Comments
Abstract
Rationale: Not all patients with acute pulmonary embolism (PE) have a high risk of an adverse short-term outcome.
Objectives: This prospective cohort study aimed to develop a multimarker prognostic model that accurately classifies normotensive patients with PE into low and high categories of risk of adverse medical outcomes.
Methods: The study enrolled 848 outpatients from the PROTECT (PROgnosTic valuE of Computed Tomography) study (derivation cohort) and 529 patients from the Prognostic Factors for Pulmonary Embolism (PREP) study (validation cohort). Investigators assessed study participants for a 30-day complicated course, defined as death from any cause, hemodynamic collapse, and/or adjudicated recurrent PE.
Measurements and Main Results: A complicated course occurred in 63 (7.4%) of the 848 normotensive patients with acute symptomatic PE in the derivation cohort and in 24 patients (4.5%) in the validation cohort. The final model included the simplified Pulmonary Embolism Severity Index, cardiac troponin I, brain natriuretic peptide, and lower limb ultrasound testing. The model performed similarly in the derivation (c-index of 0.75) and validation (c-index of 0.85) cohorts. The combination of the simplified Pulmonary Embolism Severity Index and brain natriuretic peptide testing showed a negative predictive value for a complicated course of 99.1 and 100% in the derivation and validation cohorts, respectively. The combination of all modalities had a positive predictive value for the prediction of a complicated course of 25.8% in the derivation cohort and 21.2% in the validation cohort.
Conclusions: For normotensive patients who have acute PE, we derived and validated a multimarker model that predicts all-cause mortality, hemodynamic collapse, and/or recurrent PE within the following 30 days.
Risk stratification of patients with pulmonary embolism (PE) may identify patients at high risk of early death who may benefit from more intensive surveillance or aggressive therapy. Alternatively, patients deemed low risk for early complications might be considered for partial or complete outpatient treatment of their PE.
The present study derived and validated a multimarker model that predicts all-cause mortality, hemodynamic collapse, and/or recurrent PE in normotensive patients diagnosed with acute symptomatic PE in an emergency department. The model, which includes clinical, echographic, and biochemical variables, may provide guidance for decision making in PE care.
Because patients with acute symptomatic pulmonary embolism (PE) who present with arterial hypotension or shock have a high risk of death, treatment guidelines recommend strong consideration of thrombolysis for such patients (1–3). For hemodynamically stable patients with PE, contemporary risk stratification should aim to differentiate the group of patients deemed as having a low risk for early complications (who might benefit from an abbreviated hospital stay or outpatient therapy) from the group of patients deemed as having a high risk for adverse clinical events (who might benefit from an escalation of therapy) (4).
Clinical prediction models (e.g., the Pulmonary Embolism Severity Index [PESI], and the simplified PESI [sPESI]) may accurately identify patients at low risk of short-term mortality after the diagnosis of PE, and such patients might benefit from an abbreviated hospital stay or outpatient therapy (5–9). However, studies have not elucidated if the addition of a cardiac biomarker assay to the clinical prediction model improves the prognostication and subsequently improves patient safety (10). For potentially low-risk patients admitted to the hospital with acute symptomatic PE, some clinicians typically order an echocardiogram and a lower extremity venous compression ultrasound before hospital discharge in an attempt to improve risk stratification, but the indications for their use require further study (11, 12). Alternatively, although some evidence suggests that a subset of high-risk normotensive patients with acute symptomatic PE may have a reasonable risk-to-benefit ratio for thrombolytic therapy, single markers of right ventricular dysfunction (e.g., echocardiography, spiral computed tomography, or brain natriuretic peptide testing [BNP]) and myocardial injury (e.g., cardiac troponin T or I testing) have an insufficient positive predictive value for PE-specific mortality to drive decision making toward such therapy (13). Thus, recommendations for outpatient treatment or thrombolytic therapy for normotensive patients with acute symptomatic PE necessitate further development of prognostic models and conduct of clinical trials that assess various treatment strategies (14).
The PROgnosTic valuE of Computed Tomography scan in hemodynamically stable patients with acute symptomatic pulmonary embolism (PROTECT) study was designed to prospectively assess the prognostic significance of multidetector computed tomographic pulmonary angiography (MCTPA) findings and other prognostic tools in normotensive patients with acute symptomatic PE (15). The purpose of this study was to derive a multimarker model for estimating risk in normotensive patients with acute symptomatic PE. We also propose a risk-stratification strategy for classification of normotensive patients with acute PE into low versus high categories of risk of adverse medical outcomes, and the eStiMaTe (score [S], markers [M], thrombus [T]) calculator may then be used to estimate individual risk for certain patients in whom a more precise estimate may affect clinical decision making. We also performed an external validation of the model in an independent cohort of normotensive outpatients with objectively confirmed acute symptomatic PE.
A detailed description of the methods is available in the online supplement.
