Pulmonary artery size on ct and mortality risk: a dose-response analysis

CHEST(2023)

引用 0|浏览2
暂无评分
摘要
SESSION TITLE: Pulmonary Vascular Disease Posters 5 SESSION TYPE: Original Investigation Posters PRESENTED ON: 10/11/2023 12:00 pm - 12:45 pm PURPOSE: An enlarged pulmonary artery (PA) on CT imaging has been associated with mortality risk in several cohorts. Two measures of abnormal PA size have been defined based on the distribution of sizes in healthy individuals: PA diameter (PAd; abnormal >27mm [female], >29mm [male]) and PAd ratio to the ascending aorta (PA:AA; abnormal > 0.9). It is not known if these thresholds correspond to the PA size beyond which risk of adverse outcomes increases. We hypothesized that mortality risk would increase in a size-dependent manner for both metrics. The average AA diameter increases faster with age; thus we also hypothesized that higher PA:AA ratios would be needed to confer excess mortality risk in younger patients. METHODS: We reviewed 1000 randomly-selected CTPAs completed in 2009-2010 in two EDs within an integrated healthcare system in Utah. Patients with acute pulmonary emboli and duplicate scans were excluded. We measured the PAd and AA at the same level on each CTPA. Mortality status was obtained from the EHR and death registries. Restricted cubic splines were used to model the association of PA:AA and PAd with death in Cox regression models that also control for age and sex. Hazard Ratios (HR) are compared to the healthy population median value (PA:AA 0.77; PAd 25.1mm). RESULTS: The median age (n=912 included) was 51y (interquartile range [IQR] 37-67) and 36.5% were male. 28.8% died during follow-up (median 7 years). The median age of the n=318 patients with high PA:AA was 41.5 (IQR. 30-61). Patients with enlarged PAd were older (n= 258, median age 62 IQR 46-77). The PA:AA ratio associated with the lowest mortality risk was higher in younger patients. For the entire sample, all ratios below 0.75 predicted equivalent and minimized risk, while patients <50y had mortality risk nadir at PA:AA of 0.83. Similarly, the PA:AA conferring a HR of 1.5 was higher in younger patients (PA:AA ratio of 0.96 entire sample; PA:AA 1.05 under age 50y) with near log-linear increase for further size increase in all ages. The PAd associated with lowest risk of death was also smaller in the entire cohort (PAd of 22.1 mm [F] and 24.1 mm [M]) than in patients <50y (25.6 mm [F] and 27.6 mm [M]). The size associated with HR 1.5 was also larger in younger patients (entire cohort: PAd of 28.8 mm [F] and 30.8 mm [M]; under 50y: 30.7 mm [F] and 32.7 mm [M]) CONCLUSIONS: We found that both enlarged PAd and elevated PA:AA confer excess mortality risk in a dose-dependent manner for patients receiving CTPAs in the ED. Therefore, dichotomization as “normal or abnormal” may discard prognostic information. For both PAd and PA:AA, a larger PA size was required in younger patients before mortality risk increased. Whether this is due to differing timelines of mortality risk, increased vascular compliance in younger adults, or other mechanisms warrants further investigation. CLINICAL IMPLICATIONS: Dichotomized classification of pulmonary artery size on CTPA has previously been shown to predict mortality risk, but we show that age adjustment and accounting for the degree of enlargement may improve predictive performance. DISCLOSURES: Consultant relationship with BD Please note: $5001 - $20000 by Joseph Bledsoe, value=Consulting fee Advisory Committee Member relationship with Baxter Health Please note: 11/4-11/21 2022 Added 03/29/2023 by Joseph Bledsoe, source=Web Response, value=Honoraria No relevant relationships by Meghan Cirulis No relevant relationships by Karen Conner I received research funds from the entit relationship with Janssen Pharmaceuticals Please note: 10/1/2019-9/30/2022 Added 03/29/2023 by Mark Dodson, source=Web Response, value=Grant/Research No relevant relationships by C Elliott No relevant relationships by Daniel Knox No relevant relationships by Brian Locke No relevant relationships by Brittany Scarpato No relevant relationships by Gregory Stoddard
更多
查看译文
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要