Measures Predict Need for Ventilation in COPD

PHILADELPHIA — Three easily obtained clinical measures can predict who will probably need mechanical ventilation among newly hospitalized patients with an acute exacerbation of chronic obstructive pulmonary disease, based on an analysis of data from almost 100,000 patients.

Patients presenting to an emergency department with an acute exacerbation of COPD with a BUN level above 25 mg/dL, altered mental status, and a pulse of more than 109 beats/minute had about an 11% rate of mechanical ventilation during their index hospitalization, Dr. Andrew F. Shorr reported at the annual meeting of the American College of Chest Physicians.

In contrast, similar patients who lacked all three of those clinical signs had a 0.3% rate of mechanical ventilation later during their hospitalization, said Dr. Shorr, associate director for pulmonary and critical care medicine at the Washington (D.C.) Hospital Center.

Determining a patient's risk for needing mechanical ventilation early during hospitalization is important, Dr. Shorr said in an interview, because “if you know there is a high risk, you can arrange closer monitoring and an earlier start to ventilatory support. That's better than waiting until the patient is so sick that intubation is tenuous.”

Also, “you don't want to put a patient [who has a high risk for needing mechanical ventilation] in an unmonitored room,” he added. “With identification of high risk, you can put them in higher-level care.”

Dr. Shorr and his associates reviewed 98,036 patients admitted to any of 191 U.S. hospitals for acute exacerbation of COPD during 2004–2006. The sample was randomly divided into a derivation cohort and a validation cohort.

The researchers then took the derivation cohort and used classification and regression tree analysis to assess a long series of demographic, clinical, and hospital characteristics to find parameters that best distinguished patients who required mechanical ventilation from those who did not.

That analysis showed that three parameters worked well together to segregate patients into low- and high-risk groups. The three parameters were then tested using the validation cohort, and the results confirmed the initial finding. (See box.)

In both cohorts, the three parameters were able to account for slightly more than three-quarters of the risk for mechanical ventilation, Dr. Shorr said.

“These three markers don't have anything to do with the lungs,” he noted. “Our hypothesis is that they are simple markers for end-stage organ dysfunction.” A BUN level of greater than 25 mg/dL is a marker for volume depletion. Altered mental status is a marker for a patient who is hypoxic or hypercarbic. And a pulse rate of more than 109 beats/minute is a marker for shock, hypoxia, or acidosis.

In addition, all three are simple measures that don't require blood gas measurements or invasive testing, and they can be assessed with little interobserver variability.

At last year's annual meeting of the American College of Chest Physicians, Dr. Shorr and his associates reported that the same three measures could help predict the risk of death in patients hospitalized for an acute exacerbation of COPD. Patients who met all three criteria had a mortality rate of nearly 14%, compared with a rate of less than 1% among patients who met none of the three criteria.

The new finding that the scoring method also helps predict need for mechanical ventilation “adds to the biologic validity” of the assessment, Dr. Shorr said, adding that predicting the need for mechanical ventilation can affect management of a patient as much as predicting a high risk of death.

'If you know there is a high risk, you can arrange closer monitoring and an earlier start to ventilatory support.' DR. SHORR

ELSEVIER GLOBAL MEDICAL NEWS