This was a preplanned secondary analysis of the prospective, multicenter, observational PROTECT study (15). The institutional review board of each center approved the protocol and consent forms.
Patients underwent recruitment from the emergency department of five academic and seven general urban hospitals in Spain between January 1, 2009, and May 31, 2011.
The study required confirmation of the diagnosis of PE with a positive MCTPA (16), and it excluded patients who had hemodynamic instability at presentation (defined as cardiogenic shock, systolic blood pressure < 90 mm Hg, or use of inotropic support), treatment with thrombolytics at the time of PE diagnosis, life expectancy less than 3 months, pregnancy, geographic inaccessibility that precluded follow-up, and age younger than 18 years (15).
PROTECT a priori defined MCTPA-assessed right ventricular dysfunction (RVD) as a ratio of the right ventricle (RV) to the left ventricle short-axis diameters of greater than 0.9 (15, 17).
The study required that patients undergo transthoracic echocardiography within 24 hours after diagnosis of PE. The study defined echocardiographic RVD as the presence of at least two of the following: dilatation of the RV, hypokinesis of the RV free wall, and estimated systolic pulmonary artery pressure greater than 30 mm Hg (15, 18).
The protocol recommended that patients undergo complete bilateral lower limb venous compression ultrasonography (CCUS) testing in the 48 hours that followed the diagnosis of PE. Vein incompressibility was the sole diagnostic criterion for deep vein thrombosis (DVT).
Central laboratory personnel, blinded to the patients’ baseline characteristics and clinical outcome, measured cardiac troponin I (cTnI) levels (19) and BNP levels (20).
This study used a complicated course through 30 days after the diagnosis of PE as the primary endpoint. The study protocol defined a complicated course as death from any cause, hemodynamic collapse, or adjudicated recurrent PE (15). The Adjudication Committee, whose members were blinded to initial prognostic test results, adjudicated all serious adverse events.
The validation cohort for this study consisted of the subset of 529 normotensive outpatients enrolled in the Prognostic Factors for Pulmonary Embolism (PREP) study who had acute symptomatic PE and complete baseline and follow-up data required for this study (21).
The study used logistic regression to assess for an independent association between potential baseline predictors and complicated course in the derivation cohort. For the manual backward stepwise multivariable logistic regression model, we assessed variables that had a significance level of P < 0.1 in univariate analyses. The model retained variables associated with the outcome at a significance level of P < 0.05. Regression diagnostics were performed to assess model assumptions and the effects of outliers and influential cases. The collinearity of the full model was assessed with the variance inflation factor. We assessed performance of the model by evaluating discrimination through use of the overall c-statistic (22, 23). The study used the Brier score to quantify the overall accuracy of predictions (24) and used bootstrapping to calculate 95% bias-corrected bootstrap confidence intervals (CIs) for the c-index. Investigators evaluated model calibration with the modified Hosmer-Lemeshow Chi-square statistic (25).
We recursively partitioned the derivation cohort sample into progressively more homogeneous subgroups by sequentially using the predictor variables that discriminated between patients who had a complicated course and those who did not in the multivariate logistic regression model.
To assist clinicians with individualized estimation of adverse outcome for normotensive outpatients with acute symptomatic PE, we developed a risk calculator that we will make available at www.PEprognosis.org.
Study staff screened 999 consecutive outpatients with acute PE for eligibility. Hemodynamic instability excluded 42 (4.2%) patients from participation. Of the remaining 957 hemodynamically stable patients, the study excluded 6.8% (65 of 957 patients) because they did not have a technically adequate MCTPA (n = 28, 2.9%) or a transthoracic echocardiogram (n = 37, 3.9%). Other reasons for exclusion included anticipated unavailability for follow-up (n = 23) and refusal to give informed consent (n = 21). The ineligible patients had an even distribution of disease severity (measured by sPESI), and they did not differ significantly from the eligible cohort with respect to baseline characteristics. The remaining eligible 848 patients (416 men and 432 women) enrolled in the study (Figure 1).
Figure 1. Patient flow diagram. *Hemodynamic instability at presentation was defined as cardiogenic shock, systolic blood pressure < 90 mm Hg, or use of inotropic support. ED = emergency department; MCTPA = multidetector computed tomography; PE = pulmonary embolism; TTE = transthoracic echocardiography.
