RESUMO
BACKGROUND: Acute abdominal pain is a common complaint, caused by a variety of conditions, ranging from acutely life-threatening to benign and self-limiting, with symptom overlap complicating diagnosis. Signs and symptoms may be valuable when assessing a patient to guide clinical work. AIM: Summarising evidence on the accuracy of signs and symptoms for diagnosing serious illness in adults with acute abdominal pain in an ambulatory care setting. DESIGN & SETTING: We performed a systematic review, searching for prospective diagnostic accuracy studies that included adults presenting with acute abdominal pain to an ambulatory care setting. METHOD: Six databases and guideline registers were searched, using a comprehensive search strategy. We assessed the risk of bias, and calculated descriptive statistics and measures of diagnostic accuracy. Results were pooled when at least four studies were available. RESULTS: Out of 18 923 unique studies, 16 studies with moderate to high-risk bias were included. Fourteen clinical features met our criteria, including systolic blood pressure <100 mmHg (positive likelihood ratio [LR+]7.01), shock index >0.85, uterine cervical motion tenderness (LR+5.62 and negative likelihood ratio [LR-]8.60), and a self-assessment questionnaire score >70 (LR+12.20) or <25 (LR-0.19). Clinical diagnosis made by the clinician had the best rule-in ability (LR+24.6). CONCLUSIONS: We identified 14 signs and symptoms that can influence the likelihood of a serious illness, including pain characteristics, systemic signs, gynaecological signs, and clinician's overall assessment. The risk of bias was moderate to high, leading to uncertainty and preventing us from making firm conclusions. This highlights the need for better research in this setting.