Trends and Outcomes of Noninvasive and Invasive Ventilation in Acute Respiratory Failure Due to Pneumonia

Abstract

Listen

Translate article

SESSION TITLE: Airway/Mechanical Ventilation SESSION TYPE: Original Investigation Slide PRESENTED ON: Wednesday, November 1, 2017 at 08:45 AM - 10:00 AM PURPOSE: Non-invasive ventilation (NIV) started gaining popularity in early 2000s and since then NIV use in acute respiratory failure (ARF) has been increasing through out the world. Currently, NIV is first choice to treat ARF due to COPD, CHF and also to prevent post extubation failure in COPD patients. Role of NIV in PNA is not well defined. On the other hand, invasive mechanical ventilation (IMV) has been the cornerstone of ARF due to PNA. The purpose of this study is to compare over all and annual trends and outcomes associated with NIV use in PNA and its impact on IMV, length of stay (LOS), cost and mortality in USA METHODS: We used the Nationwide Inpatient Sample (NIS) for years 2002-2012. The Clinical Classifications Software (CCS) for ICD-9-CM was used to extract data for pneumonia as the primary diagnosis on discharge. ICD-9 code 93.90 was used for NIV and 96.70, 96.71 and 96.72 were used for IMV. Pearson and spearman correlation coefficients were calculated for annual trends in in-hospital mortality, length of stay (LOS), and total charges for pneumonia patients, along with trends in the use of NIV and IMV. Mortality, LOS, and total charges were also compared between patients with NIV only, with NIV and IMV and IMV only, along with demographics. RESULTS: We extracted 4,162,668 cases of PNA from 2002 to 2012. NIV was used in 102,338(2.4%) and IMV in 232,087 (5.5%) cases. Mean age of the population who had NIV and IMV was (mean/SEM) {66.53(0.06) vs 61.99(0.04), P<0.001} respectively. Further analysis revealed that the use of NIV was lowest in population 80 years or older. There were no gender or racial differences noted. LOS was higher in IMV as compared to NIV (mean/SEM) {15.5(0.03) vs 10.1(0.03), P<0.001}, mortality was higher {0.29(0.0) vs 0.17(0.0) P<0.001} and cost was also higher in IMV group {124141.5(311.71)vs 69967(293.14), P<0.001}. Because of the confounding factor that patients who fail NIV would be intubated and hence resulted in higher mortality and LOS for IMV, we looked for differences in LOS and mortality and cost for PNA with IMV alone and NIV + IMV. LOS and cost was higher in NIV + IMV group (mean/SEM) {17.5(0.12) vs 15.4 (0.04) P<0.001} & {142444.8(1114.3) vs 123575.67(324.1) P<0.001) respectively. Mortality is slightly lower in the NIV+IMV as compared to IMV alone {0.27(0) vs 0.29(0)}. Annual incidence of PNA has been increasing in the hospitalized patients since 2002 to 2012 but it is not significant when adjusted for rate of growth in population. The use of NIV has increased significantly from 2002 (0.97%) to 2012 (4.37%). The trend for IMV in pneumonia is unchanged (5.6%) and there is significant improvement in mortality from 6.5% in 2002 to 4.3% in 2012 CONCLUSIONS: Our study shows the increasing trend of using NIV for PNA in USA. This trend highlights that there is an observable clinical benefit of NIV in PNA. There is 12% mortality benefit with NIV as compared to IMV but it doesn’t decrease intubation rates, this is similar to prior studies CLINICAL IMPLICATIONS: Our study will encourage physicians to consider earlier institution of NIV in PNA. it will promote further research to better define patients who would benefit from NIV DISCLOSURE: The following authors have nothing to disclose: Faraz Siddiqui, Saqib Abbassi, A H Siddiqui, Nauree Narula, Amina Saqib, Michel Chalhoub No Product/Research Disclosure Information