Development and Validation of Prediction Model to Estimate Risk of Acute Pulmonary Embolism: Retrospective Cohort Study

doi:10.21203/rs.3.rs-628946/v1

Background: Venous thromboembolism (VTE), clinically presenting as deep vein thrombosis (DVT) or pulmonary embolism (PE). Not all DVT patients carry the same risk of developing acute pulmonary embolism (APE).

Methods: Patients who were diagnosed as symptomatic VTE between June 2013 and June 2018 (n=1582) at the First Affiliated Hospital of Xi’an Jiaotong University were enrolled in this study, of whom 783 had the DVT and APE confirmed by angiography or computed tomography pulmonary angiography (CTPA). Using the LASSO and logistics regression, we derived predictive model with 16 candidate variables to predict the risk of APE and completed internal validation.

Results: Overall, 52.9% patients had DVT+APE (773 vs 1460), 47.1% patients only had DVT (687 vs 1460). The APE risk prediction model included one pre-existing disease or condition (respiratory failure), one risk factors(infection), three symptoms (dyspnea, hemoptysis, syncope), five signs (skin cold clammy, tachycardia, diminished respiration, pulmonary rales, accentuation/splitting of P₂, tricuspid murmur), and six ECG indicators (S_ⅠQ_ⅢT_Ⅲ, right axis deviation, left axis deviation, S₁S₂S₃, T wave inversion, Q/q wave), of which all were positively associated with APE. The ROC curves of the model showed AUC of 0.79 (95% CI, 0.77-0.82) and 0.80 (95% CI, 0.76-0.84) in the training set and testing set. The model showed good predictive accuracy (calibration slope, 0.83 and Brier score, 0.18).

Conclusion: Based on a retrospective single-center population study, we developed a novel prediction model to identify patients with different risks for APE in DVT patients, which may be useful for quickly estimating the probability of APE before obtaining definitive test results and speeding up emergency management processes.

Nuclear Medicine & Medical Imaging

acute pulmonary embolism

predictive model

clinical assessment

Pulmonary embolism (PE) is a common emergency and critical illness in clinical practice, with a sudden dramatic onset and often results in poor outcomes. Clinical evidence shows that about 90% of PE originates from the crumbling away and migration of deep vein thrombosis (DVT), both are collectively referred to venous thromboembolism (VTE), it represents globally the third most frequent acute cardiovascular syndrome, behind myocardial infarction and stroke¹. Acute pulmonary embolism (APE) is the most serious clinical type of VTE, which means a common complication of hospitalized patients. Because of occult onset and nonspecific symptoms, APE is usually ignored, which is an important cause of unexpected death and perioperative death of hospitalized patients, and also the main cause of increased medical expenses, prolonged hospital stay and medical disputes^1–3. Therefore, how to achieve early identification and diagnosis, timely and effective treatment, standardized follow-up and management to reduce the mortality and recurrence rate of pulmonary embolism patients and improve the prognosis is a major health problem facing China even the world.

The standard diagnosis of PE mainly includes pulmonary angiography and computed tomography pulmonary angiography (CTPA). Due to the contraindications of contrast imaging in the tested patients, the lack of relevant examination equipment in primary hospitals⁴ and the high cost of examination⁵, it is difficult to widely apply the diagnosis examination in clinical work, other relatively economical and accurate examination means are difficult to be implemented quickly. The results of prospective studies and guidelines lend support to the concept that clinical probability assessment is a fundamental step in the diagnosis of pulmonary embolism^1,6,7. The pulmonary embolism risk assessment scale recommended by the current guidelines mainly includes the Wells score⁸ and the revised Geneva score⁹. In the Chinese population, the diagnostic value of the Wells scores and revised Geneva score still needs to be verified by multi-center, prospective validation studies in a large cohort. Although some studies demonstrated the usefulness of these traditional scores for identifying suspected patients at risk of developing PE, they are not focused on accurately estimating risk in DVT patients.

Our study was aim to develop and validate a predictive model for developing risk of APE in DVT patients, using clinically variables that are readily available in primary hospitals and at the time of thrombotic events. To enhance visual presentation and facilitate subsequent clinical applications, heatmap shows the distribution of the all sample's predictor variables, generated nomograph provides a quick visual technique to assess the clinical probability of acute pulmonary embolism, which can direct personalised decision-making for preventative therapy.

