Diabetes does not increase in-hospital or short-term mortality in patients undergoing surgical repair for type A aortic dissection: insight from the national readmission database Authors: Hamza Chaudhry, Soha Dargham, Amin Jayyousi, Jassim Al Suwaidi, Charbel Abi Khalil Published: 2024 ABSTRACT AbstractBackgroundPrevious studies have reported a protective effect of type 2 diabetes on the incidence and progression of aortic aneurysms. We investigated whether this protective effect extends to aortic dissections.MethodsData from the US Nationwide Readmission Database (2016–2019) were analyzed. Patients admitted for open surgery repair of acute type A aortic dissection (TAAD) were initially analyzed (index group). Those discharged alive were followed for up to 30 days (readmission group). The co-primary outcomes were in-hospital and 30-day mortality.ResultsBetween 2016 and 2019, 7,324 patients were admitted for open surgical repair of acute TAAD, of whom 965 (13.2%) had diabetes. Patients with diabetes were older and had a higher prevalence of obesity, hypertension, smoking, dyslipidemia, and chronic kidney disease (CKD). 15.2% of patients with diabetes and 14.6% without diabetes died; hence, diabetes did not have a significant impact on in-hospital mortality (adjusted odd ratio [aOR] = 1.02 [0.84–1.24]). Similarly, diabetes was not associated with a higher adjusted risk of atrial fibrillation (aOR = 1.03 [0.89–1.20]), stroke (aOR = 0.83 [0.55–1.26]), cardiogenic shock (aOR = 1.18 [0.98–1.42]), but increased the risk of acute renal failure (aOR = 1.20 [1.04–1.39]). Within 30 days of discharge, 154 (15.9%) patients with diabetes and 952 (15%) from the non-diabetes group were readmitted. Readmitted patients with diabetes were older and had a higher prevalence of cardiovascular comorbidities. We didn’t observe any significant difference in the adjusted risk of 30-day mortality between the diabetes and non-diabetes groups (adjusted hazard ratio [aHR] = 0.81 [0.41–1.60]). However, diabetes was associated with a lower risk of readmission (aHR = 0.81 [0.68–0.97]). Age was the most significant predictor of all outcomes. CKD was the most significant predictor of 30-day mortality, with the risk increasing five-fold in patients with diabetes (HR = 5.58 [2.58–6.62]. Cardiovascular-related conditions were the most common causes of readmission in both groups. However, respiratory-related conditions were more prevalent in the diabetes group compared to the non-diabetes group (19.5% vs. 13%, respectively, p = 0.032).ConclusionsDiabetes does not increase in-hospital or short-term mortality in patients undergoing surgical repair for Type A aortic dissection.Graphical abstract Keywords: Aortic dissection, Type A aortic dissection, Diabetes, Vascular medicine, Cardiology Background Aortic dissection is a life-threatening cardiovascular condition defined by a tear in the aorta’s inner layer, separating the aortic layers and forming a false lumen [1]. Dissections are classified according to the anatomical location of the tear, with Type A aortic dissections (TAAD) involving the ascending aorta and Type B aortic dissections (TBAD) involving the descending aorta. The most significant risk factor for aortic dissection is hypertension, which is present in approximately 80% of patients with aortic dissection and accounts for 54% of the population-attributable risk [2]. Other important risk factors include advanced age, male sex, bicuspid aortic valve, genetic conditions such as Marfan syndrome, and smoking [2]. The incidence of aortic dissection is estimated to be around 2.9 cases per 100,000 person-years, with TAADs being more common and more fatal [3]. Without surgical intervention, the mortality rate of TAADs reaches nearly 50% within 48 h [4]. Due to the high mortality associated with TAAD, it is considered a surgical emergency, with the gold standard method of repair being open surgical repair [5]. Although type 2 diabetes (T2D) is traditionally considered a risk factor for cardiovascular disease [6], an unclear and complex association has been reported between T2D and aortic diseases. Previous data