key: cord-0952585-zglimqrs
authors: Nazerian, Peiman; Morello, Fulvio; Prota, Alessio; Betti, Laura; Lupia, Enrico; Apruzzese, Luc; Oddi, Matteo; Grosso, Federico; Grifoni, Stefano; Pivetta, Emanuele
title: Diagnostic accuracy of physician’s gestalt in suspected COVID‐19: Prospective bicentric study
date: 2021-03-15
journal: Acad Emerg Med
DOI: 10.1111/acem.14232
sha: 0d1a1f71b181af072e44363cd083a5d73081b360
doc_id: 952585
cord_uid: zglimqrs

OBJECTIVES: Physicians’ gestalt is central in the diagnostic pipeline of suspected COVID‐19, due to the absence of a single tool allowing conclusive rule in or rule out. The aim of this study was to estimate the diagnostic test characteristics of physician's gestalt for COVID‐19 in the emergency department (ED), based on clinical findings or on a combination of clinical findings and bedside imaging results. METHODS: From April 1 to April 30, 2020, patients with suspected COVID‐19 were prospectively enrolled in two EDs. Physicians prospectively dichotomized patients in COVID‐19 likely or unlikely twice: after medical evaluation of clinical features (clinical gestalt [CG]) and after evaluation of clinical features and results of lung ultrasound or chest x‐ray (clinical and bedside imaging–integrated gestalt [CBIIG]). The final diagnosis was adjudicated after independent review of 30‐day follow‐up data. RESULTS: Among 838 ED enrolled patients, 193 (23%) were finally diagnosed with COVID‐19. The area under the curve (AUC), sensitivity, and specificity of CG and CBIIG for COVID‐19 were 80.8% and 91.6% (p < 0.01), 82.9% and 91.4% (p = 0.01), and 78.6% and 91.8% (p < 0.01), respectively. CBIIG had similar AUC and sensitivity to reverse transcription–polymerase chain reaction (RT‐PCR) for SARS‐CoV‐2 on the first nasopharyngeal swab per se (93.5%, p = 0.24; and 87%, p = 0.17, respectively). CBIIG plus RT‐PCR had a sensitivity of 98.4% for COVID‐19 (p < 0.01 vs. RT‐PCR alone) compared to 95.9% for CG plus RT‐PCR (p = 0.05). CONCLUSIONS: In suspected COVID‐19, CG and CBIIG have fair diagnostic accuracy, in line with physicians’ gestalt for other acute conditions. Negative RT‐PCR plus low probability based on CBIIG can rule out COVID‐19 with a relatively low number of false‐negative cases.

A single test for conclusive diagnosis of coronavirus disease 19 is currently unavailable. Patient's medical history, signs, symptoms, physical examination, and routine laboratory findings per se are largely inaccurate. 1 Reverse transcription-polymerase chain reaction (RT-PCR) for severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) on a nasopharyngeal swab, representing the key microbiological tool, can give false-negative results with the need of repeating the test after days. [1] [2] [3] [4] Bedside diagnostic imaging tests, represented by lung ultrasound (LUS) and chest x-ray (CXR), share some advantages such as low cost, large availability, rapid bedside performance, and adaptability to local facilities (e.g., triage, low, high intensity, isolation ED areas, and hospital wards). However, both LUS and CXR are unsuitable per se for conclusive diagnosis. [5] [6] [7] [8] In this challenging scenario, physicians' gestalt is the pivot player orchestrating and integrating the other diagnostic tools. The aim of this study was to formally evaluate the diagnostic testing characteristics of physicians' gestalt for COVID-19, within a standard ED diagnostic pipeline.

This was an observational, prospective bicentric diagnostic study approved by the ethical committees of the involved hospitals.

Informed consent was obtained from study patients.

