QUALITATIVE AND QUANTITATIVE EMPHYSEMA IS ASSOCIATED WITH HOSPITAL ADMISSION IN A LARGE CT LUNG SCREENING COHORT

Abstract

SESSION TITLE: Tuesday Electronic Posters 3 SESSION TYPE: Original Inv Poster Discussion PRESENTED ON: 10/22/2019 01:00 PM - 02:00 PM PURPOSE: Stratifying risk of hospital admission can improve healthcare efficiency through better prospective direction of resource utilization. We conducted a prospective study analyzing the association between emphysema, assessed both qualitatively and quantitatively, and risk for hospital admission in a large clinical CT lung-screening (CTLS) cohort METHODS: Patients undergoing CTLS at Lahey Hospital and Medical Center between January 1, 2012 and September 30, 2014 who had a Lahey primary care physician were eligible for this study (n=1703). Follow up data was collected through September 30, 2017. Quantitative emphysema scores, defined as the percentage of low-attenuation areas measuring less than 950 Hounsfield units (%LAA -950), was obtained from baseline CTLS exams using Chest Imaging Platform workstation v 3.4.8 (www.chestimagingplatform.org). Qualitative emphysema scores, graded as none, mild, moderate, or marked, were reported on each baseline CTLS exam by a trained CTLS radiologist. Quantitative scores were divided by quartiles. The lowest quartile for %LAA and none for qualitative scores served as reference groups. Hospital admission data was collected through Lahey administrative coding data, and COPD, MI, CHF, PNA and stroke were characterized based on diagnosis codes per 2018 Centers for Medicare & Medicaid Services condition-specific measures. Cox proportional hazard regression for admission was adjusted for variables with a univariate p<0.2. Significance levels were set at p<0.05. All statistical analyses were performed using STATA14.1 software. RESULTS: 1,681 patients were included in our analysis with a mean age of 62.5 ± 6.2 years. 56.1% of the patients were male and 98.2% were white. A total of 508 (30%), 59 (3.5%) and 48 (2.9%) patients had at least one all-cause, COPD and PNA admission, respectively. Quartile 4 of %LAA and marked emphysema was associated with all cause hospital admission when adjusted for age, BMI, pack years, years quit, cancer and eligible group, HR 1.31 [1.02, 1.68, p=0.032] vs. 2.20 [1.54, 3.15, p<0.001]. CONCLUSIONS: Our results demonstrate that CTLS patients are at high risk for hospital admission and that both quantitative and qualitative emphysema is associated with all-cause hospital admission and with principle diagnosis of COPD and PNA. CLINICAL IMPLICATIONS: Quantitative emphysema scoring could be incorporated into admission prediction models to better direct resources towards highest-risk patients DISCLOSURES: no disclosure on file for Raul Estepar; No relevant relationships by Melissa Gawlik, source=Web Response No relevant relationships by Lee Gazourian, source=Web Response No relevant relationships by William Long, source=Web Response Speaker/Speaker's Bureau relationship with Covidien Please note: $1001 - $5000 Added 03/15/2019 by Andrea McKee, source=Web Response, value=travel Removed 03/15/2019 by Andrea McKee, source=Web Response Speaker/Speaker's Bureau relationship with Covidien Please note: $1001 - $5000 Added 03/15/2019 by Andrea McKee, source=Web Response, value=Salary Speaker/Speaker's Bureau relationship with Covidien Please note: $1-$1000 Added 03/15/2019 by Andrea McKee, source=Web Response, value=Travel No relevant relationships by Brady McKee, source=Web Response No relevant relationships by Lori Lyn Price, source=Web Response No relevant relationships by Shawn Regis, source=Web Response No relevant relationships by Giulia Rizzo, source=Web Response No relevant relationships by Ava Sanayei, source=Web Response No relevant relationships by Sandeep Somalaraju, source=Web Response No relevant relationships by Cristina Stefanescu, source=Web Response No relevant relationships by William Thedinger, source=Web Response Advisory relationship with Philips HealthTech Please note: $1001 - $5000 Added 03/15/2019 by Christoph Wald, source=Web Response, value=Honoraria<br no disclosure on file for George Washko