Association of Body Mass Index With Postoperative Early Adverse Outcomes in Acute Type a Aortic Dissection Patients in Fujian Province, China: A Retrospective Study


 Background: Abnormal body mass index (BMI) has been related to a higher risk of adverse perioperative outcomes in patients undergoing cardiac surgery. However, the effects of BMI in patients with acute type A aortic dissection (AAAD) on postoperative outcomes remain unclear. The aims of this study were to explore the relationships of BMI and postoperative early adverse outcomes in AAAD patientsMethods: Patients who underwent AAAD surgery at Fujian Province Cardiac Medical Research Center from June 2013 to March 2020 were retrospectively evaluated. They were divided into three groups on the basis of Chinese BMI classification established by the World Health Organization: normal group (BMI 18.5-23.9 kg/m2), overweight group (BMI 24-27.9 kg/m2), and obesity group (BMI > 28kg/m2). Preoperative, intraoperative, and postoperative data were collected. Multivariable and univariable logistic regression analysis models were performed to identify whether BMI was independently associated with postoperative adverse outcomes.Results: Of 777 cases, 31.9% were normal weight, 52.5% were overweight, and 15.6% were obese. The percentage of prolonged mechanical ventilation (44.9% vs 55.8% vs 66.1%, respectively; P<0.001) and the median duration of intensive care unit stays (8.1 vs 9.5 vs 12.0 days, respectively; P<0.001) were higher and longer in the overweight and obese group. Multivariable logistic regression analysis demonstrated that a higher risk of postoperative early adverse outcomes in the overweight (odds ratio [OR]: 2.374, 95%CI: 1.647–3.422), and obese patients (OR: 3.659, 95%CI: 2.122–6.308) with reference to the normal BMI patients, and age, heart rate, and surgery duration were also associated with postoperative early adverse outcomes (P<0.05).Conclusion: Overweight and obese patients are independently associated with higher postoperative early adverse outcomes in patients who underwent AAAD surgery.


Background
Obesity is generally considered to be a strong risk factor for morbidity and mortality during the perioperative period in patients with heart disease [1] . Studies have reported obesity patients who underwent cardiac surgery were associated with adverse outcomes in recent years [2,3] . However, some reports suggest that obesity is unable to increase the incidence rate of in-hospital mortality and even reduce it [4,5] . The improved survival and functional outcomes in overweight and obese cardiac surgery patients are termed as the "obesity paradox" in the literature [6] .
So far, the "obesity paradox" has been demonstrated in patients undergoing coronary artery bypass grafting or valve surgery [7][8][9] . However, it has not been fully studied in patients with acute type A aortic dissection (AAAD). Only two studies have evaluated the effect of body mass index (BMI) on postoperative clinical outcomes in patients with AAAD [10,11] , but the study was in western countries whose cases were categorized by the World Health Organization (WHO) standard. Considering ethnic differences, the WHO standard of BMI is not suitable for the Chinese population which the results of the mentioned studies might therefore not be valid for. No existing studies have reported the associations between BMI and early clinical outcomes in AAAD patients undergoing surgery. Moreover, it is unknown whether the "obesity paradox" in Chinese AAAD patients exists. This study adopted the Chinese BMI classi cation standard to determine the relationship between BMI and postoperative adverse outcomes in AAAD patients.

Study population
This retrospective study enrolled 777 consecutive patients aged 18- All patients were divided into three groups on the basis of Chinese BMI classi cation established by WHO: BMI = 18.5-23.9 kg/m 2 (normal group); BMI = 24-27.9 kg/m 2 (overweight group); BMI 28 kg/m 2 (obesity group). Underweight patients (BMI<18.5) were excluded to avoid biased results because of the small sample size. Ethics approval has been obtained from the ethics committee of Fujian Medical University Union Hospital (approval number:2013002) and conformed to the Declaration of Helsinki.

Data collection
All data were collected by investigators who had been uniformly trained, including preoperative characteristics, intraoperative, and postoperative data. The endpoints were postoperative early adverse outcomes during hospitalization or in the rst 30 postoperative days, including in-hospital mortality, prolonged mechanical ventilation (MV), respiratory failure, hypoproteinemia, acute renal failure, acute liver failure, aortic dissection rupture, gastrointestinal bleeding, neurological complications (cerebral hemorrhage, cerebral infarction, epilepsy, and delirium), arrhythmia, septicemia and Multiple Organ Dysfunction Syndrome (MODS). Besides, the duration of intensive care unit (ICU) stays and hospital stays were also recorded.