Study design and data source

Figure 1 illustrates the workflow. These patients were from the First Affiliated Hospital of Xi’an Jiaotong University (Xi 'an, China) for acute symptomatic VTE between June 1, 2013 and June 1, 2018. A total of 1,460 patients was determined as the required sample size to be enrolled in the study. Approximately 30% of patients refered from inpatient or emergency departments to peripheral vascular department; about 70% came from peripheral vascular department. Each patient was examined at baseline according to a standardized protocol, following recommended international standards. Diagnostic procedures carried out at our institution dedicated diagnostic unit, including pulmonary angiography or CTPA, and compression venous ultrasonography. Angiography was performed after obtaining written informed consent from the patients. Exclusion criteria were: 1) recurrent pulmonary embolism, 2) incomplete clinical data, 3) contraindication to CTPA/ angiography, 4) The patient refuses to complete diagnostic test. This research was approved by the Ethics Committee of the First Affiliated Hospital of Xi’an Jiaotong University (Approval No. XJTU1AF2018LSK-144).

Predictor variables

In a first step, we searched PubMed and Web of Science databases without language or time restrictions to retrieve relevant studies. The prediction factors were mainly derived from 2019 ESC Guidelines for the diagnosis and management of acute pulmonary embolism¹ and a systematic review and meta-analysis that was designed to identify factors for VTE in hospitalized medical patients¹⁰. To maximize safety and model usability, we tend to choose reasonable and clinically relevant predictors that are easily available in medical hospital, especially for primary hospital. Because of some biochemical tests are not routinely available, we did not consider biomarkers endorsed by guidelines (D-dimer or pro-BNP). Age as one of continuous variables, was transformed into binary variables using pre-specified cut-offs either (>65 years and <65 years) derived from literature⁹. Meta-analysis found low-certainty evidence of association between risk of any VTE and CVC use¹⁰, we did not choose central venous cathete (CVC) use, this risk factor is less common in our sample population. There is probably an association between risk of any VTE and elevated heart rate(>100 beats per minute), therefore, we selected tachycardia (>100 beats per minute) and heart rate (as continuous variables).

From the medical records and hospital chart review, trained study doctors recorded all clinical and electrocardiogram data on a standard form, including demographic factors and clinical diagnoses. At the time of diagnosis, all the eligible cases underwent by trained clinical doctor to determine the presence or absence of signs and symptoms related to VTE, as dichotomous variables (yes/no), including dyspnea, hemoptysis, chest pain, syncope, swelling pain in the lower limbs and so on. The doctor should be careful to identify potential factors associated with APE and exclude pre-existing medical history, that are similar with the clinical manifestations of pulmonary embolism.

Based on literature and research reports, we screened more than 10 kinds of electrocardiogram sign associated with APE^11–15. The ECGs obtained within the first 24 hours of hospital admission were included in the study. Patients with acute cor pulmonale were deemed present if we identified at least one of the following: 1) S_ⅠQ_ⅢT_Ⅲ, 2) T-wave inversion in right precordial leads, 3) S₁S₂S₃, 4) pseudo infarction, 5) transient right bundle branch block. If the above signs had appeared in the past, they would be excluded.

Outcome variables

The primary outcomes of this study were as follows:1) an easy-to-use predictive model for acute pulmonary embolism was derived and validated, 2) a reasonable pipeline of disease risk prediction and factor analysis was introduced. All patients had a complete diagnostic examination for a definite diagnosis. Pulmonary embolism was diagnosed by pulmonary angiography or CTPA, and deep vein thrombosis was diagnosed by compression venous ultrasonography.

