Does this patient have Pulmonary Embolism?


Does this patient have Pulmonary Embolism?

Original Article: Chunilal SD et al. Does this patient have Pulmonary Embolism? JAMA,2003; 209:2849-2857

Bottom Line:

§ The true value of the clinical prediction rules is the ability to formulate a pre-test probability based on the risk category in which a patient is placed. Then use further testing (D-dimer, CT, V-Q Scan, LE doppler) to develop a post test probability..

§ Clinical Assessment alone is insufficient to diagnose or rule out pulmonary embolism

§ Clinical prediction rules can be used to risk stratify patients with symptoms (dyspnea and chest pain) and/or risk factors for pulmonary embolism

§ A negative D-dimer in a patient with a low pre-test probability of PE appears to effectively rule out PE

SORT Grade of Recommendation: A Validated clinical prediction rules

Comments/Hints/Suggestions

Choose one decision rule with which you are comfortable and can reliably and efficently reproduce

Prevalence: The overall prevalence of pulmonary embolism is 1-2 per 1000 people annually. The prevalence in the populations studied for the prediction rules varied from 9.5% to 29%.

Accuracy of Exam: Clinical Gestalt

Based on several studies experienced physicians using history, physical exam usually coupled with ECG, CXR and ABG can divide patients into risk categories as follows

.

Risk Category

Prevalence

+ LR

Low

8-19 %

0.13-0.53

Moderate

26-47%

0.67-1.1

High

46-91%

12-1.9

Accuracy of Prediction Rules: Data from Prospective Validations Wells Criteria

Study

Sanson et al, 2000

Wells et al, 2001

Chagnon et al, 2002

(414 patients)

(930 patients)

(277 patients)

Risk Category

Prevalence %

LR

Prevalence %

LR

Prevalence %

LR

Low

28

0.93

1.3

0.13

12

0.39

Moderate

30

1.0

16.2

1.9

40

2

High

38

1.4

40.6

5.9

91

29

Genevea Criteria

Study

Chagnon et al, 2000

Perrier et al, 2004*

(277 patients)

(771 patients)

Risk Category

Prevalence %

LR

Prevalence %

LR

Low

13

0.44

7

0.23

Moderate

38

1.8

34

1.7

High

67

5.8

85

18.9

Description of Prediction Rules

Simplified Wells

Score

Geneva

Score

Clinical signs/symptoms of DVT (leg swelling

Age 60-79

1

and pain with palpation of deep veins of leg)

3.0

Age >80

2

No alternate diagnosis as likely or more likely than PE

3.0

Previous DVT or PE

2

Heart reate > 100 beats/min

1.5

Recent Surgery

3

Immobilization or surgery in last 4 weeks

1.5

Pulse rate >100

1

Previous history of DVT or PE

1.5

PaCO2 < 4.8 kPa

2

Hemoptysis

1.0

4.8-5.19 kPa

1

Cancer actively treated within the last 6 months

1.0

PaO2 <6.5 kPa

4

Probability categories: low <2, moderate 2-6, high>6

6.5-7.99 kPa

3

8-9.49 kPa

2

9.5-10.99 kPa

1

CXR with platelike atelectasis

1

with elevated hemidiaphragm

1

Probability categories: low 0-4, moderate 5-8, high 9-16

There are three sets of clinical prediction rules (Simplified Wells, Geneva, and Charlotte), however, only the two mentioned in this review have been prospectively validated

Precision: The extended Wells criteria (precursor to simplified Wells) had a Kappa of 0.86. One would expect the easier to use simplified Wells to have similar agreement. Kappa not available for Geneva criteria.

Studies Description

Inclusion Criteria

Included studies had to 1) provide an estimate of pretest probability for PE using clinical gestalt or clinical prediction rule 2) persons performing clinical assessment must be blinded to results of diagnostic testing 3) provide comparison of assessments to validated methods of ruling in or out pulmonary embolism (proof of comparison to gold standard) Studies using clinical prediction rules had to sytematically collect all data on consecutive patients and have a sufficient number of confirmed pulmonary embolisms (N>50).

Search Date

MEDLINE search between 1966 and March 2003, english language only

Limitations

The criteria for which to use the clinical prediction rules is not explicitly stated (i.e. all patients who enter ER?, patients with acute dyspnea?, patients with risk factors?) Studies are limited to outpatients presenting to ER Validation studies have different protocols for ruling in or out pulmonary embolus Large variation in PE prevelance between validation studies

Updated Search: March 2003 to March 2005

*Perrier et al. Diagnosing pulmonary embolism in outpatients with clinical assessment, D-dimer mearsurement, venous ultrasound, and helical computed tomography: A multicenter management study. The American Journal of Medicine. March 2004; 116:291-299

A prospective cohort study of 1290 patients presenting to 3 general hospital emergency rooms with acute dyspnea or chest pain without other obvious cause. Twenty five percent were excluded for various reasons leaving 965 patients in the study sample. The Geneva criteria was not calculated in 194 of these because an ABG was not available. Furthermore a subset of the patients in this study were used in the Chagnon study which is first study to validate the Geneva criteria.

Reviewed by: Tim Von Fange, Dr. M Lee Chambliss Date: March 2005


2009-10-23

Osteoporosis RCE 2009-10-23
Stroke RCE 2009-10-23
Urinary Tract Infection RCE Final 2009-10-23
Strep Throat RCE 2009-10-23
Shoulder Instablity - RCE Synopsis 2009-10-23
Mole or melanoma RCE 2009-10-23
Sinusitis RCE 2009-10-23
Opiates in Acute Abdomen RCE 2009-10-23
Meningitis - RCE Synopsis 2009-10-23