have shown that patients with diabetes develop smaller aortic aneurysms with lower growth rates compared to those without diabetes [7]. Further, some studies suggest that aortic aneurysm rupture-related mortality is lower in patients with diabetes [8]. Diabetes is also associated with a lower risk of acute aortic dissection [9]. A paradoxical inverse relationship between diabetes and the risk of aortic dissection has been reported [10]. While endothelial dysfunction is common in both conditions, it leads to acute tearing in aortic dissection but progressive wall degradation in aneurysm rupture. Hence, extrapolating a protective effect of diabetes from aortic aneurysms to rupture is not accurate. We aimed to bridge this knowledge gap by conducting an analysis that evaluated the impact of diabetes on type A aortic dissection. Our primary objective is to assess in-hospital outcomes of diabetes patients who undergo open surgery repair for TAAD. Additionally, we sought to assess short-term mortality and readmission in these patients and identify the most common underlying causes. Methods Data source Our data source was the Nationwide Readmissions Database (NRD), derived from the Healthcare Cost and Utilization Project’s (HCUP) Inpatient Databases. The database includes data on all-payer hospital inpatient stays from 30 states, accounting for 61.2% of the total US population and 59.2% of all hospitalizations in the US [11]. Furthermore, the database contains detailed data on diagnostic and procedural codes, patient demographics, in-hospital and post-discharge outcomes, and readmission rates using the International Classification of Diseases, Tenth Revision- ICD-10 [12]) [12]. This study was approved by Weill Cornell Medicine-Qatar’s institutional review board (record number 21 − 0002), and a consent form was not required. Data collection and outcome measures Patients with a primary diagnosis of type A aortic dissection between 2016 and 2019 who underwent open surgery were first identified (ICD-10 code: I71.0, I71.01, I71.02, and I71.03), which consisted of our initial study group (index group). Patients who successfully underwent surgery and were discharged alive were followed for up to 30 days (readmission group). We excluded patients with aortic aneurysms, TBAD, and those treated by endovascular repair. Patients under the age of 18 or those with missing data on age, mortality, or gender were also excluded. The primary outcome for the index group was in-hospital mortality, and the co-primary outcome for the readmitted patients was 30-day mortality. Secondary outcomes were atrial fibrillation (AF), acute renal failure, cardiogenic shock, stroke, and 30-day readmission. Further, we assessed the predictors and etiologies of readmission. Analysis methods and statistics Demographic information and comorbidities were analyzed using descriptive statistics. Categorical variables were described as numbers and percentages, and continuous variables as median (IQR) or mean (SD) as appropriate. To compare patients with and without diabetes, we utilized the Pearson chi-square test for categorical variables and the Student T-test for continuous variables. For in-hospital outcomes, odds ratios with their corresponding 95% confidence intervals were calculated using binary logistic regression, with adjustments made for age, sex, and baseline characteristics that were significantly different among both groups (obesity, smoking, hypertension, dyslipidemia, and chronic kidney disease). Kaplan-Meier (KM) curves were created to estimate and visually depict the probability of survival and readmission over a 30-day period, as they allow for straightforward comparison between groups based on diabetes status. KM curves are well-suited to handle time-to-event data, especially when censoring is present, and offer an intuitive way to compare survival distributions between the two groups. Hazard ratios for mortality and readmission were determined using the log-rank test. We applied a Cox proportional hazards model to adjust for confounding factors and obtain hazard ratios that account for differences in baseline