Diagnostic protocols for the management of pregnancy of unknown location

Pregnancy of unknown location (PUL) is a term used to describe a non-diagnostic transvaginal ultrasound scan in early pregnancy. The scan may be non-diagnostic because the pregnancy is too small to identify, too difficult to identify, or absent, having already undergone the process of miscarriage. Many patients are surprised to learn that urine pregnancy tests are more sensitive in detecting a pregnancy than ultrasound scans, but as long as this remains the case, or until we discover a biomarker that can locate a pregnancy, the management of women with PUL will continue to form a significant part of the workload of Early Pregnancy Units. The meta-analysis of the various protocols described in the literature reported by Bobdiwala et al. is welcome as an attempt to compare the different strategies for the initial triage of patients with PUL. The authors acknowledge that there was a great deal of heterogeneity between studies in the definition of PUL, the defined outcomes, and the risk of verification bias. They chose to assess the likelihood of an underlying ectopic pregnancy as the primary outcome, and their meta-analysis demonstrates that the M4 model has the best overall performance for predicting an ectopic pregnancy later diagnosed on ultrasound. The model is based on serum human chorionic gonadotropin (hCG) levels over 48 hours and defines a high risk of ectopic pregnancy to be anything >5%. Follow-up is more intensive for this subgroup of patients, comprising of about a third of the PUL population. The alternative strategy for managing PUL is to identify a subgroup of patients at low risk of intervention, accepting that there will be ectopic pregnancies within this group, but that they are likely to resolve spontaneously. With regards to this strategy of identifying failing PULs, Bobdiwala et al. BJOG 2019;126:190–8. found that the hCG ratio performed best, and it was a variation of this protocol that was adopted in the 2012 National Institute for Health and Care Excellence (NICE) guidance (Newbatt et al. BMJ 2012;345:e8136). The current meta-analysis could not take into account the usability of the different models in terms of losses to follow-up and protocol deviations. The problem with deciding on the best protocol to adopt in clinical practice is that there is huge variation in early pregnancy care that may not be reflective of practice in the units generating research into PUL. Early pregnancy units are still in the process of being established in many countries (Rovner et al. J Ultrasound Med. 2018;37:1533–8). UK hospitals vary in terms of access to Early Pregnancy Units, with some restrictions according to presumed gestational age and with limited opening hours, particularly over the weekend. There is also variation in diagnostic thresholds for the ultrasound diagnosis of intrauterine pregnancies, the ability to recognise ectopic pregnancies, and in the training, experience, and continuity of sonographers. In addition to patient factors, such as variation within local populations affecting compliance with recommended follow-up schedules, the quality and availability of clinical leadership within departments will also influence the ability to successfully implement specific protocols. Thus, it is important that every clinic audits their own practice and adopts a triage strategy and follow-up protocol for PUL that performs best for them. None declared. Completed disclosure of interest form is available to view online as supporting information. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.