Predictors of hospital mortality in hemodynamically stable patients with pulmonary embolism

Abstracts

BACKGROUND: Pulmonary embolism is associated with high mortality in patients with hypotension or circulatory shock. However, the association between some clinical variables and mortality is still unclear in hemodynamically stable patients. OBJECTIVES: To derive an in-hospital mortality risk stratification model in hemodynamically stable patients with pulmonary embolism. METHODS: This is a prospective multicenter cohort study of 582 consecutive patients admitted in emergency units or intensive care units with clinically suspected pulmonary embolism and whose diagnosis was confirmed by one or more of the following tests: pulmonary arteriography, spiral CT angiography, magnetic resonance angiography, Doppler echocardiography, pulmonary scintigraphy, or venous duplex scan. Data on demographics, comorbidities and clinical manifestations were collected and included in a logistic regression analysis so as to build the prediction model. RESULTS: Overall mortality was 14.1%. The following parameters were identified as independent death risk variables: age > 65 years, bed rest > 72h, chronic cor pulmonale, sinus tachycardia, and tachypnea. After risk stratification, mortalities of 5.4%, 17.8%, and 31.3% were found in the low, moderate and high-risk subgroups, respectively. The model showed 65.5% sensitivity and 80% specificity, with a 0.77 area under the curve. CONCLUSION: In hemodynamically stable patients with pulmonary embolism, age > 65 years, bed rest > 72h, chronic cor pulmonale, sinus tachycardia and tachypnea were independent predictors of in-hospital mortality. However, further validation of the prediction model in other populations is required so that it can be incorporated into the clinical practice.

Hospital mortality; pulmonary embolism; pulmonary heart disease; tachycardia; sinus

FUNDAMENTO: A embolia pulmonar apresenta alta mortalidade em pacientes com hipotensão arterial ou choque circulatório. Entretanto, em pacientes hemodinamicamente estáveis, a associação de algumas variáveis clínicas com a mortalidade ainda não está claramente estabelecida. OBJETIVOS: Derivar um modelo de estratificação do risco de mortalidade intra-hospitalar em pacientes com embolia pulmonar hemodinamicamente estáveis. MÉTODOS: Estudo de coorte multicêntrico prospectivo de 582 pacientes consecutivos que foram admitidos em unidades de emergência ou de terapia intensiva, com suspeita clínica de embolia pulmonar, e que tiveram o diagnóstico confirmado por meio de um ou mais dos seguintes exames: arteriografia pulmonar, angiotomografia computadorizada helicoidal, angioressonância magnética, ecodopplercardiograma, cintilografia pulmonar ou duplex-scan venoso. Os dados sobre características demográficas, comorbidades e manifestações clínicas foram coletados e incluídos em uma análise de regressão logística para compor o modelo de predição. RESULTADOS: A mortalidade global foi de 14,1%. Foram identificadas como variáveis independentes de risco de óbito: idade > 65 anos; repouso no leito > 72h; cor pulmonale crônico; taquicardia sinusal e taquipnéia. Após a estratificação por faixas de risco, observaram-se mortalidades de 5,4%, 17,8% e 31,3%, respectivamente nos subgrupos de baixo, moderado e alto riscos. O modelo mostrou sensibilidade de 65,5% e especificidade de 80%, com uma área sob a curva de 0,77. CONCLUSÃO: Em pacientes hemodinamicamente estáveis com embolia pulmonar, a idade > 65 anos, o repouso no leito > 72h, o cor pulmonale crônico, a taquicardia sinusal e a taquipnéia foram preditores independentes da mortalidade intra-hospitalar. Entretanto o modelo de predição necessita ser validado em outras populações para sua incorporação à prática clínica.