Derivation and validation of the models

The initial cohort comprised 1,582 symptomatic VTE patients. 36 patients were excluded due to incomplete data, 86 were excluded due to acute pulmonary embolism only, hence, 1,460 patients (DVT + APE vs DVT 773:687) were included in this study. Then, we randomly classified samples as training set (1,095) and testing set (365) in a 3:1 ratio. The training set was used to generate the prediction model, and testing set was used to evaluate the prediction performance of the model. Firstly, we performed univariate analysis to select predictor variables those significantly linked with APE diagnosis, using a cutoff of p < 0.05. To avoid overfitting, LASSO regression analysis was used to screened those APE diagnostic-related variables. Later, all APE diagnostic-related predictor variables were included in the multivariate analysis to assess independent predictor factors using logistics regression. Ultimately, we constructed sixteen APE diagnostic-related predictors as candidates for the prediction model. The area under the receiver operator characteristics (AUC) curves was used to evaluate the diagnostic efficiency of the model. Based on the AUC,

Brier score and calibration curves were used to evaluate the concordance between predicted diagnosis outcomes in training set and testing set. The prediction model distribution of patients in different risk levels, the number of censored patients, and the heatmap of APE diagnostic-related predictors were displayed. Establishment of the nomogram based on independent risk factors resulting from multivariate logistics regression to predict the APE probabilities for patients with DVT.

Handling of missing data

Except for age and gender, there were very few missing data for all variables. We eliminated the missing variables and analyzed the complete data.

Statistical analysis

The statistical analysis was performed in R software (version 4.1). p < 0.05 was regarded as statistically significant.

Study population

Baseline characteristics according to risk groups are shown in Table 1. Among the analyzed patients, 703 patients (48.1%) were males, 757 patients (51.9%) were females. Median age for 687 patients with DVT only was 59 years (interquartile range [IQR], 48–68), and median age among 773 patients with DVT and PE was 62 (interquartile range [IQR], 51–70). The overall prevalence of pulmonary embolism was 52.9% (773 of 1460 patients).

Table 1

Demographic and clinical characteristics of the patients in the training and testing set
Characteristic n (%) or median (IQR)	DVT	DVT + APE
Characteristic n (%) or median (IQR)	n = 687	n = 773
Sex
Male	330 (48)	373 (48)
Female	357 (52)	400 (52)
Age	59 (48, 68)	62 (51, 70)
Pre-existing disease or condition
Heart failure	21 (3.1)	19 (2.5)
Respiratory failure	1 (0.1)	31 (4.0)
Previous history of VTE	76 (11)	43 (5.6)
Autoimmune Disease	17 (2.5)	26 (3.4)
Malignant tumor	77 (11)	59 (7.6)
Risk factors
Fracture of lower limb	65 (9.5)	96 (12)
Severe trauma	32 (4.7)	36 (4.7)
Spinal cord injury	5 (0.7)	15 (1.9)
Arthroscopic operation	23 (3.3)	14 (1.8)
Blood transfusion	25 (3.6)	32 (4.1)
Hormone replacement therapy	22 (3.2)	19 (2.5)
Infection	27 (3.9)	96 (12)
Paralytic stroke	36 (5.2)	46 (6.0)
Superficial venous thrombosis	19 (2.8)	6 (0.8)
Postpartum period	21 (3.1)	14 (1.8)
Stay in bed (> 3 days)/Undergo surgery	181 (26)	210 (27)
Long time of sitting (> 6 hours)	81 (12)	29 (3.8)
Undergo hysteroscopy/Laparoscopy surgery	29 (4.2)	31 (4.0)
Laricose vein of lower limb	46 (6.7)	43 (5.6)
Smoke	192 (28)	195 (25)
Symptoms
Dyspnea	28 (4.1)	302 (39)
Hemoptysis	3 (0.4)	33 (4.3)
Chest pain	13 (1.9)	90 (12)
Swelling and pain in the lower limbs	657 (96)	593 (77)
Fever	32 (4.7)	52 (6.7)
Syncope	8 (1.2)	99 (13)
Cough	33 (4.8)	74 (9.6)
Palpitation	5 (0.7)	31 (4.0)
Delirium/Disturbance of consciousness	1 (0.1)	5 (0.6)
Signs
Skin cold clammy	7 (1.0)	27 (3.5)
Cyanosis of the lips	1 (0.1)	19 (2.5)
Tachycardia	36 (5.2)	109 (14)
Diminished respiration	1 (0.1)	43 (5.6)
Pulmonary rales	7 (1.0)	78 (10)
Accentuation/Splitting of P₂	100 (15)	178 (23)
Distention of jugular vein/Hepatojugular reflex	2 (0.3)	8 (1.0)
ECG
Heart rate	78 (69, 89)	82 (72, 94)
S_ⅠQ_ⅢT_Ⅲ	21 (3.1)	137 (18)
Nodal tachycardia	54 (7.9)	103 (13)
Right ventricular hypertrophy	0 (0)	13 (1.7)
Right axis deviation	4 (0.6)	19 (2.5)
Left axis deviation	55 (8.0)	176 (23)
S₁S₂S₃	2 (0.3)	41 (5.3)
Low voltage	17 (2.5)	35 (4.5)
Clockwise rotation of cardiac electric axis	1 (0.1)	9 (1.2)
ST-segment elevation	10 (1.5)	14 (1.8)
ST-segment depression	33 (4.8)	91 (12)
T wave inversion(V₁-V₃/V₄)	34 (4.9)	175 (23)
ST-segment depression(Ⅱ/Ⅲ/aVF)	14 (2.0)	71 (9.2)
Q/q wave(Ⅱ/aVF)	17 (2.5)	74 (9.6)
T wave inversion(Ⅱ/aVF)	6 (0.9)	56 (7.2)
Right bundle branch block	25 (3.6)	50 (6.5)
ECG: electrocardiogram; APE: acute pulmonary embolism; DVT: deep vein thrombosis; VTE, venous thromboembolism; P₂: pulmonary valve second heart sound.