covariates. This model assesses survival analysis, accommodates both the time-dependent nature of survival data, and adjusts for multiple covariates simultaneously. We also used the receiver operating characteristic (ROC) curve to evaluate the predictive accuracy of our statistical model for in-hospital and 30-day outcomes related to diabetes. A significance level of p < 0.05 was used for statistical significance. All statistical analyses were performed using SPSS Statistics version 26.0 (IBM Corp.). Results Study group Between 2016 and 2019, 7,537 cases of acute type A aortic dissection were admitted for open surgical repair (Fig. 1). After excluding patients with missing data, our study population consisted of 7,324 patients. Of those, 965 (13.2%) had a history of type 2 diabetes at the time of their initial hospitalization. Among patients with diabetes, 154 (15.9%) were readmitted within 30 days. In comparison, 952 of the 6359 (15.0%) non-diabetes patients were readmitted within the same time frame. Flowchart of Study. A total of 7,324 acute type A aortic dissection cases were admitted for open surgical repair and included in our analysis. Of these patients, 13.2% had type 2 diabetes during the initial hospitalization (index group), and 15.9% of them were readmitted within 30 days (readmission group) In-hospital outcomes Table 1 presents the demographic characteristics of patients with and without diabetes. Patients with diabetes were older than those without (64 [12] vs. 60 [13], respectively, p < 0.001). Notably, a higher percentage of diabetes patients were over the age of 75. Additionally, patients with diabetes exhibited a higher percentage of comorbidities, including obesity, hypertension, smoking, dyslipidemia, and chronic kidney disease (CKD), with all differences being statistically significant (p < 0.05 for all). We did not observe any significant differences in the surgical procedures performed. Table 1Comparison of baseline characteristics of acute type a aortic dissection patients who underwent open surgical repair, with and without diabetesDiabetesn = 965Non-diabetesn = 6359P-valueAge < 55219(22.7%)2103(33.1%)< 0.001 55–64258(26.7%)1607(25.3%) 65–74274(28.4%)1558(24.5%) > 75214(20.2%)1091(17.1%) Mean (SD)64.05(12.0)60.54(13.7)< 0.001Gender Male567(58.8%)4293(67.5%)< 0.001Income Low265(27.7%)1569(25.0%)0.200 Low-Mid234(24.5%)1528(24.3%) High-Mid244(25.5%)1637(26.1%) High212(22.2%)1548(24.6%)Los (Days) Median (IQR)11 (7–18)10 (6–17)0.109Comorbidities Obesity331(34.3%)1259(19.8%)< 0.001 Hypertension878(91.0%)5109(80.3%)< 0.001 Smoking194(20.1%)1110(17.4%)0.049 Dyslipidemia513(53.2%)1936(30.4%)< 0.001 Valvular heart disease337(34.9%)2357(37.1%)0.198 Chronic kidney disease257(26.6%)972(15.3%)< 0.001 Cerebrovascular disease64(6.6%)267(4.2%)< 0.001Surgical procedures Ascending Aortic Replacement452(7.1%)58(6.0%)0.212 Aortic Root Replacement1461(23.0%)204(21.1%)0.205 Aortic Valve Repair or Replacement1014(15.9%)146(15.1%)0.517 Hemiarch Replacement452(7.1%)58(6.0%)0.212 Total Aortic Arch Replacement125(2.0%)15(1.6%)0.385 Teflon Felt Reinforcement or Glue Repair1014(15.9%)146(15.1%)0.517 Frozen Elephant Trunk Procedure452(7.1%)58(6.0%)0.212Baseline characteristics of patients in the index group showed that those with diabetes were older and had a significantly higher prevalence of comorbidities. However, there was no difference in the type of surgical procedures Comparison of baseline characteristics of acute type a aortic dissection patients who underwent open surgical repair, with and without diabetes 219 (22.7%) 2103 (33.1%) 258 (26.7%) 1607 (25.3%) 274 (28.4%) 1558 (24.5%) 214 (20.2%) 1091 (17.1%) 64.05 (12.0) 60.54 (13.7) 567 (58.8%) 4293 (67.5%) 265 (27.7%) 1569 (25.0%) 234 (24.5%) 1528 (24.3%) 244 (25.5%) 1637 (26.1%) 212 (22.2%) 1548 (24.6%) 331 (34.3%) 1259 (19.8%) 878 (91.0%) 5109 (80.3%) 194 (20.1%) 1110 (17.4%) 513 (53.2%) 1936 (30.4%) 337 (34.9%) 2357 (37.1%) 257 (26.6%) 972 (15.3%) 64 (6.6%) 267 (4.2%) 452 (7.1%) 58 (6.0%) 1461 (23.0%) 204 (21.1%) 1014 (15.9%) 146 (15.1%) 452 (7.1%) 58 (6.0%) 125 (2.0%) 15 (1.6%) 1014 (15.9%) 146 (15.1%) 452 (7.1%) 58 (6.0%) Baseline characteristics of patients in the index group showed that those with diabetes were older and had a significantly higher prevalence of comorbidities. However, there was no difference in the type of surgical procedures One-hundred and forty-seven (15.2%) diabetes patients and 926 (14.6%) non-diabetes died, showing no significant difference in mortality between the two groups; the adjusted odds ratio (aOR) for in-hospital mortality was 1.02 (95% CI 0.84–1.24) (Table 2). The ROC curve showed that diabetes does not provide a predictive value for in-hospital mortality with an area under the curve (AUC) of 0.480 and a p-value of 0.604 (Fig. 2A). Table 3 outlines the predictors of in-hospital mortality. As expected, age is associated with a higher risk of in-hospital mortality in all patients. Gender did not appear to influence mortality, but cerebrovascular disease increased the risk by nearly 60% in diabetes patients and 30% in non-diabetes ones. Table 2Comparison of in-hospital outcomes in acute type a aortic dissection patients who underwent open surgical repair, with and without diabetesOutcomesNumber of eventsDiabetesNon-diabetesUnadjusted OR(95% CI)Adjusted OR(95% CI)P- value*Mortality147(15.2%)926(14.6%)1.05(0.87–1.27)1.02(0.84–1.24)0.847Atrial fibrillation398(41.2%)2291(36.0%)1.25(1.09–1.43)1.03(0.89–1.20)0.664Acute renal failure510(52.8%)2920(45.9%)1.32(1.15–1.51)1.20(1.04–1.39)0.015Stroke28(2.9%)223(3.5%)0.55(0.55–1.23)0.83(0.55–1.26)0.384Cardiogenic shock175(18.1%)1000(15.7%)1.22(1.05–1.32)1.18(0.98–1.42)0.08During the initial hospitalization, diabetes was not associated with a higher adjusted risk of in-hospital mortality, atrial fibrillation, stroke, or cardiogenic shock. However, the risk of acute renal failure was increased by 20%* p-value for adjusted risk Comparison of in-hospital outcomes in acute type a aortic dissection patients who underwent open surgical repair, with and without diabetes 147 (15.2%) 926 (14.6%) 1.05 (0.87–1.27) 1.02 (0.84–1.24) 398 (41.2%) 2291 (36.0%) 1.25 (1.09–1.43) 1.03 (0.89–1.20) 510 (52.8%) 2920 (45.9%) 1.32 (1.15–1.51) 1.20 (1.04–1.39) 28 (2.9%) 223 (3.5%) 0.55 (0.55–1.23) 0.83 (0.55–1.26) 175 (18.1%) 1000 (15.7%) 1.22 (1.05–1.32) 1.18 (0.98–1.42) During the initial hospitalization, diabetes was not associated with a higher adjusted risk of in-hospital mortality, atrial fibrillation, stroke, or cardiogenic shock. However, the risk of acute renal failure was increased by 20% * p-value for adjusted risk Receiver operating characteristic (ROC) curve to evaluate the predictive accuracy of diabetes for in-hospital (A) and 30-day (B) mortality. Both models have no discrimination ability with an area under the curve (AUC) < 5 Table 3Predictors of in-hospital mortality, 30-day mortality, and 30-day readmissionPredictorsIn-hospital mortality30-day mortality Diabetes Non-diabetes Diabetes Non-diabetes OR (95% CI)P-valueOR (95% CI)P-valueOR (95% CI)P-valueOR (95% CI)P-valueAge1.98(1.16–3.38)0.0131.45(1.18–1.77) < 0.0011.71(1.44–1.89)0.0081.12(1.02–1.14)0.006SexMale (ref)Female0.916(0.64–1.31)0.6331.13(0.98–1.31)0.0930.48(0.05–4.58)0.5201.78(0.87–3.64)0.115IncomeLow (ref)Low-Mid1.31(0.80–2.16)0.2791.14(0.94–1.38)0.1771.09(0.07–7.58)0.9600.72(0.36–1.44)0.157High-Mid1.41(0.87–2.28)0.1680.94(0.77–1.14)0.5170.77(0.22–2.75)0.6890.61(0.30–1.23)0.201High1.04(0.61–1.77)0.8810.78(0.64–0.96)0.0194.71(0.41–33.67)0.2110.72(0.36–1.42)0.101ObesityNo (ref)Yes0.85(0.58–1.24)0.4040.89(0.74–1.06)0.2010.85(0.09–8.20)0.8900.92(0.32–2.63)0.874HTNNo (ref)Yes0.66(0.38–1.15)0.1400.84(0.52–1.13)0.3420.22(0.02–2.17)0.1950.60(0.26–1.34)0.211SmokingNo (ref)Yes0.92(0.59–1.44)0.7290.85(0.71–1.03)0.1051.11(0.12–10.68)0.9270.37(0.13–1.07)0.066DyslipidemiaNo (ref)Yes0.85(0.60–1.20)0.3560.94(0.80–1.09)0.3995.85(0.18–18.68)0.3230.99(0.46–2.12)0.979Valvular heart diseaseNo (ref)Yes0.86(0.59–1.25)0.4160.81(0.70–0.94)0.0051.27(0.13–12.34)0.8340.78(0.33–1.81)0.557Chronic kidney diseaseNo (ref)Yes1.26(0.86–1.85)0.2361.23(1.02–1.48)0.0275.58(2.58–6.62)0.0373.15(2.53–6.49)0.002Cerebrovascular diseaseNo (ref)Yes1.62(1.18–1.86)0.0021.32(1.11–1.84)0.0272.27(2.14–2.98)0.0072.12(1.98–2.41)0.004Age