Mortalidade hospitalar; embolia pulmonar; doença cardiopulmonar; taquicardia sinusal

FUNDAMENTO: La embolia pulmonar presenta alta mortalidad en pacientes con hipotensisn arterial o shock circulatorio. Sin embargo, en pacientes hemodinamicamente estables, la asociacisn de algunas variables clmnicas con la mortalidad todavma no esta claramente establecida. OBJETIVO: Derivar un modelo de estratificacisn del riesgo de mortalidad intrahospitalaria en pacientes con embolia pulmonar hemodinamicamente estables. MITODOS: Estudio de cohorte multicintrico prospectivo de 582 pacientes consecutivos que ingresaron en unidades de emergencia o de terapia intensiva, con sospecha clmnica de embolia pulmonar, y que tuvieron el diagnsstico confirmado por medio de uno o mas de los examenes: arteriografma pulmonar, angiotomografma computadorizada helicoidal, angiorresonancia magnitica, ecocardiografma doppler, centellografma pulmonar o duplex scan venoso. Los datos sobre caractermsticas demograficas, comorbilidades y manifestaciones clmnicas se recolectaron e se incluyeron en un analisis de regresisn logmstica para componer el modelo de prediccisn. RESULTADOS: La mortalidad global fue del 14,1%. Se identificaron como variables independientes de riesgo de sbito: edad > 65 aqos; reposo en la cama > 72h; cor pulmonale crsnico; taquicardia sinusal y taquipnea. Tras la estratificacisn por franjas de riesgo, se observaron mortalidades del 5,4%, el 17,8% y el 31,3%, respectivamente en los subgrupos de bajo, moderado y alto riesgos. El modelo revels sensibilidad del 65,5% y especificidad del 80%, con un area bajo la curva de 0,77. CONCLUSISN: En pacientes hemodinamicamente estables con embolia pulmonar, la edad > 65 aqos, el reposo en la cama > 72h, el cor pulmonale crsnico, la taquicardia sinusal y la taquipnea fueron predictores independientes de la mortalidad intrahospitalaria. Sin embargo, el modelo de prediccisn necesita validarse en otras poblaciones para su incorporacisn a la practica clmnica. (Arq Bras Cardiol 2009; 93(2) : 132-137)

Mortalidad hospitalaria; embolia pulmonar; enfermedad cardiopulmonar; taquicardia sinusal

ORIGINAL ARTICLE

Predictors of hospital mortality in hemodynamically stable patients with pulmonary embolism

André Volschan^I; Denilson Albuquerque^II; Bernardo Rangel Tura^II; Marcos Knibel^VIII; José Péricles Esteves^III; Luiz Carlos Bodanese^V; Francisco Silveira^IX; João Pantoja^IV; Paulo Cesar Pereira da Silva e Souza^VII; João Mansur^VI; Evandro Tinoco Mesquita^I,*

^IHospital Pró-Cardíaco, Rio de Janeiro, RJ

^IIUniversidade do Estado do Rio de Janeiro, Rio de Janeiro, RJ

^IIIHospital Português, Salvador, BA

^IVHospital Copa D'Or, Rio de Janeiro, RJ

^VHospital São Lucas, Porto Alegre, RS

^VIHospital Samaritano, Rio de Janeiro, RJ

^VIIHospital de Clínicas de Niterói, Niterói, RJ

^VIIIHospital CardioTrauma, Rio de Janeiro, RJ

^IXHospital Prontocor, Belo Horizonte, MG - Brazil

Mailing address

SUMMARY

BACKGROUND: Pulmonary embolism is associated with high mortality in patients with hypotension or circulatory shock. However, the association between some clinical variables and mortality is still unclear in hemodynamically stable patients.

OBJECTIVE: To derive an in-hospital mortality risk stratification model in hemodynamically stable patients with pulmonary embolism.

METHODS: This is a prospective multicenter cohort study of 582 consecutive patients admitted in emergency units or intensive care units with clinically suspected pulmonary embolism and whose diagnosis was confirmed by one or more of the following tests: pulmonary arteriography, spiral CT angiography, magnetic resonance angiography, Doppler echocardiography, pulmonary scintigraphy, or venous duplex scan. Data on demographics, comorbidities and clinical manifestations were collected and included in a logistic regression analysis so as to build the prediction model.