Model development

A total of 54 variables were obtained from systematic review and meta-analysis, which has previously been reported to be associated with VTE. Univariate regression analysis was performed on 54 selected variables, we found that 34 variables were significantly linked with diagnosis of APE in DVT patients (p < 0.05). Lasso regression analysis and multivariate logistics regression analysis were adopted for the 34 APE diagnostic-related variables. Based on the results of the univariate analysis, 23 variables are included in the Lasso regression model (Fig. 2). After selecting the above 23 variables through multiple logistic regression again, 20 variables were independently associated with APE. We included 16 variables with OR value > 1 to build a prediction model, and named the model as APE risk prediction model (Table 2).

Table 2

The APE risk prediction model based on independent predictors of acute pulmonary embolism in training set
Characteristic	Coefficient	S.E.	OR	95% CI for OR		p-value
Characteristic	Coefficient	S.E.	OR	Lower	Upper	p-value
Pre-existing disease or condition
Respiratory failure	1.85	1.09	6.38	1.11	122.08	0.09
Risk factors
Infection	0.85	0.32	2.33	1.25	4.42	0.01
Symptoms
Dyspnea	2.14	0.26	8.52	5.22	14.50	< 0.00
Hemoptysis	1.36	0.68	3.89	1.16	17.90	0.05
Syncope	1.38	0.49	3.99	1.63	11.30	0.00
Signs
Skin cold clammy	0.69	0.55	2.00	0.71	6.19	0.20
Tachycardia	0.70	0.27	2.01	1.19	3.44	0.01
Diminished respiration	1.87	1.09	6.49	1.11	124.78	0.09
Pulmonary rales	0.93	0.59	2.53	0.87	9.29	0.12
Accentuation/Splitting of P₂	0.48	0.19	1.62	1.12	2.34	0.01
ECG
S_ⅠQ_ⅢT_Ⅲ	1.00	0.32	2.71	1.46	5.23	0.00
Right axis deviation	1.52	0.86	4.58	0.96	33.04	0.08
Left axis deviation	1.14	0.22	3.11	2.03	4.84	0.00
S₁S₂S₃	2.78	1.06	16.16	3.04	299.29	0.01
T wave inversion(V₁-V₃/V₄)	0.64	0.27	1.89	1.12	3.24	0.02
Q/q wave(Ⅱ/aVF)	0.91	0.40	2.49	1.16	5.63	0.02
Constant	-1.00	0.10	0.37	0.30	0.45	< 0.00
CI: confidence interval.

