Comparison of the accuracy of four diagnostic prediction rules for pulmonary embolism in patients admitted to the emergency department.

INTRODUCTION AND OBJECTIVES: Ruling out pulmonary embolism (PE) through a combination of clinical assessment and D-dimer level can potentially avoid excessive use of computed tomography pulmonary angiography (CTPA). We aimed to compare the diagnostic accuracy of the standard approach based on the Wells and Geneva scores combined with a standard D-dimer cut-off (500 ng/ml), with three alternative strategies (age-adjusted and the YEARS and PEGeD algorithms) in patients admitted to the emergency department (ED) with suspected PE. METHODS: Consecutive outpatients admitted to the ED who underwent CTPA due to suspected PE were retrospectively assessed. Sensitivity, specificity, positive and negative predictive values, likelihood ratios and diagnostic odds ratios were calculated and compared between the different diagnostic prediction rules. RESULTS: We included 1402 patients (mean age 69±18 years, 54% female), and PE was confirmed in 25%. Compared to the standard approach (p<0.001), an age-adjusted strategy increased specificity with a non-significant decrease in sensitivity only in patients older than 70 years. Compared to the standard and age-adjusted approaches, the YEARS and PEGeD algorithms had the highest specificity across all ages, but were associated with a significant decrease in sensitivity (p<0.001), particularly in patients aged under 60 years (sensitivity of 81% in patients aged between 51 and 60 years). CONCLUSION: Compared to the standard approach, all algorithms were associated with increased specificity. The age-adjusted strategy was the only one not associated with a significant decrease in sensitivity compared to the standard approach, enabling CTPA requests to be reduced safely.

Artificial intelligence-based diagnosis of acute pulmonary embolism: Development of a machine learning model using 12-lead electrocardiogram.

INTRODUCTION: Pulmonary embolism (PE) is a life-threatening condition, in which diagnostic uncertainty remains high given the lack of specificity in clinical presentation. It requires confirmation by computed tomography pulmonary angiography (CTPA). Electrocardiography (ECG) signals can be detected by artificial intelligence (AI) with precision. The purpose of this study was to develop an AI model for predicting PE using a 12-lead ECG. METHODS: We extracted 1014 ECGs from patients admitted to the emergency department who underwent CTPA due to suspected PE: 911 ECGs were used for development of the AI model and 103 ECGs for validation. An AI algorithm based on an ensemble neural network was developed. The performance of the AI model was compared against the guideline recommended clinical prediction rules for PE (Wells and Geneva scores combined with a standard D-dimer cut-off of 500 ng/mL and an age-adjusted cut-off, PEGeD and YEARS algorithm). RESULTS: The AI model achieves greater specificity to detect PE than the commonly used clinical prediction rules. The AI model shown a specificity of 100% (95% confidence interval (CI): 94-100) and a sensitivity of 50% (95% CI: 33-67). The AI model performed significantly better than the other models (area under the curve 0.75; 95% CI 0.66-0.82; p<0.001), which had nearly no discriminative power. The incidence of typical PE ECG features was similar in patients with and without PE. CONCLUSION: We developed and validated a deep learning-based AI model for PE diagnosis using a 12-lead ECG and it demonstrated high specificity.

The value of multiparametric prediction scores in heart failure varies with the type of follow-up after discharge: a comparative analysis.

AIMS: Multiple prediction score models have been validated to predict major adverse events in patients with heart failure. However, these scores do not include variables related to the type of follow-up. This study aimed to evaluate the impact of a protocol-based follow-up programme of patients with heart failure regarding scores accuracy for predicting hospitalizations and mortality occurring during the first year after hospital discharge. METHODS AND RESULTS: Data from two heart failure populations were collected: one composed of patients included in a protocol-based follow-up programme after an index hospitalization for acute heart failure and a second one-the control group-composed of patients not included in a multidisciplinary HF management programme after discharge. For each patient, the risk of hospitalization and/or mortality within a period of 12 months after discharge was calculated using four different scores: BCN Bio-HF Calculator, COACH Risk Engine, MAGGIC Risk Calculator, and Seattle Heart Failure Model. The accuracy of each score was established using the area under the receiver operating characteristic curve (AUC), calibration graphs, and discordance calculation. AUC comparison was established by the DeLong method. The protocol-based follow-up programme group included 56 patients, and the control group, 106 patients, with no significant differences between groups (median age: 67 years vs. 68.4 years; male sex: 58% vs. 55%; median ejection fraction: 28.2% vs. 30.5%; functional class II: 60.7% vs. 56.2%, I: 30.4% vs. 31.9%; P = not significant). Hospitalization and mortality rates were significantly lower in the protocol-based follow-up programme group (21.4% vs. 54.7%; P < 0.001 and 5.4% vs. 17.9%; P < 0.001, respectively). When applied to the control group, COACH Risk Engine and BCN Bio-HF Calculator had, respectively, good (AUC: 0.835) and reasonable (AUC: 0.712) accuracy to predict hospitalization. There was a significant reduction of COACH Risk Engine accuracy (AUC: 0.572; P = 0.011) and a non-significant accuracy reduction of BCN Bio-HF Calculator (AUC: 0.536; P = 0.1) when applied to the protocol-based follow-up programme group. All scores showed good accuracy to predict 1 year mortality (AUC: 0.863, 0.87, 0.818, and 0.82, respectively) when applied to the control group. However, when applied to the protocol-based follow-up programme group, a significant predictive accuracy reduction of COACH Risk Engine, BCN Bio-HF Calculator, and MAGGIC Risk Calculator (AUC: 0.366, 0.642, and 0.277, P < 0.001, 0.002, and <0.001, respectively) was observed. Seattle Heart Failure Model had non-significant reduction in its acuity (AUC: 0.597; P = 0.24). CONCLUSIONS: The accuracy of the aforementioned scores to predict major events in patients with heart failure is significantly reduced when they are applied to patients included in a multidisciplinary heart failure management programme.