RESULTS: Overall mortality was 14.1%. The following parameters were identified as independent death risk variables: age > 65 years, bed rest > 72h, chronic cor pulmonale, sinus tachycardia, and tachypnea. After risk stratification, mortalities of 5.4%, 17.8%, and 31.3% were found in the low, moderate and high-risk subgroups, respectively. The model showed 65.5% sensitivity and 80% specificity, with a 0.77 area under the curve.

CONCLUSION: In hemodynamically stable patients with pulmonary embolism, age > 65 years, bed rest > 72h, chronic cor pulmonale, sinus tachycardia and tachypnea were independent predictors of in-hospital mortality. However, further validation of the prediction model in other populations is required so that it can be incorporated into the clinical practice.

Key words: Hospital mortality; pulmonary embolism; pulmonary heart disease; tachycardia, sinus.

Introduction

Pulmonary embolism (PE) is the third cardiovascular cause of hospital admission¹, following acute coronary syndrome and stroke. It has a broad spectrum of clinical manifestations, especially when associated with decompensated heart failure² and chronic obstructive pulmonary disease. Massive PE characterized by hemodynamic instability defines the subgroup of more severely ill patients among whom high mortality rates are observed^3,4, and who can benefit from a more aggressive therapeutic approach. On the other hand, clinically stable patients considered at low risk require a shorter length of hospital stay and, in some cases, can be treated as outpatients with low-molecular-weight heparin^5,6. Thus, prognostic assessment becomes useful to guide the therapeutic strategy and other care.

If, on one hand, hypotension characterizes patients with PE at a higher risk of death, on the other hand echocardiogram, troponin, and brain natriuretic peptide levels help identify cases with a worse prognosis in hemodynamically stable patients. Since the early 1990's, the presence of right ventricular dysfunction as detected in the echocardiogram of normotensive patients classified as having submassive PE has been related to higher mortality^7,8. Despite its ability to stratify patients at a higher risk, echocardiography is not always available 24 hours a day, and this limits its use.

Troponin and brain natriuretic peptide are parameters that indirectly express ventricular involvement^9,10 and help select patients at a higher risk^11,12. Although the use of these tests has progressively increased, they are still uncommonly used in less complex health centers that also admit patients with PE.

Prognostic clinical markers in stable patients that may contribute to death risk stratification in patients with PE have not been frequently studied^13-15. Demographic data, risk factors, signs and symptoms comprise a group of low-complexity variables that can be assessed at any hospital environment, regardless of ancillary tests, and this makes this strategy universally usable.

Objective

The objective of this study was to elaborate a model based on clinical stratification markers of in-hospital mortality risk in hemodynamically stable patients with PE.

Methods

This is a prospective multicenter cohort study conducted in 24 investigation centers of 20 Brazilian tertiary-care hospitals from January 1988 to May 2003. From an initial group of 727 consecutive patients with suspected PE, those presenting hemodynamic instability were excluded; thus a sample of 582 consecutive patients (42.1% males, median age of 73 years, ranging from 18 to 102) admitted in emergency or intensive care units was analyzed. PE was clinically suspected by the physician who evaluated the patient, based on risk factors, signs and symptoms of the disease. Systolic blood pressure > 90 mmHg was considered a criterion of hemodynamic stability. In addition to clinical suspicion, PE had to be documented using one of the following ancillary methods:

1. Pulmonary arteriography with visualization of the thrombus in pulmonary artery.

2. Spiral CT angiography with visualization of the thrombus in pulmonary artery.

3. Magnetic resonance angiography with visualization of the thrombus in pulmonary artery.

4. Echocardiography with visualization of the thrombus in pulmonary artery.

5. Ventilation/perfusion pulmonary scintigraphy with a high probability of PE.

6. Duplex scan with visualization of the thrombus and reduced compressibility in the deep venous system.

The following variables were considered for the univariate analysis: age, gender, previous history of venous thromboembolism, hip or lower limb fracture in the past 90 days, abdominal or pelvic surgery in the past 30 days, neoplasia, bed rest > 72h, chronic cor pulmonale, cigarette smoking, heart failure, stroke, chest pain, sinus tachycardia (heart rate > 100 bpm), syncope, dyspnea, tachypnea (respiratory rate > 20 bpm), fever (axillary temperature > 37 ºC), cough, cyanosis and hemoptisis or bloody sputum.

The information collected was included in a standardized form by investigators of each study center and later sent to the chief investigator, to be stored in a CSV-format database and exported to the R statistical package version 2.6.0, where the analyses were carried out¹⁶.

Statistical analysis

The model used demographic variables, comorbidities and clinical manifestations that could be easily collected during the baseline visit of patients with PE.

Data were described as percentages, means and standard deviation or median and interquartile deviation, according to the type of variable (categorical, normal, and nonparametric). Dichotomization of continuous variables was performed using the ROC (Receiver Operating Characteristic) curve. Student's t, Mann-Whitney, chi square or Fisher's exact tests were used to measure the association between clinical variables and the endpoint. P values < 0.20 or the existence of a strong clinical association with mortality were used as selection criteria for inclusion in the multivariate model¹⁷.

The multivariate model was adjusted using logistic regression with the selection of variables guided by the likelihood ratio, using the least possible number of variables without loss of the predictive ability; its accuracy was assessed by the C statistics. After the model was created, a score was elaborated based on the odds ratio value approximate to the unit of each variable. The risk score was analyzed using the chi square test of linear trend, and the death risk was further evaluated in each stratum.

Survival analysis was carried out in the three subgroups using the Kaplan-Meyer estimator.

Results

The overall mortality of this study was 14.1%; demographic and clinical characteristics are described in Table 1.

The median age was 73 years, ranging from 18 to 102 years, and the best cut-off point was the age of 65 years, as determined using the ROC curve¹⁶, with an area under the curve of 0.60 (Figure 1), sensitivity of 81.8%, specificity of 88.3%, and positive and negative likelihood ratios of 6.99 and 0.21, respectively. In addition to age > 65 years, the model identified bed rest, chronic cor pulmonale, sinus tachycardia, and tachypnea as variables able to independently predict death risk, using logistic regression (Table 2).

From the prediction model, the expected death risk was calculated for each patient, and an ROC curve was constructed¹⁸ (Figure 2). The area under the curve - also known as C statistics - was 0.77, which corresponds to the model accuracy, and operating characteristics with 65.5% sensitivity, 80.0% specificity, positive likelihood ratio of 3.3, and negative likelihood ratio of 0.4. Figure 3 shows the relationship between pre-test probability (14.1%), which corresponds to the overall mortality in the study, and the post-test probability of death in the case of a positive test (35.1%) and of a negative test (6.2%). Comparison of mortality curves (Figure 4) showed a significant difference between the risk groups, with p value = 0.000004.

Based on the predictive power of each variable, as guided by the respective odds ratio, the model showed a linear association for prediction of death (p>0.001).

For the calculation of the score points of each patient, values described in Table 3 were used; the risk stratification showed a death risk of 5.4% among patients considered at low risk, and of 17.8% and 33.1% in the moderate and high risk groups, respectively, as shown in Table 4.

Discussion

Most of the studies on the prognosis of PE use echocardiographic data of right ventricular dysfunction^19,20 or laboratory data such as troponin^21,22 and brain natriuretic peptide^23-25 elevation to quantify the risk of a worse outcome for the patients. Despite their good prognostic accuracy, these tests are not broadly available.

