI, VIP), Latin American and Caribbean Health Sciences Literature (LILACS), Scientific Electronic Library Online (SciELO), Evidence Search: Health and Social Care by the National Institute for Health and Care Excellence (NICE), the World Health Organization (WHO) Reproductive Health Library (RHL), and the Turning Research into Practice database (TRIP). The most recent date of search was 6 May 2019. We identified additional references by handsearching the references of articles, meta-analyses, and evidence-based guidelines retrieved from the computerised databases, on-line trial registries, and grey literature through OpenSIGLE, National Technical Information Service (NTIS), ProQuest Dissertations & Thesis Database, and Index to Theses in Great Britain and Ireland.Prospective or retrospective diagnostic case-control or cross-sectional studies, irrespective of publication date, format, and language, which evaluated the diagnostic accuracy of total serum bile acids (TSBA) or components of serum bile acid profile for the diagnosis of intrahepatic cholestasis of pregnancy in pregnant women of any age or ethnicity, in any clinical setting, symptomatic for pruritus.We selected studies by reading titles, abstracts, or full texts, and assessing their fulfilment of our inclusion criteria. We emailed primary authors to request missing data or individual participant data. Having extracted data from each included study, we built the two-by-two tables for each primary study and for all the index tests considered. We estimated sensitivity and specificity with their 95% confidence intervals (CI). We presented data in coupled forest plots, showing sensitivities and specificities of each study, and we plotted the studies in the Receiver Operating Characteristic (ROC) space. We performed meta-analyses adopting the hierarchical summary ROC model (HSROC) or the bivariate model to meta-analyse the data. We made indirect comparisons of the considered index tests by adding the index tests as covariates to the bivariate or HSROC models. We performed heterogeneity analysis and sensitivity analysis on studies assessing TSBA accuracy. We used Review Manager 5 (RevMan 5) and SAS statistical software, release 9.4 (SAS Institute Inc., Cary, NC, USA), to perform all statistical analyses. We used QUADAS-2 domains to assess the risk of bias of the included studies.Our search yielded 5073 references, but at the end of our selection process, only 16 studies fulfilled the review inclusion criteria. Nine of these provided individual participant data. We analysed only data concerning TSBA, cholic acid (CA), glycocholic acid (GCA), chenodeoxycholic acid (CDCA), and CA/CDCA because the remaining planned index tests were assessed in few studies. Only one study had low risk of bias in all four QUADAS-2 domains. The most biased domains were the patient sampling and the reference standard domains. When considering all studies with a cut-off of 10 μmol/L, TSBA overall sensitivity ranged from 0.72 to 0.98 and specificity ranged from 0.81 to 0.97. After a sensitivity analysis excluding case-control studies, TSBA sensitivity ranged from 0.48 to 0.66 and specificity from 0.52 to 0.99. After a sensitivity analysis excluding studies in which TSBA was part of the reference standard, TSBA sensitivity ranged from 0.49 to 0.65 and specificity from 0.53 to 0.99. We found the estimates of the overall accuracy for some serum bile acid components (CA, GCA, CDCA, and CA/CDCA) to be imprecise, with the CI for sensitivity and specificity very wide or impossible to calculate. Indirect comparisons between serum bile acid profile components and TSBA were not statistically significant. None of the heterogeneity analysis performed was statistically significant, except for the timing of assessment of TSBA (onset of symptoms, peak value among multiple assessments, delivery) but without clinically relevant results. We could not analyse the diagnostic accuracy of combinations of index tests because none of the included studies carried them out, and because of the small number of included studies.The overall high risk of bias, the existing concern regarding applicability of the results in clinical practice, and the great heterogeneity of the results in the included studies prevents us from making recommendations and reaching definitive conclusions at the present time. Thus, we do not find any compelling evidence to recommend or refute the routine use of any of these tests in clinical practice. So far, the diagnostic accuracy of TSBA for intrahepatic cholestasis of pregnancy might have been overestimated. There were too few studies to permit a precise estimate of the accuracy of serum bile acid profile components. Further primary clinical research is mandatory. We need both further phase II and phase III diagnostic studies.
