Description

van der Valk et al identified clinical predictors for distinguishing bacterial from nonbacterial exacerbations of chronic obstructive pulmonary disease (COPD). This can help decide if antibiotics should be used or withheld in initial management of the patient pending further studies. The authors are from Medisch Spectrum Twente, Regional Laboratory of Public Health, and University Medical Centre Nijmegen in The Netherlands.

A bacterial infection was associated with the exacerbation in 19% of cases.

Parameters:

(1) Gram stain of sputum

(2) number of exacerbations in past 12 months

(3) clinically relevant decrease in FEV1 (> 12% decrease and > 200 mL)

decrease in FEV1 in mL in 1 second =

= (baseline FEV1 in mL) - (admission FEV1 in mL)

percent decrease in FEV1 =

= ((baseline FEV1) - (admission FEV1)) / (baseline FEV1) * 100%

Parameter

Finding

Points

Gram stain

negative

0

positive

1

number of exacerbations in past 12 months

0 or 1

0

>= 2

1

clinically relevant decrease in FEV1

no

0

yes

1

where:

• Change in FEV1 was used to define lung function. A clinically relevant decrease was > 12% and > 200 mL from baseline.

• A positive sputum Gram stain was based on identification of pathogenic or potentially pathogenic bacteria. White blood cells were not considered.

total number of predictors =

= SUM(points for all 3 parameters)

Interpretation:

• minimum number of predictors: 0

• maximum number of predictors: 3

Number of Predictors

Percent with Bacterial Exacerbation

Percent with Nonbacterial Exacerbation

0

0%

100%

1 or 2

20%

80%

3

67%

33%