[More] [Minimize]Table 1 shows the patients’ clinical symptoms, predisposing conditions, and relevant findings at presentation. Overall, 533 patients (63%; 95% CI, 60–66%) had RVD detected by MCTPA, and 192 (23%; 95% CI, 20–25%) had echocardiographic RVD. On MCTPA evaluation, right-to-left ventricular diameter ratio ranged from 0.5 to 3.25 and had a median value of 1.0 (25th–75th percentile, 0.82–1.24). Patients from the validation cohort had younger age, less comorbidity (e.g., immobilization, chronic pulmonary disease), and fewer signs of clinical severity (e.g., syncope, tachycardia, hypoxemia, hypotension) compared with those from the PROTECT cohort (Table 1).
| Derivation Cohort (N = 848) | Validation Cohort (N = 529) | |
|---|---|---|
| Clinical characteristics | ||
| Age, yr, median (25th–75th percentiles) | 72 (59–80) | 68 (52–78) |
| Age > 80 yr | 184 (22) | 98 (19) |
| Male sex | 416 (49) | 246 (47) |
| Risk factors for VTE | ||
| Cancer* | 144 (17) | 77 (15) |
| Recent surgery† | 89 (10) | 59 (11) |
| Previous VTE | 121 (14) | 134 (25) |
| Immobilization‡ | 167 (20) | 18 (3) |
| Comorbid diseases | ||
| COPD | 108 (13) | 33 (6) |
| Congestive heart failure | 51 (6.0) | 25 (5) |
| Clinical symptoms and signs at presentation | ||
| Syncope | 128 (15) | 25 (5) |
| Chest pain | 410 (48) | 255 (48) |
| Dyspnea | 680 (80) | 423 (80) |
| Heart rate ≥ 110/min | 176 (21) | 68 (13) |
| SaO2 < 90% | 185 (22) | 35 (6.6) |
| SBP < 100 mm Hg | 34 (4.0) | 6 (1.1) |
| Simplified PESI (7) | ||
| Low-risk | 313 (37) | 265 (50) |
| High-risk | 535 (63) | 264 (50) |
| Echocardiography and cardiac biomarkers | ||
| RV dysfunction (echocardiogram) | 192 (23) | 37 (7) |
| BNP > 100 pg/ml | 380 (45) | 219 (41) |
| cTnI > 0.05 ng/ml | 139 (16) | 132 (25) |
| Treatment | ||
| Insertion of an IVC filter | 8 (0.9) | 22 (4.1) |
The study had complete primary outcome information for all patients at the end of the 30-day follow-up. A complicated course occurred in 63 of the 848 (7.4%; 95% CI, 5.7–9.2%) normotensive patients with acute symptomatic PE who entered into the study. Complicated course was due to death in 38 patients, hemodynamic collapse in 28 patients, and nonfatal symptomatic recurrent PE in 1 patient (Table 2). Of the 38 patients (4.5%; 95% CI, 3.1–5.9%) who died during the 30-day follow-up period, the Adjudication Committee considered death as “definitely” or “possibly” due to fatal PE in 11 of these patients (29.0% of the total deaths; 95% CI, 14.5–43.4%), and this corresponded to a cumulative rate of definite or possible fatal PE of 1.3% (95% CI, 0.5–2.1%) at 30 days after PE diagnosis. Cancer (29%; 11 of 38 patients), cardiopulmonary disease (24%; 9 of 38 patients), bleeding (7.9%; 3 of 38 patients), and other miscellaneous diseases (11%; 4 of 38 patients) caused the other deaths. Regarding other outcomes during the 30-day follow-up period, 3 of 848 patients (0.3%; 95% CI, 0–0.7%) had recurrent symptomatic PE (one nonfatal and two fatal), and 19 patients (2.2%; 95% CI, 1.2–3.2%) suffered a major bleeding episode.
| Events | Patients, n (%) |
|---|---|
| PROTECT derivation cohort, N = 848 | |
| Complicated course* | 63 (7.4) |
| All-cause death | 38 (4.5) |
| PE-related death | 11 (1.3) |
| Hemodynamic collapse† | 28 (3.0) |
| Systolic blood pressure < 90 mm Hg | 22 |
| Catecholamine administration | 10 |
| Emergency thrombolysis | 14 |
| Endotracheal intubation | 7 |
| Cardiopulmonary resuscitation | 3 |
| Nonfatal recurrent PE | 1 (0.1) |
| PREP validation cohort, N = 529 | |
| Complicated course* | 24 (4.5) |
| All-cause death | 14 (2.6) |
| PE-related death | 6 (1.1) |
| Hemodynamic collapse† | 14 (2.6) |
| Nonfatal recurrent PE | 4 (0.7) |
In univariable analyses, echocardiographic RVD (odds ratio [OR], 2.62; 95% CI, 1.54–4.45; P < 0.001), elevated cTnI (OR, 2.84; 95% CI, 1.62–4.98; P < 0.001), elevated BNP (OR, 3.21; 95% CI, 1.80–5.73; P < 0.001) objectively confirmed concomitant DVT (OR, 1.84; 95% CI, 1.08–3.12; P = 0.02), a high-risk sPESI (OR, 7.49; 95% CI, 2.97–18.88; P < 0.001), or immobilization (OR, 2.02; 95% CI, 1.15–3.53; P = 0.01) at the time of acute PE diagnosis predicted a complicated course during follow-up (Table 3). The multivariable model found that elevated troponin level (ORadj, 1.96; 95% CI, 1.06–3.63; P = 0.03), elevated BNP level (ORadj, 2.12; 95% CI, 1.13–3.99; P = 0.02), objectively confirmed concomitant DVT (ORadj, 2.08; 95% CI, 1.19–3.65; P = 0.01), and a high-risk sPESI (ORadj, 5.62; 95% CI, 2.19–14.45; P < 0.001) independently predicted a complicated course (Table 3). We found no significant interaction terms between the variables in the final model.