The APE risk prediction model included one pre-existing disease or condition(respiratory failure),one risk factors(infection), three symptoms(dyspnea, hemoptysis, syncope), five signs(skin cold clammy, tachycardia, diminished respiration, pulmonary rales, accentuation/splitting of P₂, tricuspid murmur), and six ECG indicators(S_ⅠQ_ⅢT_Ⅲ, right axis deviation, left axis deviation, S₁S₂S₃, T wave inversion, Q/q wave), of which all were positively associated with APE in DVT patients. The area under the ROC curve was 0.79 (95% CI, 0.77–0.82) (Fig. 3).

Internal validation

To validate the APE risk prediction model, we used an internal validation procedure based on random classify validation. The ROC curves of the model showed AUC of 0.79 (95% CI, 0.77–0.82) and 0.80 (95% CI, 0.76–0.84) in the training set and testing set, respectively, and no significant difference was found between these values, indicating the reliability of the nomogram (Fig. 3). This model had a Brier score of 0.18, calibration slope of 0.83, indicating good predictive accuracy performance (Fig. 3).

Model presentation

Since none of the prediction model performed well in all patients with APE, we try to derive a new predictive model which better identify patients at risk of deterioration. Our model had a good discriminatory power for APE in DVT patients (AUC, 0.79; 95% CI, 0.77–0.82). Heatmap showed that high-risk patients had more kinds of risk factors, which suggested that there were significant differences between the 16 diagnostic-related variables in high-risk and low-risk score APE patients (Fig. 4). To generate and validate an APE risk prediction model that could be translated to the clinic, we developed a nomogram to predict risk of APE in DVT patients (Fig. 5).

Using high-quality data from a retrospective cohort study, we derived an easy-to-use clinical score to predict the risk of developing APE in the DVT patients. The APE risk prediction model derived from a large cohort of consecutive inpatient with diagnostic examination, completely based on past medical history, clinical symptoms, physical signs, and the sign of the electrocardiogram. Sixteen clinical predictors accurately identified patients with high-risk disease who may benefit from individualized management to improve clinical outcomes. The good discriminatory power of our model was validated by internal validation.

We purposefully chose to use readily available predictors to increase clinical applicability and ease of use, especially for primary hospitals. In our study, sudden-onset dyspnea and hemoptysis are powerful predictors of pulmonary embolism, it is consistent with previous reports^16,17. Pulmonary embolism was identified in nearly one of every six patients hospitalized for a first episode of syncope¹⁸, therefore, syncope was selected as a predictive variable and was eventually included in the predictive model. Right ventricular dysfunction is associated with thrombotic load and one of the important prognostic factors of pulmonary embolism. In APE patients, there is always at least one ECG sign of right ventricular strain, including S_ⅠQ_ⅢT_Ⅲ, right bundle branch block and T wave inversions¹⁴. Our model included a total of 6 ECG sign, these indicators have previously been reported to be related to pulmonary embolism.

Why choose Electrocardiogram as a predictor of APE? Electrocardiogram is an irreplaceable examination method to explore and measure abnormal electrocardiogram activity. It has the advantages of non-invasive, timeliness and simple operation, and has become a necessary examination for patients with unexplained dyspnea or chest pain¹⁹. The changes of electrocardiogram frequency, rhythm and conduction in APE patients, throughout the disease course and during treatment phases, may better assess risk stratification, prognosis and outcome of the disease and hence the opportunity for more applicable and balanced targeted preventative strategies^13,20,21.

Is this prediction model clinically generalized? Firstly, as one of the most common clinical examination methods, electrocardiogram is often used to assist early screening of suspected patients. Typically, the sign of the electrocardiogram requires physicians to provide scientific and medical expertise, and the electrocardiographic abnormalities with acute cor pulmonale are well-defined criteria, which have been known and applied for many years²². Except electrocardiogram, all the data required for the prediction model are routinely collected in the context of suspected acute pulmonary embolism and are available from the patient’s history and physical examination. Since the model was derived from multidisciplinary patients, we believe that the prediction model is applicable to all clinical departments and easy to calculate.

Is this prediction model valid and accurate? In terms of prediction accuracy, all patients received a diagnosis by a gold standard criterion, our prediction model could be considered accurate for predicting pulmonary embolism and superior to other people's reports. The model displayed good discrimination in the training set and testing sets (area under ROC curve, 0.79 and 0.80, respectively).