Consecutive patients presenting to the EDs of two Italian university hospitals (one in northern and one in central Italy) were enrolled from April 1 to April 30, 2020. Inclusion criteria were: 1) age ≥ 18 years; 2) presence of any sign or symptom for COVID-19 (declared or found at ED triage) among cough, fever (temperature > 37.5°C), pharyngodynia, dyspnea, oxygen saturation ≤ 94% in room air, respiratory rate ≥ 20 breaths/min, need of additional oxygen or ventilation, close contact with a suspected or confirmed COVID-19 case; and 3) first ED evaluation during the study period. Exclusion criteria were: 1) known diagnosis of COVID-19, 2) patient's loss at follow-up after ED discharge with a negative RT-PCR test result, and 3) patient's refusal to participate in the study.

Since February 2020, patients with predefined risk factors for COVID-19 detected by a pretriage nurse using standardized assessment were assigned to a dedicated ED area. Study patients were evaluated by one or more ED physicians (staff or in training physicians). Before the study was started, participating physicians were exposed to COVID-19 cases for more than 30 days.

All physicians were instructed regarding COVID-19 signs and symptoms and LUS and CXR signs. RT-PCR for SARS-CoV-2 on a nasopharyngeal swab was performed in all cases of suspected COVID-19. The results of RT-PCR were available on average within 6 hours from the request. Bedside diagnostic imaging (LUS and CXR) was performed at the discretion of the physicians, with CXR representing the standard reference imaging modality per local protocol.

Lung ultrasound was performed by the physician after his clinical evaluation, with the aim of detecting unilateral or bilateral interstitial syndrome, consolidations, and pleural effusion. All the above-mentioned findings were defined and diagnosed according to international recommendations on point-of-care LUS. 9 All physicians performing LUS had completed an ultrasound training course accredited by the Italian Society of Emergency Medicine or a similar hospital course and had performed at least 40 LUS examinations before starting the study. 10 CXR was performed by a radiology technician at the patient's bedside and interpreted by physicians. If both LUS and CXR were executed, LUS was performed before CXR.

The physicians were asked to dichotomize patients in COVID-19 likely or unlikely twice during the index visit. The first gestalt adjudication (clinical gestalt [CG]) was made immediately after a medical visit encompassing full medical history, vital sign assessment (body temperature, blood pressure, heart rate, oxygen saturation, respiratory rate), physical examination, electrocardiogram, and blood gas analysis (at the physician's discretion). The second gestalt adjudication, indicated as clinical and bedside imaging-integrated gestalt (CBIIG), was made after availability of bedside imaging with LUS or CXR. In a subgroup of patients, independent gestalt adjudication by a second physician was used to evaluate interobserver agreement. In case of discordant adjudication among the two experts, a third expert adjudicated the final diagnosis.

We aimed at including enough patients to provide accurate estimates of the physician's gestalt for COVID-19. In the absence of previous data, the study was powered to test the null hypothesis that the diagnostic accuracy of the physicians gestalt for COVID-19 is 80%, as previously reported for pneumonia, and exceeds 70%, assuming a prevalence of 25% of COVID-19 among suspected patients. 11 Using a type I error of 0.025 (α-level, two-sided) and type II error (β-level) of 0.1, we estimated that at least 468 needed patients to be included. Figure 2 , a Fagan nomogram represents the effect of RT-PCR plus CG or CBIIG on the posttest probability of COVID-19.

In this study, CG showed fair diagnostic accuracy for COVID-19. imaging, thus producing some degree of differential verification bias.

Fourth, the study was conducted during the first COVID-19 wave in Italy, in a period of strict national lockdown. In this phase, total ED visits substantially declined, visits for alternative diagnoses (e.g., non-COVID-19 lung diseases, decompensated heart failure) underwent major reduction, and COVID-19 was a key ED diagnosis.

Therefore, the diagnostic accuracy of CG and bedside imaging may change in a context of lower COVID-19 prevalence, usual ED overcrowding, and wider case mix. Finally, few patients lost at follow-up were excluded from further analyses, thus potentially introducing a small selection bias.