| Risk Factor | Unadjusted OR (95% CI) | P Value | Adjusted OR (95% CI) | P Value |
|---|---|---|---|---|
| Age, per yr | 1.01 (0.99–1.03) | 0.17 | — | — |
| Male sex | 0.82 (0.49–1.37) | 0.45 | — | — |
| Cardiac troponin I > 0.05 ng/ml | 2.84 (1.62–4.98) | <0.001 | 1.96 (1.06–3.63) | 0.03 |
| BNP > 100 pg/ml | 3.21 (1.80–5.73) | <0.001 | 2.12 (1.13–3.99) | 0.02 |
| RV dysfunction (echocardiogram) | 2.62 (1.54–4.45) | <0.001 | — | — |
| RV dysfunction (MCTPA) | 1.29 (0.75–2.24) | 0.36 | — | — |
| Presence of DVT by CCUS | 1.84 (1.08–3.12) | 0.02 | 2.08 (1.19–3.65) | 0.01 |
| Simplified PESI > 0 points | 7.49 (2.97–18.88) | <0.001 | 5.62 (2.19–14.45) | <0.001 |
| Dyspnea | 2.05 (0.92–4.59) | 0.08 | — | — |
| Chest pain | 0.73 (0.43–1.23) | 0.24 | — | — |
| Syncope | 0.93 (0.45–1.93) | 0.85 | — | — |
| Immobilization* | 2.02 (1.15–3.53) | 0.01 | — | — |
| Recent surgery | 0.56 (0.20–1.57) | 0.27 | — | — |
Analyses suggested that the final model had good predictive performance. For the total population of 848 patients, the receiver operating characteristic curve had an area under the curve of 0.75 (95% CI, 0.69–0.80), and the Brier score was 0.06 (Figure 2). Optimism, which is the tendency of the model to perform better with the data from which it was constructed than on new data, was low. The internally validated area under the curve and Brier score were 0.73 and 0.07, respectively. The final model was well calibrated (Hosmer-Lemeshow Chi-square statistic, 0.75; P value for the lack of fit = 0.99). When only all-cause mortality was considered, the receiver operating characteristic curve had an area under the curve of 0.76 (95% CI, 0.70–0.82).
Figure 2. Receiver operating characteristic curve for 30-day complicated course. Area under the curve (AUC) (c-statistic), 0.75; 95% confidence interval (CI), 0.69–0.80; P < 0.001. When we excluded the two patients who received thrombolytic therapy without another event qualifying for a complicated course, the predictive performance of the model for a complicated course was similar: AUC (c-statistic), 0.77; 95% CI, 0.71–0.82; P < 0.001. Complicated course defined as all-cause mortality or hemodynamic collapse or recurrent pulmonary embolism. Hemodynamic collapse was defined as at least one of the following: systolic blood pressure < 90 mm Hg for at least 15 minutes, need for catecholamine administration because of persistent arterial hypotension or shock, need for thrombolysis, need for endotracheal intubation, or need for cardiopulmonary resuscitation.
[More] [Minimize]On the basis of the predictor variables from the multivariate model, we used a recursive partitioning technique to develop a risk-stratification strategy for identification of low-risk and high-risk normotensive patients with acute PE. For identifying low-risk normotensive patients with acute PE, the program generated a model that included the sPESI and BNP levels on admission. By combining the prognostic information provided by these two variables, we found that of 216 low-risk patients (25% of the entire study population) with an sPESI of 0 points and BNP levels less than 100 pg/ml, only 2 (0.9%; 95% CI, 0–2.2%) experienced a complicated course within the first 30 days, and none of them died (Figure 3). Thus, the combination model showed an excellent prognostic sensitivity and negative predictive value of 97% (95% CI, 88–99%) and 99% (95% CI, 96–100%), respectively. None of the 12 patients (12 of 216; 5.6%; 95% CI, 2.5–8.6%) with echocardiographic RVD among those with a sPESI of 0 points and BNP levels less than 100 pg/ml had a complicated course within the first 30 days.
Figure 3. Recursive partitioning analysis tree for the study data set. *Variables forced into the model. Predicted probabilities for combinations not shown in the algorithm will be available at www.PEprognosis.org. BNP = brain natriuretic peptide; CCUS = complete lower limb compression ultrasound testing; DVT = deep vein thrombosis.