In fact, our prediction model is extending and externally validating in multiple centers, preliminary experimental results prove the feasibility of our ideas, the ability of the model to distinguish patients’ risk for APE in the validation cohort was at least as good as in the original cohort. To facilitate clinical visualization management, instead of using points proportional to their beta regression coefficient values, we estimate the probability of acute pulmonary embolism directly from Nomogram.

There are potential limitations of our study. Firstly, the original intention of this model was to serve primary hospital and simplify the diagnosis process, so we did not include the biochemical indicators recommended by the guidelines, such as D-dimer, pro-BNP, etc. Secondly, as is often the case in clinical diagnostic studies, in our study,we did not account for the uncertainty around predictions, but focused on the clinical possibility assessment. Finally, although we retrieved a large number of patients from the high-volume center, this study assessed single-center cohorts retrospectively and was not a population-based study or nationwide survey. Hopefully, this model will be further validated in a large, multi-center, prospective validation study before providing benefits for Chinese patients.

In conclusion, this study reports the derivation and initial validation of a sixteen variable clinical prediction model that demonstrated good overall accuracy in predicting risk of acute pulmonary embolism for patients with deep vein thrombosis. The above means the prediction model appear more suitable for primary hospital. Pending external validation, this study now provides the basis and information for risk assessment of patients with acute pulmonary embolism.

Competing interests

The authors declare no competing interests

Author Contributions

YL and HT designed the study; YL, YM, BF, SX, MK, ZD and YC involved with patient care and data collection; LY, YM, BF and YW analyzed the patient samples and interpreted data; LY wrote the initial paper; YL, YM, BF and HT revised the paper; YL and HT had primary responsibility for final content. All authors read and approved the final manuscript.

Funding

This study was supported by the fund for key research and development program project of Shaanxi province, China (No.2017SF-254) and the Clinical Research Award of the First Affiliated Hospital of Xi’an Jiaotong University, China (No. XJTU1AF-CRF-2018e023).

Availability of data and materials

All data generated or analyzed during this study are included in this published article.

Ethics approval

This research was approved by the Ethics Committee of the First Affiliated Hospital of Xi’an Jiaotong University (Approval No. XJTU1AF2018LSK-144).

