Abstract

OBJECTIVES: Twin anemia polycythemia sequence (TAPS) is associated with increased perinatal mortality and morbidity. Inconsistencies in the diagnostic criteria for TAPS exist, which hinder the ability to establish robust evidence-based management or monitoring protocols. The main aim of this study was to determine, by expert consensus using a Delphi procedure, the key diagnostic features and optimal monitoring approach for TAPS. METHODS: A Delphi process was conducted among an international panel of experts. Panel members were provided with a list of literature-based parameters for diagnosing and monitoring TAPS. They were asked to rate their importance on a five-point Likert scale. Consensus was sought to determine the cut-off values for accepted parameters, as well as parameters used in the monitoring and assessment of outcome of twin pregnancy complicated by TAPS. RESULTS: A total of 132 experts were approached, 50 joined the first round; of whom 33 (66%) completed all three rounds. There was agreement that the monitoring interval for the development of TAPS should be every 2 weeks and that the severity should be assessed antenatally using a classification system based on the middle cerebral artery (MCA) peak systolic velocity (PSV), but no agreement on the gestation at which to start. Once the diagnosis is made, monitoring should be scheduled weekly. For the diagnosis of TAPS, the combination of MCA PSV ≥ 1.5MoM in the anemic twin and ≤ 0.8MoM in the polycythemic twin was agreed. Alternatively, MCA PSV discordance ≥ 1MoM can be used to diagnose TAPS. Postnatally, hemoglobin difference ≥ 8 g/dL and inter-twin reticulocyte ratio ≥ 1.7 were agreed criteria. There was no agreement on the cut-off of the MCA PSV or its discordance for prenatal intervention. The panel agreed on prioritising perinatal and long-term survival outcomes in follow-up studies. CONCLUSIONS: Consensus-based diagnostic features of TAPS, as well as cut-off values for the parameters involved, were agreed upon by a panel of experts. Future studies are needed to validate these diagnostic features before they can be used in clinical trials of interventions. This article is protected by copyright. All rights reserved.