Polena V., Huchon C., Ramos C. V., Rouzier R., Dumont Alexandre, Fauconnier A. (2015). Non-invasive tools for the diagnosis of potentially life-threatening gynaecological emergencies : a systematic review. Plos One, 10 (2), p. e0114189 [17 p.]. ISSN 1932-6203.
Titre du document
Non-invasive tools for the diagnosis of potentially life-threatening gynaecological emergencies : a systematic review
Objective To identify non- invasive tools for diagnosis of the major potentially life-threatening gynaecological emergencies (G-PLEs) reported in previous studies, and to assess their diagnostic accuracy. Methods MEDLINE; EMBASE; Cochrane Central Register of Controlled Trials (CENTRAL; The Cochrane Library) were searched to identify all eligible studies published in English or French between January 1990 and December 2012. Studies were considered eligible if they were primary diagnostic studies of any designs, with a gold standard and with sufficient information for construction of a 2 x 2 contingency table, concerning at least one of the following G-PLEs: complicated ectopic pregnancy, complicated pelvic inflammatory disease, adnexal torsion and haemoperitoneum of any gynaecological origin. Extraction of data and assessment of study quality were conducted by two independent reviewers. We set the thresholds for the diagnostic value of signs retrieved at Sensibility >= 95% and LR- <= 0.25, or Specificity >= 90% and LR+ >= 4. Results We identified 8288 reports of diagnostic studies for the selected G-PLEs, 45 of which met the inclusion criteria. The methodological quality of the included studies was generally low. The most common diagnostic tools evaluated were transvaginal ultrasound (20/45), followed by medical history (18/45), clinical examination (15/45) and laboratory tests (14/45). Standardised questioning about symptoms, systolic blood pressure < 110 mmHg, shock index>0.85, identification of a mass by abdominal palpation or vaginal examination, haemoglobin concentration < 10 g/dl and six ultrasound and Doppler signs presented high performances for the diagnosis of G-PLEs. Transvaginal ultrasound was the diagnostic tool with the best individual performance for the diagnosis of all G-PLEs. Conclusion This systematic review suggests that blood pressure measurement, haemoglobin tests and transvaginal ultrasound are cornerstone examinations for the diagnosis of G-PLEs that should be available in all gynaecological emergency care services. Standardised questioning about symptoms could be used for triage of patients.