[More] [Minimize]For identification of high-risk normotensive patients with acute PE, the recursive partitioning technique included the four tests used in the predictive model (i.e., cTnI, BNP, sPESI, and CCUS) for prediction of a 30-day complicated course (Figure 3). The combination of all modalities had the highest positive predictive value for the prediction of a complicated course during follow-up (25.8%; 95% CI, 10.4–41.2%).
Predicted probabilities for combinations not shown in the manuscript are available at www.PEprognosis.org.
Of the 529 patients included in the PREP validation cohort, 24 (4.5%; 95% CI, 2.8–6.3%) patients experienced a complicated course during the first month of follow-up, compared with 7.4% (63 of 848 patients; 95% CI, 5.7–9.2%) in the PROTECT cohort. The predictive model had a c-index of 0.85 (95% CI, 0.77–0.93) in the external validation cohort. For all-cause mortality, the c-index was 0.83 (95% CI, 0.72–0.94). The combination of the sPESI and BNP levels on admission classified 36% (193 of 529) of patients in the PREP validation cohort as having low risk of an adverse event, and the overall 30-day complicated course of this group was 0%, compared with 7.1% (95% CI, 4.4–9.9%) in the high-risk group. In the PREP validation cohort, the model had a negative predictive value of 100%, and a negative likelihood ratio of 0. In the PREP external validation cohort, the combination of all prognostic tests had a positive predictive value of 21.2% (7 of 33; 21.2%; 95% CI, 9.0–38.9%) for the prediction of a complicated course during follow-up.
The present study derived and validated a multimarker prognostication that consists of sPESI, BNP, cTnI, and CCUS imaging for concomitant DVT for normotensive patients diagnosed with acute symptomatic PE in an emergency department. Although early mortality rates are very low in outpatients with hemodynamically stable PE, the majority of deaths from PE in absolute terms occur in this subset. The combination of a clinical score (i.e., sPESI) and a marker of ventricular dysfunction (i.e., BNP testing) showed an excellent prognostic ability for identification of low-risk patients with acute PE. This study also demonstrated that the combination of a clinical score (i.e., sPESI), markers of RV dysfunction (i.e., BNP) and myocardial injury (i.e., cTnI), and assessment of thrombotic burden (i.e., CCUS) improved identification of normotensive patients with acute symptomatic PE at high risk for a 30-day complicated course. We also developed the multimarker eStiMaTe calculator to predict complicated course in normotensive PE. We propose that the multimarker model be used as a simple risk “screening” method for patients with acute PE and that clinicians use the eStiMaTe calculator to estimate risk for an individual patient. The results of our study supported the model’s generalizability through external validation in a multicenter, multinational, European cohort of patients who had a lower severity of illness.
Different studies suggest the need for clinical prognostic models (i.e., sPESI) for identification of low-risk (i.e., for a complicated course) patients with PE who might safely undergo outpatient PE therapy (8, 9). This study further supports the applicability of the sPESI as a first step for identifying low-risk patients. Laboratory biomarkers offer a number of theoretical advantages when used in addition to clinical prediction rules. Standardized, readily available assays yield “objective” numerical results that may assist in the quantitative assessment of ventricular dysfunction, myocardial injury, and/or comorbidity. Particularly, BNPs are very sensitive indicators of neurohormonal activation due to ventricular overload and dysfunction. A recent prospective cohort study (26) suggested that BNPs might assist with selection of candidates for home PE treatment because of their very high sensitivity and negative predictive value (27). This study’s findings support this concept, particularly when BNP testing is combined with a clinical score (i.e., sPESI). Indeed, we found that patients with an sPESI of 0 and BNP less than or equal to 100 pg/ml had an 87% reduction in the risk of experiencing a complicated course during the 30-day follow-up period in comparison to those with an sPESI greater than or equal to 1 and/or BNP greater than 100 pg/mL.
Single tests have not demonstrated a sufficient positive predictive value to identify the subgroup of normotensive patients with acute symptomatic PE at high risk of PE-related complications. Multimarker models combining CCUS testing with prognostic tools indicating myocardial injury or RV dysfunction (i.e., cardiac troponin and transthoracic echocardiography, respectively) have improved the identification of such high-risk patients (13). This study validated such a multimarker prognostic model, in which the combination of a high-risk sPESI, an elevated BNP, an elevated cTnI, and concomitant DVT predicted a fivefold increase in the risk of an adverse 30-day outcome. Of note, the predictive model only classified 3.6% of patients in the derivation cohort and 4.1% of patients in the validation cohort as high risk, suggesting that a very small proportion of normotensive patients with acute symptomatic PE might benefit from an escalation of therapy. In fact, this study further confirms the low mortality rate for normotensive patients diagnosed with PE in the emergency department setting (28). Particularly, the data from this study and the PREP study suggest that such patients have a remarkably low complication rate directly attributable to PE.