is associated with a higher risk of all outcomes. CVD increases the risk of in-hospital and 30-day mortality. Chronic kidney disease was associated with a nearly fivefold increase in mortality risk among patients with diabetes and a threefold increase among those without diabetes Predictors of in-hospital mortality, 30-day mortality, and 30-day readmission 1.98 (1.16–3.38) 1.45 (1.18–1.77) 1.71 (1.44–1.89) 1.12 (1.02–1.14) 0.916 (0.64–1.31) 1.13 (0.98–1.31) 0.48 (0.05–4.58) 1.78 (0.87–3.64) 1.31 (0.80–2.16) 1.14 (0.94–1.38) 1.09 (0.07–7.58) 0.72 (0.36–1.44) 1.41 (0.87–2.28) 0.94 (0.77–1.14) 0.77 (0.22–2.75) 0.61 (0.30–1.23) 1.04 (0.61–1.77) 0.78 (0.64–0.96) 4.71 (0.41–33.67) 0.72 (0.36–1.42) 0.85 (0.58–1.24) 0.89 (0.74–1.06) 0.85 (0.09–8.20) 0.92 (0.32–2.63) 0.66 (0.38–1.15) 0.84 (0.52–1.13) 0.22 (0.02–2.17) 0.60 (0.26–1.34) 0.92 (0.59–1.44) 0.85 (0.71–1.03) 1.11 (0.12–10.68) 0.37 (0.13–1.07) 0.85 (0.60–1.20) 0.94 (0.80–1.09) 5.85 (0.18–18.68) 0.99 (0.46–2.12) 0.86 (0.59–1.25) 0.81 (0.70–0.94) 1.27 (0.13–12.34) 0.78 (0.33–1.81) 1.26 (0.86–1.85) 1.23 (1.02–1.48) 5.58 (2.58–6.62) 3.15 (2.53–6.49) 1.62 (1.18–1.86) 1.32 (1.11–1.84) 2.27 (2.14–2.98) 2.12 (1.98–2.41) Age is associated with a higher risk of all outcomes. CVD increases the risk of in-hospital and 30-day mortality. Chronic kidney disease was associated with a nearly fivefold increase in mortality risk among patients with diabetes and a threefold increase among those without diabetes Further, diabetes did not significantly impact the adjusted risk of AF (aOR = 1.03 [0.89–1.20]), cardiogenic shock (aOR = 1.18 [0.98–1.42]), or stroke (aOR = 0.83 [0.55–1.26]), but it increased the risk of acute renal failure (aOR = 1.20 [1.04–1.39]). 30-day outcome A total of 154 (15.9%) patients with diabetes and 952 (15%) non-diabetes patients were readmitted within 30 days. Baseline characteristics of readmitted patients showed that those with diabetes were, on average, five years older (65 [12] vs. 60 [13], diabetes vs. non-diabetes, p < 0.001) (Table 4). The majority of patients in both groups were male, but the proportion of males was higher in non-diabetes patients compared to patients with diabetes. Additionally, patients with diabetes had a significantly higher prevalence of comorbidities, obesity, hypertension, dyslipidemia, VHD, CKD, and cerebrovascular disease. Table 4Comparison of baseline characteristics of readmitted patients, with and without diabetesDiabetes n = 154 (15.9%)*Non-diabetes n = 952 (15%)*P-valueAge < 5531(20.1%)325(34.1%) < 0.001 55–6431(20.1%)235(24.7%) 65–7450(32.5%)226(23.7%) 75–8439(25.3%)147(15.4%) > 843(1.9%)19(2.0%) Mean (SD)65.89(12.30)60.59(13.55) < 0.001Gender Male87(56.5%)623(65.4%)0.032Income Low52(33.8%)235(24.9%)0.052 Low-Mid40(26.0%)222(23.5%) High-Mid30(19.5%)244(25.9%) High32(20.8%)242(25.7%)Comorbidities Obesity42(27.3%)144(15.1%) < 0.001 Hypertension139(90.3%)786(82.6%)0.017 Smoking38(24.7%)257(27.0%)0.546 Dyslipidemia79(51.3%)327(34.3%) < 0.001 Valvular heart disease33(21.4%)266(27.9%)0.020 Chronic kidney disease58(37.7%)198(20.8%) < 0.001 Cerebrovascular disease12(7.7%)41(4.3%) < 0.001Baseline characteristics of readmitted patients indicate that those with diabetes were, on average, five years older. Obesity, hypertension, dyslipidemia, valvular heart disease, chronic kidney disease, and cerebrovascular disease were more common among patients with diabetes. * N (%) of readmitted patients among the ones discharges Comparison of baseline characteristics of readmitted patients, with and without diabetes 31 (20.1%) 325 (34.1%) 31 (20.1%) 235 (24.7%) 50 (32.5%) 226 (23.7%) 39 (25.3%) 147 (15.4%) 3 (1.9%) 19 (2.0%) 65.89 (12.30) 60.59 (13.55) 87 (56.5%) 623 (65.4%) 52 (33.8%) 235 (24.9%) 40 (26.0%) 222 (23.5%) 30 (19.5%) 244 (25.9%) 32 (20.8%) 242 (25.7%) 42 (27.3%) 144 (15.1%) 139 (90.3%) 786 (82.6%) 38 (24.7%) 257 (27.0%) 79 (51.3%) 327 (34.3%) 33 (21.4%) 266 (27.9%) 58 (37.7%) 198 (20.8%) 12 (7.7%) 41 (4.3%) Baseline characteristics of readmitted patients indicate that those with diabetes were, on average, five years older. Obesity, hypertension, dyslipidemia, valvular