Rationalizing the management of pregnancies of unknown location: Diagnostic accuracy of human chorionic gonadotropin ratio‐based decision tree compared with the risk prediction model M4

Conventional and modern markers of pregnancy of unknown location: Update and narrative review

Pregnancy of unknown location (PUL) is a temporary pathologic or physiologic phenomenon of early pregnancy that requires follow up to determine the final pregnancy outcome. Evidence indicated that PUL patients suffer a remarkably higher rate of adverse pregnancy outcomes, represented by ectopic gestation and early pregnancy loss, than the general population. In the past few decades, discussion about PUL has never stopped, and a variety of markers have been widely investigated for the early and accurate evaluation of PUL, including serum biomarkers, ultrasound imaging features, multivariate analysis, and the diagnosis of ectopic pregnancy based on risk stratification. So far, machine learning (ML) methods represented by M4 and M6 logistic regression have gained a level of recognition and are continually improving. Nevertheless, the heterogeneity of PUL markers, mainly caused by the limited sample size, the differences in population and technical maturity, etc., have hampered the management of PUL. With the advancement of multidisciplinary integration and cutting-edge technologies (e.g. artificial intelligence, prediction model development, and telemedicine), novel markers, and strategies for the management of PUL are expected to be developed. In this review, we summarize both conventional and novel markers (represented by artificial intelligence) for PUL assessment and management, investigate their advancements, limitations and challenges, and propose insights on future research direction and clinical application.

Development of a single-visit protocol for the management of pregnancy of unknown location following in vitro fertilization: a retrospective study

STUDY QUESTION

Can women with pregnancy of unknown location (PUL) following in vitro fertilization (IVF) be risk-stratified regarding the subsequent need for medical intervention, based on their demographic characteristics and the results of serum biochemistry at the initial visit?

SUMMARY ANSWER

The ratio of serum hCG to number of days from conception (hCG/C) or the initial serum hCG level at ≥5 weeks' gestation could be used to estimate the risk of women presenting with PUL following IVF and needing medical intervention during their follow-up.

WHAT IS KNOWN ALREADY

In women with uncertain conception dates presenting with PUL, a single serum hCG measurement cannot be used to predict the final pregnancy outcomes, thus, serial levels are mandatory to establish a correct diagnosis. Serum progesterone levels can help to risk-stratify women at their initial visit but are not accurate in those taking progesterone supplementation, such as women pregnant following IVF.

STUDY DESIGN, SIZE, DURATION

This was a retrospective study carried out at two specialist early pregnancy assessment units between May 2008 and January 2021. A total of 224 women met the criteria for inclusion, but 14 women did not complete the follow-up and were excluded from the study.

PARTICIPANTS/MATERIALS, SETTING, METHODS

We selected women who had an IVF pregnancy and presented with PUL at ≥5 weeks' gestation.

MAIN RESULTS AND THE ROLE OF CHANCE

A total of 30/210 (14.0%, 95% CI 9.9-19.8) women initially diagnosed with PUL required surgical intervention. The hCG/C was significantly higher in the group of women requiring an intervention compared to those who did not (P = 0.003), with an odds ratio of 3.65 (95% CI 1.49-8.89, P = 0.004). A hCG/C <4.0 was associated with a 1.9% risk of intervention, which accounted for 25.7% of the study population. A similar result was obtained by substituting hCG/C <4.0 with an initial hCG level <100 IU/l, which was associated with 2.0% risk of intervention, and accounted for 23.8% of the study population (P > 0.05).

LIMITATIONS, REASONS FOR CAUTION

A limitation of our study is that it is retrospective in nature, and as such, we were reliant on existing data.

WIDER IMPLICATIONS OF THE FINDINGS

A previous study in women with PUL after spontaneous conception found that a 2% intervention rate was considered low enough to eliminate the need for close follow-up and serial blood tests. Using the same 2% cut-off, a quarter of women with PUL after IVF could also avoid attending for further visits and investigations.

STUDY FUNDING/COMPETING INTEREST(S)

No external funding was required for this study. No conflicts of interest are required to be declared.