Criteria
In-hospital mortality was de ned as both all-cause deaths occurring during the hospitalization or in the rst 30 postoperative days [12] . Prolonged mechanical ventilation was de ned as the duration of mechanical ventilation ( 48hours) [13] . MODS was de ned as the Sequential Organ Failure Assessment (SOFA) score of ≥6 on two or more consecutive days after surgery [14] . Postoperative complications were identi ed as any of the following: respiratory failure, hypoproteinemia, acute renal failure, aortic dissection rupture, acute liver failure, gastrointestinal bleeding, neurological complications (cerebral hemorrhage, cerebral infarction, epilepsy, and delirium), septicemia, MODS, prolonged MV, and arrhythmia. Postoperative early adverse outcomes were de ned as one or two of the following: in-hospital mortality and postoperative complications.

Statistical Analysis
Continuous variables were presented as mean±SD or median (interquartile range), based on Gaussian distribution, and categorical variables were presented as numbers/percentages. Continuous variables were analyzed using analysis of variance (ANOVA) or the Kruskal-Wallis tests, and categorical variables were done using Chi-squared test or Fisher's exact test. Univariable and multivariable logistic regression analyses were done to assess the relationship between BMI and postoperative early adverse outcomes, and variables with P ≤ 0.1 at univariable analysis were included in the multivariable analysis. Data were performed using the Statistical Package for the Social Science (SPSS version 23.0). P < 0.05 was considered to be statistically signi cant.

Results
From June 2013 to March 2020, a total of 807 AAAD patients aged 18-80 years received surgical treatment. And a cohort of 777 AAAD patients met inclusion criteria. The ow chart of patient inclusion is demonstrated in Figure 1. Table 1 shows the baseline characteristics of the three groups. Among the 777 AAAD patients, 248 (31.9%) were classi ed into the normal group with a BMI of (21.7 ± 1.5) kg/m 2 , 408 (52.5%) were classi ed into the overweight group having a BMI of (25.4 ± 1.0) kg/m 2 , and 121(15.6%) patients were classi ed into the obese group with a BMI of (31.6 ± 6.5) kg/m 2 . The oldest age was seen in the normal group (age of 55.7 ± 11.8 years, P 0.05) as compared with the overweight and obesity groups, and with a high proportion of females, and drinker P 0.05). However, there was no signi cant difference in the history of smoker, and hypertension, coronary artery heart disease (CHD), type 2 diabetes (DM2), and previous cardiac surgery (PCS) among the three groups (P > 0.05).
Patient preoperative, intraoperative, and postoperative clinical variables are summarized in Table 2. The mean of systolic blood pressure (BP), diastolic BP, and pulse pressure were signi cantly lower in the normal group compared with the overweight and obesity groups (P < 0.05). There were signi cant differences in serum creatinine (Scr), hemoglobin, and white blood cell (WBC) among the three groups (P 0.05). No signi cant differences were observed among other preoperative and operative variables (P 0.05).
Compared with the normal group, the ICU stay and hospital stay was longer in the overweight group and obesity group (P 0.05). But the in-hospital mortality was no signi cant difference between the three groups (16.9% vs 21.8% vs 24.0%, respectively; P=0.198). No signi cant difference was present in respiratory failure, hypoproteinemia, acute renal failure, aortic dissection rupture, acute liver failure, gastrointestinal complications, neurological complications, septicemia, arrhythmia, and MODS among the three groups (P 0.05) except prolonged MV, postoperative complications, and adverse outcomes (P 0.05).
The results of the univariable logistic regression analysis of risk factors are presented in Table 3. BMI would predict postoperative adverse outcomes (P<0.05). Besides, age, hypertension, surgery duration, cardiopulmonary bypass (CPB), aortic cross-clamping duration, diastolic BP, and heart rate were related to postoperative adverse outcomes (P <0.05).
The results of the multivariable logistic regression analyses are presented in Table 4. After adjusting for aortic clamping time, CPB time, diastolic BP, and the history of hypertension, BMI, age, heart rate, and surgery duration were associated with postoperative adverse outcomes (P <0.05). It also demonstrated a higher risk of postoperative early adverse outcomes in the overweight (OR 2.374, 95%CI: 1.647-3.422), and obese patients (OR 3.659, 95%CI: 2.122-6.308) with reference to the normal BMI patients.