Bleeding is an important potential complication of anticoagulation. However, because the aim of the study was to develop a multimarker prognostication useful both for identifying low-risk patients with PE (who might benefit from outpatient therapy) and high-risk normotensive patients with PE (who might benefit from thrombolysis), we decided not to include major bleeding as an endpoint of the study. The risk assessment focused on risks associated with PE. The study did adjudicate cause of death, and the contribution of bleeding was included as part of this specific assessment.
We developed a very simple model that integrates prognostic information from baseline clinical parameters (i.e., sPESI), in combination with easily obtained biomarkers and lower limb venous compression ultrasound testing. The model did not support using echocardiographic RV dysfunction as a predictor. In addition, echocardiographic criteria for defining RV dysfunction are poorly standardized and may vary widely between hospitals, ultrasound laboratories, and examiners (10). This simple algorithm and the calculator provide clinicians and patients with a framework for discussing prognosis, policy-makers with a tool for investigating stage-specific management options, and researchers with the ability to identify at-risk study populations that maximize the efficiency and power of clinical trials.
Limitations of the study include the retrospective analysis of the validation cohort, although prospectively developed. Despite the large number of patients assessed for this study, the low number of primary endpoints did not allow for more precision in our estimates. One weakness of the study included the restriction of recruitment to patients who could safely undergo the extra tests within 24 hours after diagnosis of PE. It is not apparent whether this affected the results. Because this study did not directly assess the impact of the clinical prediction models on the management of these patients, further studies are needed to address this important question. Also, the model does not address some other factors that affect decision making, such as bleeding risk and ability to comply with therapy (e.g., outpatient therapy) (29). The predictive model may not generalize to all settings, and future studies may further address its external validity and if it can change clinician behavior and improve patient outcomes.
In summary, the PROTECT study validated a multimarker model that predicts 30-day all-cause mortality, hemodynamic collapse, and/or recurrent PE in normotensive patients diagnosed with acute symptomatic PE in an emergency department.
Coordinator of the PROTECT Study: David Jiménez
PROTECT Steering Committee Members: David Jiménez, José Luis Lobo, Manuel Monreal, Remedios Otero, Roger D. Yusen
PROTECT Study Coordinating Center: S & H Medical Science Service
Adjudication Committee: Francisco Conget, Dolores Nauffal, Mikel Oribe, Fernando Uresandi
Radiology Panel: Ignacio Gallego, Luis Gorospe, Agustina Vicente
Blood Sample Processing: José Manuel del Rey
Statistician: Víctor Abraira, Javier Zamora, Alfonso Muriel
Investigators of the PROTECT study: Consolación Rodríguez, Jorge Vivancos, Jesús Marín (Bormujos), Mikel Oribe, Aitor Ballaz, Jose María Abaitúa, Sonia Velasco (Galdakao), Manuel Barrón, María Lladó, Carmen Rodrigo, Luis Javier Alonso (Logroño), Ramón Rabuñal, Olalla Castro, Concepción Iglesias, Ana Testa (Lugo), David Jiménez, Vicente Gómez, Luis Gorospe, Sem Briongos, José Manuel del Rey (Madrid), Celso Álvarez, Nuria Rodríguez, Amador Prieto, María Martín (Oviedo), Carmen Navarro, Mónica López, Eva Castañer, Eva Guillaumet (Sabadell), Remedios Otero, Teresa Elías, Pilar Serrano, Francisco López (Sevilla), Reina Valle, María Victoria Piret, Pilar Lucio, José María Cuesta (Sierrallana), Dolores Nauffal, Marta Ballester, José Pamies, Ana Osa (Valencia), José Luis Lobo, Vanesa Zorrilla, Delfina Pozo, Ángel Alonso (Vitoria), Francisco Conget, Miguel Ángel Santolaria, Mariano González, José Luis de Benito (Zaragoza)
| 1. | Kearon C, Akl EA, Comerota AJ, Prandoni P, Bounameaux H, Goldhaber SZ, Nelson ME, Wells PS, Gould MK, Dentali F, et al.; American College of Chest Physicians. Antithrombotic therapy for VTE disease: antithrombotic therapy and prevention of thrombosis, 9th ed: American College of Chest Physicians evidence-based clinical practice guidelines. Chest 2012;141:e419S-e494S. |