heart disease, chronic kidney disease, and cerebrovascular disease were more common among patients with diabetes. * N (%) of readmitted patients among the ones discharges Figure 3 illustrates the hazards of 30-day outcomes across all patients. We didn’t observe any statistically significant difference between the survival curves of diabetes and non-diabetes groups for 30-day mortality (HR = 1.11 (0.57–2.13). Interestingly, the 30-day readmission risk was lower in patients with diabetes (HR = 0.84 (0.70–0.99). These results remained unchanged after adjusting for variables that differed between the two groups (aHR = 0.81 (0.41–1. 60), 0.81 [0.68–0.97]; mortality and readmission, respectively). The ROC curve showed that diabetes does not provide predictive value for 30-day mortality, performing even worse than in-hospital mortality, with an AUC of 0.258 and a p-value of 0.644 (Fig. 2B). Hazard Ratio for 30-day outcomes in diabetes and non-diabetes patients. The Kaplan-Meier curve shows no significant difference in the risk of 30-day mortality (HR = 1.11 [0.57–2.13]) between both groups (A). However, the risk of readmission was lower in patients with diabetes (HR = 0.84 [0.70–0.99]) We further assessed the predictors of in-hospital mortality. As expected, age was a significant risk factor in all patients. CKD was strongly associated with increased risk, particularly among diabetes patients, where it increased the risk nearly fivefold. In non-diabetes patients, CKD was also associated with a higher mortality risk, but to a lesser extent, increasing the risk almost threefold (OR = 3.15 [1.53–6.49]). Cerebrovascular disease was associated with a nearly twofold increase in the risk of in-hospital mortality. We further examined the causes of readmission and stratified them by the presence of diabetes. The top five categories of readmission etiology were consistent between patients with diabetes (Fig. 4A) and non-diabetes patients (Fig. 4B), with cardiovascular causes being the most common in both groups. However, cardiovascular-related readmissions were nearly three times more prevalent in diabetes patients compared to non-diabetes patients. Additionally, patients with diabetes had a significantly higher percentage of readmissions due to respiratory-related conditions compared to non-diabetes patients (20% vs. 7%, respectively, p = 0.032). Etiologies of Readmission in patients without diabetes (A) and with diabetes (B). Cardiovascular Etiologies of readmission in patients without diabetes (C) and with diabetes(D). Cardiovascular disease was the leading cause of readmission in all patients. In patients with diabetes, hypertensive heart disease with heart failure was the most common etiology We then examined the cardiovascular causes of readmission in more detail. Hypertensive heart disease with heart failure was the leading cardiovascular cause of readmission in diabetes patients, and its prevalence was almost four times higher in the diabetes group compared to non-diabetes (Fig. 3C, D). Dissection of the thoracic aorta was the second most common cardiovascular cause of readmission in diabetes patients. However, its prevalence was not statistically significant between the two groups (13% vs. 9.1%, diabetes vs. non-diabetes, p = 0.074). Finally, pulmonary embolism was the third most frequent cardiovascular etiology among diabetes patients, with its prevalence being twice as high in diabetes patients compared to non-diabetes ones (11% vs. 5%, p = 0.046). Discussion In this nationwide cohort of patients who underwent open surgical repair for type A aortic dissection, we report that diabetes did not impact in-hospital mortality. Furthermore, our study documents that diabetes does not increase 30-day readmission or mortality. To our knowledge, our study is the first to specifically investigate and report the outcomes of acute type A aortic dissection patients who underwent open surgical repair in the context of diabetes. Recent studies in Western countries have also found a rising number of acute aortic dissection cases [13]. Although several papers have demonstrated a relative risk reduction