Konstantinides, S. V. et al. 2019 ESC Guidelines for the dignosis and management of acute pulmonary embolism development Toed in colaboration with the European Respiratory Society (ERS). European Heart Journal 41, 543-603, doi:10.1093/eurheartj/ehz405 (2020).
Di Nisio, M., van Es, N. & Bueller, H. R. Deep vein thrombosis and pulmonary embolism. Lancet 388, 3060-3073, doi:10.1016/s0140-6736(16)30514-1 (2016).
Raskob, G. E. et al. Thrombosis A Major Contributor to Global Disease Burden. Arteriosclerosis Thrombosis and Vascular Biology 34, 2363-2371, doi:10.1161/atvbaha.114.304488 (2014).
Oren, O., Kebebew, E. & Ioannidis, J. P. A. Curbing Unnecessary and Wasted Diagnostic Imaging. Jama-Journal of the American Medical Association 321, 245-246, doi:10.1001/jama.2018.20295 (2019).
Venkatesh, A. K. et al. Evaluation of Pulmonary Embolism in the Emergency Department and Consistency With a National Quality Measure Quantifying the Opportunity for Improvement. Archives of Internal Medicine 172, 1028-1032, doi:10.1001/archinternmed.2012.1804 (2012).
Mazzolai, L. et al. Diagnosis and management of acute deep vein thrombosis: a joint consensus document from the European Society of Cardiology working groups of aorta and peripheral vascular diseases and pulmonary circulation and right ventricular function. Eur Heart J 39, 4208-4218, doi:10.1093/eurheartj/ehx003 (2018).
Keller, K. et al. Trends in thrombolytic treatment and outcomes of acute pulmonary embolism in Germany. Eur Heart J 41, 522-529, doi:10.1093/eurheartj/ehz236 (2020).
Wells, P. S. et al. Derivation of a simple clinical model to categorize patients probability of pulmonary embolism: increasing the models utility with the SimpliRED D-dimer. Thrombosis and haemostasis 83, 416-420 (2000).
Le Gal, G. et al. Prediction of pulmonary embolism in the emergency department: the revised Geneva score. Annals of internal medicine 144, 165-171, doi:10.7326/0003-4819-144-3-200602070-00004 (2006).
Darzi, A. J. et al. Prognostic factors for VTE and bleeding in hospitalized medical patients: a systematic review and meta-analysis. Blood 135, 1788-1810, doi:10.1182/blood.2019003603 (2020).
Daniel, L. R., Courtney, D. M. & Kline, J. A. Assessment of cardiac stress from massive pulmonary embolism with 12-lead ECG. Chest 120, 474-481, doi:10.1378/chest.120.2.474 (2001).
Miniati, M. et al. Simple and accurate prediction of the clinical probability of pulmonary embolism. American Journal of Respiratory and Critical Care Medicine 178, 290-294, doi:10.1164/rccm.200802-207OC (2008).
Shopp, J. D., Stewart, L. K., Emmett, T. W. & Kline, J. A. Findings From 12-lead Electrocardiography That Predict Circulatory Shock From Pulmonary Embolism: Systematic Review and Meta-analysis. Academic Emergency Medicine 22, 1127-1137, doi:10.1111/acem.12769 (2015).
Bolt, L. et al. Prognostic Value of Electrocardiography in Elderly Patients with Acute Pulmonary Embolism. American Journal of Medicine 132, E835-E843, doi:10.1016/j.amjmed.2019.05.041 (2019).
Cagdas, M. et al. Diagnostic value of QRS and S wave variation in patients with suspicion of acute pulmonary embolism. American Journal of Emergency Medicine 36, 2197-2202, doi:10.1016/j.ajem.2018.03.074 (2018).
van der Hulle, T. et al. Simplified diagnostic management of suspected pulmonary embolism (the YEARS study): a prospective, multicentre, cohort study. Lancet 390, 289-297, doi:10.1016/s0140-6736(17)30885-1 (2017).
Righini, M., Robert-Ebadi, H. & Le Gal, G. Diagnosis of acute pulmonary embolism. Journal of thrombosis and haemostasis : JTH 15, 1251-1261, doi:10.1111/jth.13694 (2017).
Prandoni, P. et al. Prevalence of Pulmonary Embolism among Patients Hospitalized for Syncope. N. Engl. J. Med. 375, 1524-1531, doi:10.1056/NEJMoa1602172 (2016).
Hornick, J. & Costantini, O. The Electrocardiogram: Still a Useful Tool in the Primary Care Office. The Medical clinics of North America 103, 775-784, doi:10.1016/j.mcna.2019.04.003 (2019).
Stein, P. D. et al. ELECTROCARDIOGRAM IN ACUTE PULMONARY-EMBOLISM. Progress in Cardiovascular Diseases 17, 247-257, doi:10.1016/s0033-0620(75)80016-8 (1975).
Stein, P. D. et al. CLINICAL, LABORATORY, ROENTGENOGRAPHIC, AND ELECTROCARDIOGRAPHIC FINDINGS IN PATIENTS WITH ACUTE PULMONARY-EMBOLISM AND NO PREEXISTING CARDIAC OR PULMONARY-DISEASE. Chest 100, 598-603, doi:10.1378/chest.100.3.598 (1991).
Winsor, T. Electrocardiogram and pulmonary infarction: (acute cor pulmonale). Jama 204, 807 (1968).

No competing interests reported.

AppendixTable1.docx
Appendix Table 1 Univariate analysis of clinical predictors.
AppendixTable2.docx
Appendix Table 2 Predictiors with significant differences in training set and testing set.
AppendixTable3.docx
Appendix Table 3 Multiple Logistic regression to construct a predictive model of acute pulmonary embolism.
Rawdataset.et

Development and Validation of Prediction Model to Estimate Risk of Acute Pulmonary Embolism: Retrospective Cohort Study

Status:

Version 1

Abstract

Figures

1. Introduction

2. Methods

3. Results

4. Discussion

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1