Prognostic Accuracy of the Modified CHA2DS2-VASc Score in COVID-19 Patients Admitted to the Emergency Department Due to Clinical Worsening.

INTRODUCTION: Risk factors comprising the CHA2DS2VASc score are recognized as risk factors for venous thromboembolism and mortality in COVID-19 patients. A modified CHA2DS2VASc score (M-CHA2D2VASc), developed by changing gender criteria from female to male, has been proposed to predict in-hospital mortality in COVID-19 patients. The aim of this study was to evaluate the prognostic accuracy of M-CHA2D2VASc for adverse clinical outcomes and short-term mortality in COVID-19 patients admitted to the Emergency Department. MATERIAL AND METHODS: Retrospective study of patients admitted to the ED who underwent computed tomography pulmonary angiography due to suspected pulmonary embolism or clinical worsening. Patients were stratified into three M-CHA2DS2-VASc risk-categories: low (0 - 1 points), intermediate (2 - 3 points) and high-risk (≥ 4 points). RESULTS: We included 300 patients (median age 71 years, 59% male). The overall mortality was 27%. The M-CHA2DS2-VASc score was higher in non-survivors compared to survivors [4 (IQR:3 - 5) vs 2 (IQR: 1 - 4), respectively, p < 0.001). The M-CHA2DS2-VASc score was identified as an independent predictor of mortality in a multivariable logistic regression model (OR 1.406, p = 0.007). The Kaplan-Meier survival curves showed that the M-CHA2DS2-VASc score was associated with short-term mortality (log-rank test < 0.001), regardless of hospitalization (log-rank test p < 0.001 and p = 0.007, respectively). The survival proportion was 92%, 80% and 63% in the lower, intermediate, and higher risk-groups. As for the risk-categories, no difference was found in pulmonary embolism, Intensive Care Unit admission, and invasive mechanical ventilation. DISCUSSION: This is the first study to validate M-CHA2DS2-VASc score as a predictor of short-term mortality in patients admitted to the Emergency Department. CONCLUSION: The M-CHA2DS2-VASC score might be useful for prompt risk-stratification in COVID-19 patients during admission to the Emergency Department.

Introdução: O score CHA2DS2VASc engloba variáveis reconhecidas como fatores de risco para tromboembolismo venoso e mortalidade nos doentes com COVID-19. O score CHA2DS2VASc modificado (M-CHA2DS2-VASc), criado pela alteração do critério de género de feminino para masculino, foi proposto como preditor da mortalidade intra-hospitalar nestes doentes. O objetivo deste trabalho foi avaliar o valor prognóstico do M-CHA2DS2-VASc como preditor de eventos adversos e mortalidade a curto-prazo nos doentes com COVID-19 admitidos no Serviço de Urgência. Material e Métodos: Análise retrospetiva de doentes admitidos no Serviço de Urgência que realizaram tomografia computorizada pulmonar com administração de contraste por agravamento clínico e/ou suspeita de embolia pulmonar. Definiram-se três categorias de risco M-CHA2DS2-VASc: baixo, intermédio e alto (0 - 1; 2 - 3 e ≥ 4 pontos, respectivamente). Resultados: Incluíram-se 300 doentes (idade mediana: 71 anos, 59% homens). A mortalidade global foi 27%. O M-CHA2DS2-VASc foi maior em não sobreviventes [4 (IQR: 3 - 5) vs 2 (IQR: 1 - 4), p < 0,001) e constituiu um preditor independente de mortalidade numa análise multiparamétrica (OR: 1.406, p = 0,007). As curvas de sobrevivência demonstraram a associação do M-CHA2DS2-VASc com a mortalidade a curto-prazo (log-rank test < 0,001), independentemente dos doentes serem hospitalizados ou não (log-rank test p < 0,001 e p = 0,007, respetivamente). A taxa de sobrevida foi de 92%, 80% e 63% nos grupos de baixo, intermédio e alto risco. De acordo com as categorias de risco, não foram encontradas diferenças na incidência de embolia pulmonar, admissão em Cuidados Intensivos e ventilação mecânica invasiva. Discussão: Este é o primeiro estudo a validar o M-CHA2DS2-VASc como preditor de mortalidade a curto prazo na admissão no Serviço de Urgência. Conclusão: O M-CHA2DS2-VASc pode ser útil para estratificação de risco nos doentes com COVID-19 admitidos no Serviço de Urgência.