TRIAL REGISTRATION NUMBER

N/A.

Pregnancy of unknown location: external validation of the hCG-based M6NP and M4 prediction models in an emergency gynaecology unit

OBJECTIVE

To investigate if M6NP predicting ectopic pregnancy (EP) among women with pregnancy of unknown location (PUL) is valid in an emergency gynaecology setting and comparing it with its predecessor M4.

DESIGN

Retrospective cohort study.

SETTING

University Hospital.

PARTICIPANTS

Women with PUL.

METHODS

All consecutive women with a PUL during a study period of 3 years were screened for inclusion. Risk prediction of an EP was based on two serum human chorionic gonadotropin (hCG) levels taken at least 24 hours and no longer than 72 hours apart.

MAIN OUTCOME MEASURES

The area under the ROC curve (AUC) expressed the ability of a model to distinguish an EP from a non-EP (discrimination). Calibration assessed the agreement between the predicted risk of an EP and the true risk (proportion) of EP. The proportion of EPs and non-EPs classified as high risk assessed the model's sensitivity and false positive rate (FPR). The proportion of non-EPs among women classified as low risk was the model's negative predictive value (NPV). The clinical utility of a model was evaluated with decision curve analysis.

RESULTS

1061 women were included in the study, of which 238 (22%) had a final diagnosis of EP. The AUC for EP was 0.85 for M6NP and 0.81 for M4. M6NP made accurate risk predictions of EP up to predictions of 20% but thereafter risks were underestimated. M4 was poorly calibrated up to risk predictions of 40%. With a 5% threshold for high risk classification the sensitivity for EP was 95% for M6NP, the FPR 50% and NPV 97%. M6NP had higher sensitivity and NPV than M4 but also a higher FPR. M6NP had utility at all thresholds as opposed to M4 that had no utility at thresholds≤5%.

CONCLUSIONS

M6NP had better predictive performance than M4 and is valid in women with PUL attending an emergency gynaecology unit. Our results can encourage implementation of M6NP in related yet untested clinical settings to effectively support clinical decision-making.

External validation of risk prediction model M4 in an Australian population: Rationalising the management of pregnancies of unknown location

BACKGROUND

The prediction model M4 can successfully classify pregnancy of unknown location (PUL) into a low- or high-risk group in developing ectopic pregnancy. M4 was validated in UK centres but in very few other countries outside UK.

AIM

To validate the M4 model's ability to correctly classify PULs in a cohort of Australian women.

MATERIALS AND METHODS

A retrospective analysis of women classified with PUL, attending a Sydney-based teaching hospital between 2006 and 2018. The reference standard was the final characterisation of PUL: failed PUL (FPUL) or intrauterine pregnancy (IUP; low risk) vs ectopic pregnancy (EP) or persistent PUL (PPUL; high risk). Each patient was entered into the M4 model calculator and an estimated risk of FPUL/IUP or EP/PPUL was recorded. Diagnostic accuracy of the M4 model was evaluated.

RESULTS

Of 9077 consecutive women who underwent transvaginal sonography, 713 (7.9%) classified with a PUL. Six hundred and seventy-seven (95.0%) had complete study data and were included. Final outcomes were: 422 (62.3%) FPULs, 150 (22.2%) IUPs, 105 (15.5%) EPs and PPULs. The M4 model classified 455 (67.2%) as low-risk PULs of which 434 (95.4%) were FPULs/IUPs and 21 (4.6%) were EPs or PPULs. EPs/PPULs were correctly classified with sensitivity of 80.0% (95% CI 71.1-86.5%), specificity of 75.9% (95% CI 72.2-79.3%), positive predictive value of 37.8% (95% CI 33.8-42.1%) and negative predictive value of 95.3% (95% CI 93.1-96.9%).

CONCLUSIONS

We have externally validated the prediction model M4. It classified 67.2% of PULs as low risk, of which 95.4% were later characterised as FPULs or IUPs while still classifying 80.0% of EPs as high risk.