In this first analysis of our study, we chose to use a model comprised of clinical variables only, based on the recommendations that by using a simpler prediction rule its application becomes more encompassing²⁶. Hemodynamically unstable patients were excluded so that we could evaluate a more homogeneous subgroup at a lower death risk. Despite the fact that this population was less severely ill in comparison to patients with hypotension, the overall mortality of 14.1% was higher than those found in related publications.

Wicki et al²⁷ studied 296 consecutive patients and constructed a prediction model, the Geneva prognostic score, and observed 10.1% of adverse events, with an overall mortality of 8.4%. The predictors of adverse events used in this model were cancer, hypotension, hypoxemia, heart failure, history of previous deep-vein thrombosis, and documented deep-vein thrombosis on duplex scan.

In the PESI study²⁸, Aujesky et al derived a risk stratification model in a total of 10,534 patients distributed into five subgroups using 11 clinical variables, and observed an overall mortality of 9.2%, ranging from 1.1% (Class I, very low risk) to 24.5% (Class V, very high risk).

These two models were later compared as to their ability to select patients for treatment as outpatients, and demonstrated a significant difference for the prediction of 30-day mortality (PESI 0.9%; 95% CI, 0.3 to 2.2; vs Geneva, 5.6%; 95% CI, 3.6 to 7.6 - p < 0.0001)⁵.

The higher mortality in our case series suggests that, despite hemodynamic stability, the population studied, which was comprised of patients admitted in intensive care units, could be at a higher risk of complications. For the same reason, it is possible to understand the higher death risk in each stratum assessed in comparison with those of other studies mentioned^27,28.

The identification of a subgroup that could be treated on an outpatient basis was not possible in this analysis, since the 5.4% mortality risk attributed to the low-risk stratum is too high to permit the choice of this treatment modality. On the other hand, the 31.3% mortality in the high-risk subgroup indicates the need for more complex interventions and surveillance of these patients.

Derivation of a prediction model is the starting point of a strategy to obtain a rule for clinical decision making. Thus, internal and external validation phases are fundamental for the model to be universally applied²⁶. Similar databases are being collected in other investigation centers, so that we can validate the model in other populations.

Conclusions

In hemodynamically stable patients with PE, age > 65 years, bed rest, chronic cor pulmonale, sinus tachycardia, and tachypnea were able to independently predict death risk using the logistic regression model.

Acknowledge

We acknowledge the participation of the following EMEP (Estudo Multicêntrico de Embolia Pulmonar - Multicenter Study on Pulmonary Embolism) investigators, for data collection:

Evandro Tinoco Mesquita (Hospital PróCardíaco) Rio de Janeiro RJ, João Mansur Filho, José Kezen Camillo Jorge (Hospital Samaritano) Rio de Janeiro RJ, Gustavo Nobre (Hospital Rio Mar) Rio de Janeiro RJ, Marcos Baptista, Nelson Arthur, Marcos Knibel, Marcelo Lugarinho (Hospital Lucas Hospital) Rio de Janeiro RJ, Gláucia Maria, Ricardo Eiras (Hospital Prontocor) Rio de Janeiro RJ, Gustavo Gouvêa, Alexandre Drumond, Eduardo Eiras, Emanuel Salgueiro (Casa de Saúde São José) Rio de Janeiro RJ, Mônica Callafiori (Hospital CardioTrauma) Rio de Janeiro RJ, Paulo César Pereira da Silva e Souza, Marcelo Magalhães (Hospital de Clinicas de Niterói) Niterói RJ, Ana Paula Lacerda (Hospital Barra D`Or) Rio de Janeiro RJ, Fernando Pacheco (Hospital Copa D`Or) Rio de Janeiro RJ, Marcelo Vieira Gomes, Luis Felipe Camille (Clinica São Vicente) Rio de Janeiro RJ, Guilherme Aguiar (Hospital São Vicente de Paulo) Rio de Janeiro RJ, Maria Luiza Toscano, Roberto Caetano (Instituto Nacional do Cancer) Rio de Janeiro RJ, José Péricles Esteves, Carolina Barbosa (Hospital Português) Salvador BA, Francisco Silveira (Hospital Prontocor) Belo Horizonte MG, Alex Macedo (Hospital Santa Casa) Santos SP, Luis Carlos Bodanese (Hospital São Lucas) Porto Alegre RS.