| 2. | Jaff MR, Archer SL, Cushman M, Goldenberg N, Goldhaber SZ, Jenkins S, Kline JA, Michaels AD, Thistlethwaite P, Vedantham S, et al., on behalf of the American Heart. Association Council on Cardiopulmonary, Critical Care, Perioperative and Resuscitation; Council on Peripheral Vascular Disease; and Council on Arteriosclerosis, Thrombosis and Vascular Biology. Management of massive and submassive pulmonary embolism, iliofemoral deep vein thrombosis, and chronic thromboembolic pulmonary hypertension: a scientific statement from the American Heart Association. Circulation 2011;123:1788–1830. |
| 3. | Torbicki A, Perrier A, Konstantinides SV, Agnelli G, Galiè N, Pruszczyk P, Bengel F, Brady AJ, Ferreira D, Janssens U, et al.; ESC Committee for Practice Guidelines (CPG). Guidelines on the diagnosis and management of acute pulmonary embolism: the Task Force for the Diagnosis and Management of Acute Pulmonary Embolism of the European Society of Cardiology (ESC). Eur Heart J 2008;29:2276–2315. |
| 4. | Penaloza A, Roy PM, Kline J. Risk stratification and treatment strategy of pulmonary embolism. Curr Opin Crit Care 2012;18:318–325. |
| 5. | Aujesky D, Obrosky DS, Stone RA, Auble TE, Perrier A, Cornuz J, Roy PM, Fine MJ. Derivation and validation of a prognostic model for pulmonary embolism. Am J Respir Crit Care Med 2005;172:1041–1046. |
| 6. | Jiménez D, Moores L, Gómez V, Lobo JL, Uresandi F, Otero R, Monreal M, Muriel A, Yusen RD. RIETE Investigators. Simplification of the pulmonary embolism severity index for prognostication in patients with acute symptomatic pulmonary embolism. Arch Intern Med 2010;170:1383–1389. |
| 7. | Aujesky D, Roy PM, Le Manach CP, Verschuren F, Meyer G, Obrosky DS, Stone RA, Cornuz J, Fine MJ. Validation of a model to predict adverse outcomes in patients with pulmonary embolism. Eur Heart J 2006;27:476–481. |
| 8. | Spirk D, Aujesky D, Husmann M, Hayoz D, Baldi T, Frauchiger B, Banyai M, Baumgartner I, Kucher N. Cardiac troponin testing and the simplified Pulmonary Embolism Severity Index. The SWIss Venous ThromboEmbolism Registry (SWIVTER). Thromb Haemost 2011;106:978–984. |
| 9. | Aujesky D, Roy PM, Verschuren F, Righini M, Osterwalder J, Egloff M, Renaud B, Verhamme P, Stone RA, Legall C, et al. Outpatient versus inpatient treatment for patients with acute pulmonary embolism: an international, open-label, randomised, non-inferiority trial. Lancet 2011;378:41–48. |
| 10. | Lankeit M, Jiménez D, Kostrubiec M, Dellas C, Hasenfuss G, Pruszczyk P, Konstantinides S. Predictive value of the high-sensitivity troponin T assay and the Simplified Pulmonary Embolism Severity Index in hemodynamically stable patients with acute pulmonary embolism: a prospective validation study. Circulation 2011;124:2716–2724. |
| 11. | Sanchez O, Trinquart L, Colombet I, Durieux P, Huisman MV, Chatellier G, Meyer G. Prognostic value of right ventricular dysfunction in patients with haemodynamically stable pulmonary embolism: a systematic review. Eur Heart J 2008;29:1569–1577. |
| 12. | Jiménez D, Aujesky D, Díaz G, Monreal M, Otero R, Martí D, Marín E, Aracil E, Sueiro A, Yusen RD. RIETE Investigators. Prognostic significance of deep vein thrombosis in patients presenting with acute symptomatic pulmonary embolism. Am J Respir Crit Care Med 2010;181:983–991. |
| 13. | Jiménez D, Aujesky D, Moores L, Gómez V, Martí D, Briongos S, Monreal M, Barrios V, Konstantinides S, Yusen RD. Combinations of prognostic tools for identification of high-risk normotensive patients with acute symptomatic pulmonary embolism. Thorax 2011;66:75–81. |
| 14. | The Steering Committee, on behalf of the PEITHO investigators. Single-bolus tenecteplase plus heparin compared with heparin alone for normotensive patients with acute pulmonary embolism, who have evidence of right ventricular dysfunction and myocardial injury: rationale and design of the pulmonary embolism thrombolysis (PEITHO) trial. Am Heart J 2012;163:33–38. |
| 15. | Jiménez D, Lobo JL, Monreal M, Otero R, Yusen RD, on behalf of the PROTECT investigators. Prognostic significance of multidetector computed tomography in normotensive patients with pulmonary embolism: rationale, methodology and reproducibility for the PROTECT study. J Thromb Thrombolysis 2012;34:187–192. |
| 16. | Remy-Jardin M, Remy J, Wattinne L, Giraud F. Central pulmonary thromboembolism: diagnosis with spiral volumetric CT with the single-breath-hold-technique-comparison with pulmonary angiography. Radiology 1992;185:381–387. |
