Racial and Sex Differences in Surgical Aortic Valve Replacement Short-term Post-Operative Outcomes at a High-Volume Cardiac Surgery Center Open Access

Teichmann, Maiko Sasaki (Summer 2019)

Permanent URL: https://etd.library.emory.edu/concern/etds/rj4305762?locale=en


Background: Both African American race and female gender are under-represented in studies investigating on their effects in clinical outcomes after SAVR procedure. Disparities in care in female and African American cohorts are reported in multiple studies. Consensus on racial and gender effects independently on postoperative outcomes have not been reached, and the combined effects of gender and race have not been reported after SAVR. 

Objective: This study aimed to investigate sex and race differences in 30-day mortality and postoperative complications among the patients who received surgical aortic valve replacement (SAVR) procedure. 

Methods:  Retrospective analyses were conducted on all patient data undergoing SAVR obtained from Adult Cardiac Surgery Database Data Collection from Emory University Hospital, St. Joseph’s Hospital, and Emory University Hospital Midtown. 3232 patients who underwent surgical procedures between 1/26/2005 to 3/29/2019 (STS data versions 2.52, 2.61, 2.73, 2.81, and 2.9). Associations between both gender and race and the clinical outcomes were investigated using risk-adjusted logistic and linear regression models.

Results: A total of 3232 cases met the inclusion criteria and were included in this study. Black females constituted 6.56% of the sample, white females 32.39%, black males 8.32%, and white males 52.72%. Black patients were on average younger at the time of operation (black female: 63.2 ± 13.7 yo, black male: 59.5 ± 14.41 yo, white female: 69.9 ± 12.7 yo, white male: 66.1 ± 13.7 yo, p-value < 0.0001), and had higher frequency of requiring urgent procedures (50 %, 47.2 %, 34.9 %, 32.5 % for black female and male, white female and male respectively, p-value <0.0001). 76 (2.3%) cases were reported for 30-day mortality, a main outcome of this study.  Black males had the highest odds of 30-day mortality when compared to white males (aOR 1.677, 95 % CI: 0.793 – 3.550 against white males). Black male also had the most hospital utilization including lengths of post-operative stay in the hospital before discharge (aGMR 1.124, 95 % CI: 1.055 – 1.198).

Conclusions: These results revealed more complex relationship of race/gender and the clinical outcomes after SAVR, and requires careful interpretation of the. After risk adjustment, no significant differences in 30-day mortality were observed for the different sex and race categories. That blacks had a longer length of stay than whites at this high-volume cardiac center could explain the observed similar risk-adjusted mortality rates.  

Table of Contents

1.        Introduction. 1

1.1.        Cardiovascular valvular disease epidemiology. 1

1.2.        Aortic Valve Replacement (AVR) therapy. 2

1.2.1.    TAVR.. 2

1.2.2.    SAVR.. 3

1.2.3.    Public health implications in access to AVR.. 3

1.3.        Racial/gender disparities in SAVR outcomes: Public health perspectives 5

1.3.1.    Gender disparities. 5

1.3.2.    Racial disparities. 6

1.4.        Study aims. 6

2.        Methods. 8

2.1.        Study Design. 8

2.2.        Study Variable Selection. 8

2.3.        Data Cleaning. 9

2.4.        Statistical Analyses. 11

3.        Results. 12

3.1.        Study Sample. 12

3.1.1.    Characteristics which are worse in both male and female AA cohorts 12

3.1.2.    Characteristics which are markedly worse in female AA cohorts. 14

3.1.3.    Characteristics which are worse in female cohorts, both black and white 14

3.1.4.    Characteristics which are worse in white cohorts. 14

3.1.5.    Characteristics which are worse in male cohorts. 14

3.2.        Crude Analysis Results. 15

3.3.        Adjusted Analysis Results. 16

4.        Conclusions and Discussion. 17

4.1.        Discussion. 17

4.2.        Descriptive analyses. 17

4.2.1.    Crude analyses. 18

4.2.2.    Adjusted outcomes. 19

4.2.3.    Investigation on loss of significance. 19

4.2.4.    Limitations and strengths. 21

4.3.        Discussion on the findings and their important implications in public health 22

4.4.        Conclusions and further direction. 26

5.        Figures. 27

Figure 1. Study Flow Diagram.. 27

Table 1. Patient Characteristics. 28

Table 2. Intraoperative Details by Race and Gender 29

Table 3a. Descriptive Postoperative Outcomes by Race and Gender 30

Table 3b. Crude Postoperative Outcomes. 31

Table 3c. Adjusted Postoperative Outcomes. 32

Figure 2. Forest Plots Summarizing the Adjusted Outcomes. 33

Table 4. Description of post-procedural length of stay in hospital and the risk score. 34

Table 5. Comparison pairs of cORs for non-home discharge location. 35

6.        References. 36

7.        Appendix. 40

Supplemental Table 1. Determination of variable and version selection. 40

Supplemental Figure 2. SAS Codes. 41

About this Master's Thesis

Rights statement
  • Permission granted by the author to include this thesis or dissertation in this repository. All rights reserved by the author. Please contact the author for information regarding the reproduction and use of this thesis or dissertation.
Subfield / Discipline
  • English
Research Field
Committee Chair / Thesis Advisor
Committee Members
Last modified

Primary PDF

Supplemental Files