Discussion
Several previous studies have investigated the impact of BMI in the prognosis of cardiac disease [10,11] , but the conclusions remain in dispute. The present study is innovative to demonstrate the impact of BMI on postoperative early adverse outcomes in Chinese AAAD patients undergoing surgery. A total of 777 AAAD patients were included in the study, which we found that the incidence of in-hospital mortality was 20.6%. We observed patients who were overweight and obese were signi cantly higher in postoperative early adverse outcomes with longer ICU stays and hospital stays compared with the normal group. Besides, age, heart rate, and surgery duration were signi cantly and strongly related to postoperative early adverse outcomes in patients following AAAD surgery, even after adjusting for other risk factors.
It is unclear why overweight and obese affects the prognosis of AAAD patients, but variability in adherence to DNA methylation or adipose tissue in overweight and obese patients may explain our observations. Recently, Dr. Simone Wahl and colleagues carried out studies [15] about epigenome-wide association amongst 5,387 individuals from three different population, demonstrating that alterations in DNA methylation are more likely to be the consequence of overweight or obesity, which methylation loci are related to lipid and lipoprotein metabolism, substrate transport, and in ammatory pathways. In addition, large amounts of bioactive mediators can be released from the adipose tissue, affecting blood pressure, in ammation, and other changes, and lead to endothelial dysfunction and arteriosclerosis [16] .
Based on these analyses, it could be concluded that overweight and obese may affect postoperative in ammatory reactions in AAAD patients, thus lead to an increase in mortality and adverse outcomes.
Although there is a huge amount of literature in support of the concept of the 'obesity paradox', which indicates that a protective effect, or no effect was observed in overweight or even obese on adverse outcomes [6] . However, a recent network meta-analysis suggested that the 'obesity paradox' disappeared in Asians, which is consistent with our results [8] . Kreibich et al. pointed out that obesity was not signi cantly associated with postoperative adverse outcomes in AAAD patients [11] . Lio et al. reported that obese patients were more likely to develop postoperative mortality and adverse outcomes compared to non-obese patients [10] . Our ndings indicate that overweight and obese are at increased risk of postoperative early adverse outcomes in AAAD patients, which wasn't consistent with their studies. These inconsistencies might be explained by the three follows: rst, BMI showed great differences in different racial persons and especially between the Caucasian and the yellow race [17] ; second, the BMI cutoffs of underweight, normal weight, overweight and obese in Chinese are different from those in other countries; third, there are differences between the studies at baseline, for example, patients with acute aortic dissection (AAD) in China had an earlier onset about ten years than the International Registry of Acute Aortic Dissection [18] . Finally, the body fat content of Asian people is generally higher than white people of the same gender, age and BMI even if their BMI is less than 25 kg/m 2 , they are more likely to develop type II diabetes and cardiovascular disease [19] . Therefore, it is not surprising that the results are quite different.
In this study, we found that the postoperative adverse outcomes in AAAD patients increased with increasing age. It has been reported that age was an established risk factor for perioperative mortality in AAAD patients [20,21] . Then we have a deep analysis of reasons. In the rst instance, elderly patients are often poorly performed having asymptomatic clinical features in comparison with the young patients which usually led to a delay in diagnosis and treatment. It may also re ect that patients do not obtain timely medical attention [22] . Secondly, it is known that older patients often suffered from concomitant comorbidities, such as hypertension, coronary heart, diabetes, and a history of cardiac surgery. That means older patients are more serious when they arrive at the hospital, which may thereby lead to worse surgical prognosis. Moreover, compared with younger, older patients were more seem to have a poor prognosis due to their low cardiopulmonary reserve, and the poor responses to cardiopulmonary bypass surgery, anesthesia, various drugs, and uid infusion. The elderly people are more physically to have weakness [23,24] , which can cause physiological disorders including heart rate variability, systemic in ammatory states, a decline in immune function, and hormonal changes [25][26][27] , and thus increased risk of adverse outcomes.