| 17. | Becattini C, Agnelli G, Vedovati MC, Pruszczyk P, Casazza F, Grifoni S, Salvi A, Bianchi M, Douma R, Konstantinides S, et al. Multidetector computed tomography for acute pulmonary embolism: diagnosis and risk stratification in a single test. Eur Heart J 2011;32:1657–1663. |
| 18. | Grifoni S, Olivotto I, Cecchini P, Pieralli F, Camaiti A, Santoro G, Conti A, Agnelli G, Berni G. Short term clinical outcome of patients with pulmonary embolism, normal blood pressure and echocardiographic right ventricular dysfunction. Circulation 2000;101:2817–2822. |
| 19. | Jiménez D, Díaz G, Molina J, Martí D, Del Rey J, García-Rull S, Escobar C, Vidal R, Sueiro A, Yusen RD. Troponin I and risk stratification of patients with acute nonmassive pulmonary embolism. Eur Respir J 2008;31:847–853. |
| 20. | Pieralli F, Olivotto I, Vanni S, Conti A, Camaiti A, Targioni G, Grifoni S, Berni G. Usefulness of bedside testing for brain natriuretic peptide to identify right ventricular dysfunction and outcome in normotensive patients with acute pulmonary embolism. Am J Cardiol 2006;97:1386–1390. |
| 21. | Sanchez O, Trinquart L, Caille V, Couturaud F, Pacouret G, Meneveau N, Verschuren F, Roy PM, Parent F, Righini M, et al. Prognostic factors for pulmonary embolism: the PREP study, a prospective multicenter cohort study. Am J Respir Crit Care Med 2010;181:168–173. |
| 22. | Harrel FE Jr, Lee KL, Mark DB. Multivariable prognostic models: issues in developing models, evaluating assumptions and adequacy, and measuring and reducing errors. Stat Med 1996;15:361–387. |
| 23. | Pencina MJ, D’Agostino RB. Overall C as a measure of discrimination in survival analysis: model specific population value and confidence interval estimation. Stat Med 2004;23:2109–2123. |
| 24. | Gerds TA, Cai T, Schumacher M. The performance of risk prediction models. Biom J 2008;50:457–479. |
| 25. | Hosmer DW Jr, Lemeshow S. Applied logistic regression. New York: J Wiley; 1989. |
| 26. | Agterof MJ, Schutgens RE, Snijder RJ, Epping G, Peltenburg HG, Posthuma EF, Hardeman JA, van der Griend R, Koster T, Prins MH, et al. Out of hospital treatment of acute pulmonary embolism in patients with a low NT-proBNP level. J Thromb Haemost 2010;8:1235–1241. |
| 27. | Kucher N, Goldhaber SZ. Cardiac biomarkers for risk stratification of patients with acute pulmonary embolism. Circulation 2003;108:2191–2194. |
| 28. | Pollack CV, Schreiber D, Goldhaber SZ, Slattery D, Fanikos J, O'Neil BJ, Thompson JR, Hiestand B, Briese BA, Pendleton RC, et al. Clinical characteristics, management, and outcomes of patients diagnosed with acute pulmonary embolism in the emergency department: initial report of EMPEROR (Multicenter Emergency Medicine Pulmonary Embolism in the Real World Registry). J Am Coll Cardiol 2011;57:700–706. |
| 29. | Zondag W, Mos IC, Creemers-Schild D, Hoogerbrugge AD, Dekkers OM, Dolsma J, Eijsvogel M, Faber LM, Hofstee HM, Hovens MM, et al.; Hestia Study Investigators. Outpatient treatment in patients with acute pulmonary embolism: the Hestia study. J Thromb Haemost 2011;9:1500–1507. |
* A complete list of members may be found before the beginning of the References.
Supported by the Instituto de Salud Carlos III (FIS PI08/200; FIS PI11/00246) (D.J.), the Sociedad Española de Neumología y Cirugía Torácica (SEPAR 2008) (D.J.), and the Sociedad Madrileña de Neumología (NM 2010) (D.J.).
Author Contributions: D.J. had full access to all the data in the study and had final responsibility for the decision to submit for publication. Study concept and design: D.J., V.T., J.L.L., M.M., D.A., S.K., R.D.Y. Acquisition of data, analysis and interpretation of data, statistical analysis: D.J., D.K., V.T., B.B., D.S., J.L.L., M.M., D.A., O.S., G.M., S.K., R.D.Y. Drafting of the manuscript: D.J., D.K., V.T., B.B., D.S., J.L.L., M.M., D.A., O.S., G.M., S.K., R.D.Y. Critical revision of the manuscript for important intellectual content: D.J., D.K., V.T., B.B., D.S., J.L.L., M.M., D.A., O.S., G.M., S.K., R.D.Y. Study supervision: D.J., R.D.Y.
This article has an online supplement, which is accessible from this issue’s table of contents at www.atsjournals.org
Originally Published in Press as DOI: 10.1164/rccm.201311-2040OC on January 29, 2014
Author disclosures are available with the text of this article at www.atsjournals.org.
