Inflammatory Biomarkers and Clinical Judgment in the Emergency Diagnosis of Urgent Abdominal Pain Tobias Breidthardt, 1,2*† Nora Brunner-Schaub, 1† Catharina Balmelli, 1 Juan Jose Sancho Insenser, 3 Katrin Burri-Winkler, 1,3 Nicolas Geigy, 4 Lukas Mundorff, 4 Aristomenis Exadaktylos, 5 Julia Scholz, 1 Philip Haaf, 1,6 Christian Hamel, 7,8 Daniel Frey, 7,9 Karen Delport, 10 W. Frank Peacock, 11 Michael Freese, 1,6 Salvatore DiSomma, 12 John Todd, 13 Katharina Rentsch, 14 Roland Bingisser, 15 and Christian Mueller 1,6 for the BASEL VII Investigators BACKGROUND: The early diagnosis of urgent abdominal pain (UAP) is challenging. Most causes of UAP are associ- ated with extensive inflammation. Therefore, we hypothe- sized that quantifying inflammation using interleukin-6 and/or procalcitonin would provide incremental value in the emergency diagnosis of UAP. METHODS: This was an investigator-initiated prospective, multicenter diagnostic study enrolling patients presenting to the emergency department (ED) with acute abdominal pain. Clinical judgment of the treating physician regarding the presence of UAP was quantified using a visual analog scale after initial clinical and physician-directed laboratory assessment, and again after imaging. Two independent specialists adjudicated the final diagnosis and the clas- sification as UAP (life-threatening, needing urgent surgery and/or hospitalization for acute medical rea- sons) using all information including histology and follow-up. Interleukin-6 and procalcitonin were mea- sured blinded in a central laboratory. RESULTS: UAP was adjudicated in 376 of 1038 (36%) patients. Diagnostic accuracy for UAP was higher for interleukin-6 [area under the ROC curve (AUC), 0.80; 95% CI, 0.77– 0.82] vs procalcitonin (AUC, 0.65; 95% CI, 0.62– 0.68) and clinical judgment (AUC, 0.69; 95% CI, 0.65– 0.72; both P  0.001). Combined assessment of interleukin-6 and clinical judgment increased the AUC at presentation to 0.83 (95% CI, 0.80 – 0.85) and after imaging to 0.87 (95% CI, 0.84 – 0.89) and im- proved the correct identification of patients with and without UAP (net improvement in mean predicted prob- ability: presentation, +19%; after imaging, +15%; P  0.001). Decision curve analysis documented incremental value across the full range of pretest probabilities. A clin- ical judgment/interleukin-6 algorithm ruled out UAP with a sensitivity of 97% and ruled in UAP with a spec- ificity of 93%. CONCLUSIONS: Interleukin-6 significantly improves the early diagnosis of UAP in the ED. © 2018 American Association for Clinical Chemistry Acute abdominal pain is the most common presenting symptom in emergency department (ED) 16 patients (1– 3). Its differential diagnosis is extensive and challenging (4, 5 ). Although 20% to 25% of patients require hospital admission and 10% of all episodes are life-threatening or require urgent surgery, most episodes of acute abdominal pain are self-limiting and benign (3, 4, 6 ). Hence, rapid and accurate diagnosis of urgent causes of abdominal pain (UAP) is essential for the early initiation of effective therapy and efficient patient flow. Clinical assessment including patient history, de- tailed physical examination, and routine laboratory test- ing remains the cornerstone of initial patient care (7, 8 ). However, in isolation this strategy has poor diagnostic accuracy (9, 10 ). Imaging techniques, particularly com- puted tomography (CT) scans (10, 11 ), are of enormous value (12 ), but the appropriate selection of patients ben- 1 Cardiovascular Research Institute Basel (CRIB), Basel, Switzerland; 2 Department of In- ternal Medicine, University Hospital, Basel, and University of Basel, Switzerland; 3 Hos- pital del Mar, Barcelona, Spain; 4 Kantonsspital Baselland, Liestal, Switzerland; 5 Emer- gency Department, Inselspital, Berne, Switzerland; 6 Department of Cardiology, University Hospital, Basel, and University of Basel, Switzerland; 7 Department of Visceral Surgery, University Hospital, Basel, and University of Basel, Switzerland; 8 Department of Visceral Surgery, Kliniken des Landkreises, Lörrach, Germany; 9 Department of Surgery, Spital Wetzikon, Wetzikon, Switzerland; 10 Emergency Department, Kantonspital Ba- selland, Bruderholz, Switzerland; 11 Department of Emergency Medicine, Baylor College of Medicine, Houston, TX; 12 Emergency Department, San Andrea Hospital, University Sapienza, Rome, Italy; 13 Singulex Inc., Clinical Research, Alameda, CA; 14 Department of Laboratory Medicine, University Hospital, Basel, and University of Basel, Switzerland; 15 Emergency De- partment, University Hospital, Basel, and University of Basel, Switzerland. * Address correspondence to this author at: Clinic of Internal Medicine and Cardiovascular Research Institute Basel (CRIB), University Hospital Basel, University of Basel, Petersgra- ben 4, CH-4031 Basel, Switzerland. E-mail tobias.breidthardt@usb.ch. † T. Breidthardt and N. Brunner-Schaub contributed equally to this work and should be considered first author. Received August 24, 2018; accepted November 15, 2018. Previously published online at DOI: 10.1373/clinchem.2018.296491 © 2018 American Association for Clinical Chemistry 16 Nonstandard abbreviations: ED, emergency department; UAP, urgent abdominal pain; CT, computed tomography; VAS, visual analog scale; AUC, area under the ROC curve; NRI, net reclassification improvement; IDI, integrated discrimination improvement; WBC, white blood cells; CRP, C-reactive protein; IBS, irritable bowel syndrome. Clinical Chemistry 65:2 302–312 (2019) Evidence-Based Medicine and Test Utilization 302 Downloaded from https://academic.oup.com/clinchem/article-abstract/65/2/302/5607969 by guest on 04 July 2020