Prognostic Significance of Preoperative Neutrophil-to-Lymphocyte Ratios in Surgery for Acute Type A Aortic Dissection

P0268 

Abstract Submission 

Alexander Mills (1), Gregory Estrera (1), Akiko Tanaka (1), Lucas Ribe (1), Yuki Ikeno (1), Harleen Sandhu (1), Charles Miller (1), Anthony Estrera (1)

(1) McGovern Medical School at UTHealth, Houston, TX

Alexander Mills    -  Contact Me
McGovern Medical School at UTHealth

Gregory Estrera    -  Contact Me
McGovern Medical School at UTHealth

Akiko Tanaka    -  Contact Me
McGovern Medical School at UTHealth

Lucas Ribe    -  Contact Me
McGovern Medical School at UTHealth

Yuki Ikeno    -  Contact Me
McGovern Medical School at UTHealth

Harleen Sandhu    -  Contact Me
McGovern Medical School at UTHealth

Charles Miller    -  Contact Me
McGovern Medical School at UTHealth

*Anthony Estrera    -  Contact Me
McGovern Medical School at UTHealth

Alexander Mills    -  Contact Me
University of Texas Health Science Center at Houston (UTHealth Houston)

Objectives:
Neutrophil-to-lymphocyte ratios (NLR) are an inexpensive and readily available biomarker that measures ongoing acute inflammation. We hypothesized that an elevated preoperative NLR was an independent predictor of worse outcomes following surgical repair of acute type A aortic dissection, especially if it continued to be elevated 24 hours from symptom onset.

Methods:
We retrospectively evaluated consecutive patients presenting with acute type A aortic dissection who underwent surgical repair from 2001 to 2020 at a single institution. Preoperative NLR values were calculated. Median and quartile values of NLR were used as cutoff points, which were evaluated by a receiver operating curve and area under the curve. A subgroup of patients who underwent repair 24 hours after symptom onset were evaluated to see if a persistently elevated NLR would portend worse outcomes. The primary outcome was in-hospital death. Secondary outcomes were stroke, atrial fibrillation, coagulopathy, acute renal failure, and unplanned reoperation. Univariate and multivariate logistic regression were used to evaluate preoperative NLR as a predictor variable for these postoperative outcomes in the overall cohort as well as the pre-identified subgroup.

Results:
613 of the 650 patients (94.3%) had preoperative NLR values. The median preoperative NLR value was 7.8 (interquartile range: 4.6-12.3). The primary outcome (in-hospital death) occurred in 94 patients (14.5%), and 317 (48.7%) underwent repair more than 24 hours from symptom onset. For all patients, the median preoperative NLR was slightly higher in the primary outcome group but not statistically significant (8.0 vs. 7.7, p=0.325). For those that presented after 24 hours from symptom onset, the preoperative NLR was much higher in the primary outcome group (8.6 vs. 6.8, p=0.037). Median NLR value of 7.8 was used as a cutoff point for predicting in-hospital death [AUC; 0.620 (0.471-0.769)]. In patients who underwent repair after 24 from onset of symptoms, NLR of 7.8 was a significant predictor of the primary outcome, coagulopathy, and acute renal failure following repair (Figure). On multivariate analysis, malperfusion and age >70 were the only predictors that remained significant for in-hospital death when combined with median preoperative NLR and emergent procedure (Figure).

Conclusions:
Preoperative NLR is an accessible and inexpensive inflammatory biomarker with prognostic implications in surgical repair of ATAAD. It appears most useful to predict postoperative complications in patients undergoing repair more than 24 hours from symptom onset, which might signify ongoing acuity in illness and portend worse postoperative outcomes.

Dissection