Potential Conflict of Interest

No potential conflict of interest relevant to this article was reported.

Sources of Funding

This study was partially funded by Sanofi-Aventis.

Study Association

This article is part of the thesis of doctorla submitted by Andre Volschan, from Universidade do Estado do Rio de Janeiro.

References

• 1. Giuntini C, Di Ricco G, Marini C, Melillo E, Palla A. Pulmonary embolism: epidemiology. Chest. 1995; 107: 3S-9S.
• 2. Terzi CB, Lage SG, Dragosavac D, Terzi RGG. Insuficiência cardíaca grave em unidade de terapia intensiva: existe um índice prognóstico ideal? Arq Bras Cardiol. 2006; 87: 344-51.
• 3. Goldhaber SZ, Visani L, De Rosa M. Acute pulmonary embolism: clinical outcomes in the International Cooperative Pulmonary Embolism Registry. Lancet. 1999; 353: 1386-9.
• 4. Kurkciyan I, Meron G, Sterz F, Janata K, Domanovits H, Holzer M, et al. Pulmonary embolism as a cause of cardiac arrest: presentation and outcome. Arch Intern Med. 2000; 160: 1529-35.
• 5. Jiménez D, Yusen RD, Otero R, Uresandi F, Nauffal D, Laserna E, et al. Prognostic models for selecting patients with acute pulmonary embolism for initial outpatient therapy. Chest. 2007; 132 (1): 7-8.
• 6. Wells PS. Outpatient treatment of patients with deep-vein thrombosis or pulmonary embolism. Curr Opin Pulm Med. 2001; 7: 360-4.
• 7. Ribeiro A, Lindmarker P, Juhlin-Dannfelt A, Johnsson H, Jorfeldt L. Echocardiography Doppler in pulmonary embolism: right ventricular dysfunction as a predictor of mortality rate. Am Heart J. 1997; 134: 479-87.
• 8. Grifoni S, Olivotto I, Cecchini P, Pieralli F, Camaiti A, Santoro G, et al. Short-term clinical outcome of patients with pulmonary embolism, normal blood pressure, and echocardiographic right ventricular dysfunction. Circulation. 2000; 101: 2817-22.
• 9. Meyer T, Binder L, Hruska N, Luthe H, Buchwald AB. Cardiac troponin I elevation in acute pulmonary embolism is associated with right ventricular dysfunction. J Am Coll Cardiol. 2000; 36: 1632-6.
• 10. Tulevski II, ten Wolde M, van Veldhuisen DJ, Mulder JW, van der Wall EE, Buller HR, et al. Combined utility of brain natriuretic peptide and cardiac troponin T may improve rapid triage and risk stratification in normotensive patients with pulmonary embolism. Int J Cardiol. 2007; 116: 161-6.
• 11. Becattini C, Vedovati MC, Agnelli G. Prognostic value of troponins in acute pulmonary embolism: a meta-analysis. Circulation. 2007; 116: 427-33.
• 12. Douketis JD, Leeuwenkamp O, Grobara P, Johnston M, Sohne M, Tem Wolde M, et al. The incidence and prognostic significance of elevated cardiac troponins in patients with submassive pulmonary embolism. J Thromb Haemost. 2005; 3: 508-13.
• 13. Aujesky D, Roy PM, Le Manach CP, Verschuren F, Meyer G, Obrosky DS, et al. Validation of a model to predict adverse outcomes in patients with pulmonary embolism. Eur Heart J. 2006; 27: 476-81.
• 14. Aujesky D, Obrosky DS, Stone RA, Auble TE, Perrier A, Cornuz J, et al. A prediction rule to identify low-risk patients with pulmonary embolism. Arch Intern Med. 2006; 166 (2): 169-75.