for hospitalization due to aortic dissection in patients with diabetes [14–16], there is limited research on outcomes post-repair. An observational, matched cohort study in Sweden showed no significant difference in survival between diabetes and non-diabetes patients after three months, one year, two years, or three years [15]. However, in their study design, there was no differentiation between type A or type B aortic dissections, and patients were not stratified by the repair technique used. Diabetes is protective against aortic dissection formation and progression [17]. It might be possible that diabetes induces changes to the vascular extracellular matrix of the aortic wall and increases matrix metalloproteinase secretion [18, 19]. Another plausible mechanism would be related to inflammation, as chronic low-grade inflammation observed in the presence of diabetes may promote adaptive responses in the aorta, potentially enhancing its resistance to dissection [20]. Finally, insulin and related growth factors can have protective roles in vascular health [21], potentially influencing the behavior of vascular smooth muscle cells and promoting a more stable aortic structure. The lack of significant impact of diabetes on mortality is not fully understood. Patients with diabetes and severe comorbidities may be less likely to undergo surgery for TAAD or may die before reaching the operating theater, introducing a selection bias where only healthier diabetes patients are operated on. A recent analysis of the International Registry of Acute Aortic Dissection showed that patients who were managed medically were more likely to have diabetes and less likely to survive compared to patients who underwent surgery [22]. Further, advances in perioperative glycemic control and tailored management of diabetes during cardiac surgery may neutralize its impact on postoperative mortality. Enhanced protocols for managing diabetes in surgical settings, including strict glucose control and insulin therapy, have been shown to improve outcomes in cardiovascular surgeries [23]​. It might also be possible that the hemodynamic stress of the dissection overshadows the deleterious effects of diabetes, resulting in comparable mortality risks between diabetes and non-diabetes patients. However, diabetes patients were more likely to be readmitted within 30 days, although mortality is not statistically different. This could be attributed to slower recovery in patients with diabetes and the burden of pre-existing comorbidities at discharge, as reflected in our study by the higher number of readmissions due to conditions such as infections and heart failure. It is well known that people with diabetes are a high-risk surgical population with longer hospital stays, higher healthcare resource utilization, and more significant perioperative morbidity [24, 25]. In our analysis, patients with diabetes had significantly higher rates of comorbidities except for VHD. Although this may seem unexpected, as diabetes is a risk factor for calcified aortic stenosis [26], two recent studies have found that diabetes is associated with a lower risk of valvular regurgitation, particularly severe aortic and mitral regurgitation [27, 28]. Diabetes often leads to changes in blood flow and pressure, which may reduce stress on heart valves and contribute to lower rates of valve degeneration [29]. Further, diabetes can alter collagen production and deposition in the heart, which might influence the integrity and stiffness of heart valves [30]. However, our analysis showed that CKD was the most significant predictor of 30-day mortality in all patients, which has been reported in several cardiovascular studies, such as in acute coronary syndrome [31], heart failure [32], and vascular surgery [33]. Patients with CKD exhibit hypercoagulability, which contributes to an elevated risk of thrombosis and increased cardiovascular mortality [34]. Additionally, the occurrence of both thrombotic and hemorrhagic complications in CKD patients indicates potential abnormalities in platelet function [35]. We acknowledge the presence of limitations in