The Science Journal of the American Association for Respiratory Care

2008 OPEN FORUM Abstracts

EVALUATION OF A HYPOXEMIC/HYPERCAPNIC ALGORITHM FOR IDENTIFICATION AND TREATMENT OF PATIENTS FOLLOWING APPLICATION OF NPPV

Paul D. Luehrs1, Jack Edge1, Martin Rohrer1, Lana Shaw1, Sean Tettenhorst1, Terrence Coulter1



Background: In 2004 we conducted a chart review (Dec 2003- Feb 2004) of patients with Acute Hypoxemic Respiratory Failure (AHRF). ARDS (Acute Respiratory Distress Syndrome) is often misdiagnosed and consequently treated with noninvasive positive pressure ventilation (NPPV) for prolonged periods of time. Our data showed a high mortality rate for ARDS patients (67% with an average of 46 hours on NPPV. In Nov. 2004 we created a hypoxemic/hypercapnic algorithm to guide clinicians in the identification and treatment of these two patient populations.

Methods: We reviewed all NPPV set-ups (n=176) from Dec 2006-Feb 2007 and classified patients as having either AHRF (n=52), acute hypercapnic respiratory failure (n=69), or meeting our exclusion criteria (n=55). AHRF was determined prior to NPPV initiation and defined as having a P/F < 200 and a PaCO2 < 45 mmHg. Patients were excluded if they had been intubated; required CPR prior to NPPV; had a tracheostomy; had no ABG's; or had recent facial, esophageal or cranial trauma/surgery. NPPV settings were determined by the respiratory therapist at the bedside using our NPPV protocol. NPPV pressures were titrated up to an IPAP of 25 cmH2O and EPAP of 15 cmH2O as needed.

Results: During the period of Dec 2006-Feb 2007 we had 176 NPPV set-ups. 52 patients met our criteria for AHRF with the absence of hypercapnia. Four subsets of patients were identified: pneumonia (PNA) (n=20); cardiogenic pulmonary edema (CPE) (n=13); ARDS (n=9); and Other (n=10). We decreased the # of hours our PNA patients were on NPPV from 46.5 to 26.9 and increased the percent intubated from 11% to 30% after algorithm implementation. Hospital mortality for PNA patients decreased from 13% to 5% and ICU length of stay (LOS) increased from 4.5 to 5.5 days, while the overall hospital LOS decreased from 13.5 to 11.9 days. Our ARDS patients had a 20% decrease in the # of hours on NPPV and an 18.5% increase in invasive positive pressure ventilation (IPPV) hours. There was a decrease in hospital mortality of 47.8% post algorithm in our ARDS patients.

Conclusion: Based on the results of the chart review we identified an improvement in the outcomes of our PNA and ARDS patients. The process of identifying AHRF patients and treating them appropriately with the aid of an algorithm following NPPV application should be investigated further. Our increase in mortality rate for CPE should be scrutinized.