026 How good is your gestalt in diagnosing acute coronary syndrome? A systematic literature review. Issue 12 (21st November 2019)

• Record Type: Journal Article
• Authors: Oliver, Govind; Berg, Patricia van den; Reynard, Charles; Body, Richard
• Is Part Of: Emergency medicine journal. Volume 36:Issue 12(2019)
• Page Start: 790
• Page End: 791
• Publication Date: 2019-11-21
• Subjects: Emergency medicine -- Periodicals; 616.02505
• DOI: 10.1136/emermed-2019-RCEM.26
• Languages: English
• ISSNs: 1472-0205

Abstract

Abstract : Chest pain accounts for 6% of Emergency Departments (ED) attendances. One of the most serious diagnoses clinicians must consider is Acute Coronary Syndrome (ACS). The challenge of early identification and timely but safe discharge of those without ACS, has led to multiple decision rules being developed. Many clinicians prefer and continue to use clinical judgement alone. We aimed to examine the current evidence on the diagnostic accuracy of clinical judgement or gestalt for 'rule in' and 'rule out' of ACS. A systematic literature review using Ovid Medline, Embase and Cochrane databases was conducted using a pre-defined search strategy. Returned results were independently screened by two reviewers for inclusion followed by a hand search of the grey literature and the references of relevant studies. Quality assessment of potentially eligible studies was carried out using a study-specific QUADAS-2 tool. Results were reported using the PRISMA guidelines. 171 studies were screened, 13 underwent full text review, 6 were quality assessed using QUADAS-2 and 2 included. Meta-analysis plans were limited to a narrative description. The two included prospective cohort studies reported on 255 and 458 ED patients with suspected ACS. Gestalt identified 30.1% and 27.1% of patients as low risk. Sensitivity was 86.7% (95% confidence interval 76.8–93.4%) and 93.0% (86.1–97.1%) and negative predicative values (NPV) of 87.3% (79.0–92.7%) and 94.4% (89.0–97.2%) respectively. For the 30.6% and 38.0% of patients identified as high risk by gestalt in each study, specificity was 81.1% (74.6–86.6%) and 70.4% (65.4–75.1%) with positive predictive values of 56.4% (47.5–64.9%) and 39.1% (34.2–44.2%) respectively. A gestalt based rule out strategy incorporating ECG and arrival troponin performed with a sensitivity and NPV of 99.0%. A significant evidence gap has been identified. At present there is insufficient information to draw meaningful conclusion. The little evidence supporting a gestalt-based diagnostic strategy urgently needs prospective validation.