our study. The National Readmission Database (NRD) presents a comprehensive dataset, but its primary purpose is for reimbursement [12], which may introduce inaccuracies regarding the input of correct diagnostic or procedural codes. Moreover, patient race plays a significant role in understanding healthcare disparities and is not included in the NRD data elements. This omission could influence the results, as one study conducted in the state of Maryland reported that non-white race was an independent risk factor for hospitalizations due to aortic dissections [36]. However, another study found no association between race and in-hospital mortality for patients after surgical repair for TAAD [37]. In any case, understanding the impact of race on outcomes is essential to address healthcare inequalities effectively. The NRD database uses binary flags to identify if a particular comorbidity is present in a patient; thus, the accuracy and severity of those conditions can only be verified if the diagnostic criteria are included in the database. Further, the extent of the lesion, perioperative complications, and other indicators that could influence the clinical course and outcome were missing in the database. The NRD also does not inform patients of cardioprotective medications they are taking, as many of these medications have been shown to improve postoperative outcomes [38]. For patients with diabetes, another gap in the NRD is the absence of diabetes-related characteristics such as glycemic control, mean glycated hemoglobin, duration of diabetes, and hypoglycemic medications used. Finally, the NRD did not provide exact blood pressures, so we could not stratify patients based on the stage of hypertension they presented with upon admission. Conclusion This study’s findings suggest that, while diabetes is an important risk factor in many conditions, it does not appear to significantly influence early mortality in patients undergoing surgical repair for Type A aortic dissection. Clinicians should continue to manage diabetes and other comorbidities while prioritizing factors such as timely surgical intervention, hemodynamic management, and post-operative care. Future studies assessing long-term clinical outcomes, including the impact of diabetes parameters such as glycemic control, duration, and treatment, are essential for refining risk stratification and optimizing management strategies for this high-risk patient group. Future research should also examine how specific surgical techniques or postoperative management approaches might interact with diabetes physiology and affect patient outcomes. Nevertheless, patients with diabetes must receive specialized care to address the unique challenges posed by the coexistence of these two conditions. As the prevalence of diabetes rises, understanding its role in fatal conditions such as Type A aortic dissection remains crucial. Acknowledgements The authors thank Ziyad Mahfoud, PhD, for his guidance on the statistical analyses and Karim Saade, BS, for his assistance in creating the graphical abstract, both at Weill Cornell Medicine-Qatar. Abbreviations Coronary artery disease Chronic kidney disease National readmission database Type A aortic dissection Type B aortic dissection Type 2 diabetes Valvular heart disease Author contributions CAK was involved in the study’s design. HC and SD conducted the data analysis. HC, AJ, JAS, and CAK interpreted the data. HC completed the first draft of the manuscript. All authors were involved in the critical revision of the manuscript. All authors read and approved the final manuscript. CAK is the guarantor of this work and, as such, has full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Funding None. Data availability Data are available upon reasonable request from the lead author. Declarations Ethics approval and consent to participate This study was approved by WCM-Q’s institutional review board (record number 21−0002). Consent for publication Not applicable. Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Data Availability Statement Data are available upon reasonable request from the lead author.