Physician's gestalt, especially when informed by both clinical and bedside imaging data, is key for the ED diagnostic pipeline of suspected COVID-19. When associated with reverse transcription-polymerase chain reaction, clinical and bedside imaging-integrated gestalt allows sensitive rule out, outperforming reverse transcription-polymerase chain reaction testing alone and potentially matching alternative strategies such as multiple reverse transcription-polymerase chain reaction testing or reverse transcription-polymerase chain reaction plus chest computed tomography.

The authors have no potential conflicts to disclose. 

Diagnosing COVID-19 in the emergency department: a scoping review of clinical examinations, laboratory tests, imaging accuracy, and biases

False negative tests for SARS-CoV-2 infection -challenges and implications

False-negative results of initial RT-PCR assays for COVID-19: a systematic review

COVID-19 testing: the threat of false-negative results

Multi-organ point-of-care ultrasound for COVID-19 (PoCUS4COVID): international expert consensus

Lung ultrasound in the emergency department -a valuable tool in the management of patients presenting with respiratory symptoms during the SARS-CoV-2 pandemic

Chest X-ray in new coronavirus disease 2019 (COVID-19) infection: findings and correlation with clinical outcome

Diagnostic impact of bedside chest X-ray features of 2019 novel coronavirus in the routine admission at the emergency department: case series from Lombardy region

International evidencebased recommendations for point-of-care lung ultrasound

Clinical gestalt to diagnose pneumonia, sinusitis, and pharyngitis: a meta-analysis

Comparing the areas under two or more correlated receiver operating characteristic curves: a nonparametric approach

The measurement of observer agreement for categorical data

Can emergency physician gestalt "rule in" or "rule out" acute coronary syndrome: validation in a multicenter prospective diagnostic cohort study

Predictive criteria for acute heart failure in emergency department patients with acute dyspnea: the PREDICA study

Comparison of the unstructured clinician gestalt, the Wells score, and the revised Geneva score to estimate pretest probability for suspected pulmonary embolism

Use of radiographic features in COVID-19 diagnosis: challenges and perspectives

Frequency and distribution of chest radiographic findings in COVID-19 positive patients

Imaging features of coronavirus disease 2019 (COVID-19): evaluation on thin-section CT

Clinical course and outcomes of patients with severe acute respiratory syndrome coronavirus 2 infection: a preliminary report of the first 28 patients from the Korean Cohort Study on COVID-19

Imaging profile of the COVID-19 infection: radiologic findings and literature review

Artificial intelligence applied on chest X-ray can aid in the diagnosis of COVID-19 infection: a first experience from Lombardy

Point-of-care lung ultrasound is more sensitive than chest radiograph for evaluation of COVID-19

Accuracy of emergency department clinical findings for diagnosis of coronavirus disease 2019

Comparison of nasopharyngeal and oropharyngeal swabs for SARS-CoV-2 detection in 353 patients received tests with both specimens simultaneously

Application and optimization of RT-PCR in diagnosis of SARS-CoV-2 infection

Sensitivity of chest CT for COVID-19: comparison to RT-PCR

Diagnosis of the coronavirus disease (COVID-19): RT-PCR or CT?

A British Society of Thoracic Imaging statement: considerations in designing local imaging diagnostic algorithms for the COVID-19 pandemic

Managing high clinical suspicion COVID-19 inpatients with negative RT-PCR: a pragmatic and limited role for thoracic CT

Diagnostic test evaluation methodology: a systematic review of methods employed to evaluate diagnostic tests in the absence of gold standard -An update

Diagnostic accuracy of physician's gestalt in suspected COVID-19: Prospective bicentric study

Matteo Versan g , Fabio Steri f , Arianna Ardito f , Marialessia Capuano g , Maria Gelardi g , Giulia Silvestri g

Denise Baricocchi g , Alessia Beaux g , Beaux Valentina g