• 15. Jiménez D, Díaz G, Valle M, Martí D, Escobar C, Vidal R, et al. Prognostic value of syncope in the presentation of pulmonary embolism. Arch Bronconeumol. 2005; 41 (7): 385-8.
• 16. R Development Core Team (2007) R: a language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria; 2007.
• 17. Greeland, S. Modeling and variable selection in epidemiology analysis. Am J Public Health. 1989; 79 (3): 340-9.
• 18. Hanley JA, McNeil BJ. The meaning and use of the area under the receiver operating characteristic (ROC) curve. Radiology. 1982; 143: 29-36.
• 19. Grifoni S, Vanni S, Magazzini S, Olivotto I, Conti A, Zanobetti M, et al. Association of persistent right ventricular dysfunction at hospital discharge after acute pulmonary embolism with recurrent thromboembolic events.. Arch Intern Med. 2006; 166: 2151-6.
• 20. Sociedade Brasileira de Cardiologia. Diretriz de embolia pulmonar. Arq Bras Cardiol. 2004; 83 (supl 1): 1-8.
• 21. Konstantinides S, Geibel A, Olschewski M, Kasper W, Hruska N, Jäckle S, et al. Importance of cardiac troponins I and T in risk stratification of patients with acute pulmonary embolism. Circulation. 2002; 106: 1263-8.
• 22. Giannitsis E, Muller-Bardorff M, Kurowski V, Weidtmann B, Wiegand U, Kampmann M, et al. Independent prognostic value of cardiac troponin T in patients with confirmed pulmonary embolism. Circulation. 2000; 102 (2): 211-7.
• 23. Pruszczyk P, Kostrubiec M, Bochowicz A, Styczynski G, Szulc M, Kurzyna M, et al. N-terminal pro-brain natriuretic peptide in patients with acute pulmonary embolism. Eur Respir J. 2003; 22: 649-53.
• 24. Binder L, Pieske B, Olschewski M, Geibel A, Klostermann B, Reiner C, et al. N-terminal pro-brain natriuretic peptide or troponin testing followed by echocardiography for risk stratification of acute pulmonary embolism. Circulation. 2005; 112: 1573-9.
• 25. Sanchez O, Trinquart L, Colombet I, Durieux P, Huisman MV, Chatellier G, et al. Prognostic value of right ventricular dysfunction in patients with haemodynamically stable pulmonary embolism: a systematic review. Eur Heart J. 2008; 29 (12): 1569-77.
• 26. Thomas G, McGinn TG, Guyatt GH, Wyer PC, Naylor CD, Stiell IG, et al. for the Evidence-Based Medicine Working Group. Users' Guides to the Medical Literature XXII: How to Use Articles About Clinical Decision Rules. JAMA. 2000; 284: 79-84.
• 27. Wicki J, Perrier A, Perneger TV, Bounameaux H, Junod AF. Predicting adverse outcome in patients with acute pulmonary embolism: a risk score. Thromb Haemost. 2000; 84 (4): 548-52.
• 28. Aujesky D, Obrosky DS, Stone RA, Auble TE, Perrier A, Cornuz J, et al. Derivation and validation of a prognostic model for pulmonary embolism. Am J Resp Crit Care Med. 2005; 172: 1041-6.

*

Em nome dos investigadores do EMEP (Estudo Multicêntrico de Embolia Pulmonar)

Publication Dates

• Publication in this collection
  13 Oct 2009
• Date of issue
  Aug 2009

History

• Reviewed
  12 Sept 2008
• Received
  12 Aug 2008
• Accepted
  19 Sept 2008