Another notable nding was that independent risk factors for adverse outcomes were heart rate and surgery duration, which was consistent with the results of the previous studies [28,29] . As we known, heart rate is a powerful predictor of early adverse outcomes which increased linearly with the heart rate [30] . The effect of pain stimulus and stress response in AAAD patients can increase the heart rate. It not only improves the metabolism and accelerates the heart rate, but also lengthens the systolic period, shortens the diastolic period, and reduce the coronary artery perfusion [31,32] . Thus, the altered systemic hemodynamics could break the balance of the circulatory system and make patients more frequently affected by adverse outcomes. The strong association between surgery duration and early adverse outcomes has been observed by other investigators as well [33] . The longer the surgery duration, on the one hand, re ects the severity of the patient's condition, on the other hand, this may be vulnerable to overdose anesthetic drugs, greater volumes of blood transfusion, or the disturbance of electrolytes or acid-base balance [34] . It will affect the organ perfusion, for example cerebral perfusion, and the injury will increase with the perfusion time. Besides, due to the demanding anatomy and di culty of arterial access in overweight or obese patients, the di culty of the procedure is increased. For the above reasons, the incidence of postoperative adverse outcomes will inevitably increase.
A further important nding is the longer ICU stay and the higher proportion of prolonged MV in the overweight and obese group, whose nding was compatible with those of other studies. [35,36] . The reasons could be as follows: (1) The proportion of smokers is higher in overweight and obese groups when compared to the normal group respectively. The decreased pulmonary function in smokers may lead to the inconsistency of baseline lung function on admission in different BMI groups. In this study, the group with a higher proportion of smokers has a higher incidence of prolonged MV. (2) The adipose tissue creates a number of estrogens in overweight and obese patients that play an important role in pulmonary hypertension and remodeling [37] . It generally focuses on the changes in pulmonary and chest wall structures, which led to the increase of residual lung volume and chest wall impedance, a decrease of lung compliance and ventilation driving force, and abnormal ventilation-perfusion [38] . Furthermore, reduction or loss of lung volume due to the weight of adipose tissue can occur in obesity or overweight patients during operation [39] , leading impaired lung function and delayed extubation. And prolonged time on mechanical ventilation may result in prolonged ICU stay and vice versa [40] . In summary, medical staff should establish reasonable ventilation strategies to shorten the length in mechanical ventilation and ICU stay.
However, there are several limitations. First, this was a single-center retrospective study which has its inherent shortcomings, and the ndings may not be generalizable to other regions. Second, the measure of body proportion relies solely on BMI which does not yield information on the distribution of body fat.
Third, the follow-up period (30 days) in this study was relatively shorter, further studies with longer followup periods would be explored to provide additional insight into the impact of BMI on long-term prognosis.
Finally, we excluded the underweight patients (< 18.5 kg / m 2 ) with a small sample size in order to avoid the error of the results, which may affect the results when their cases are large enough.

Conclusions
In summary, the present study demonstrates that overweight and obese patients are independently associated with higher postoperative early adverse outcomes in patients who underwent AAAD surgery. And age, heart rate, and surgery duration are also risk factors for postoperative adverse outcomes. The focus should be placed on overweight and obese patients in risk assessment before surgery. Further works need to provide additional insight into the impact of BMI on long-term prognosis.

Consent for publication Not applicable
Availability of data and materials The datasets used and analyzed during the current study are available from the corresponding author on reasonable request.
Competing interests The Authors declare that there is no con ict of interest.
Funding This work was supported by grants from the Joint Finds for the innovation of science and Technology, Fujian province (Grant number: 2017Y9052) and Guiding project of science and technology department of Fujian province (Grant number:2017Y0038).
Authors' contributions L-YL and S-LL assessed the delirium of the subjects. Y-CP and QC collected sociodemographic and clinical data. X-ZH, Y-JL, and L-WC analyzed and interpreted the data; L-YL and QC drafted the manuscript; all authors critically revised the manuscript. All authors read and approved the nal manuscript. Patients' ow chart of the cohort. AAAD, acute type A aortic dissection.