Abstract
Objective:
To compare sex and racial differences in 180-day infection rates after coronary artery bypass grafting (CABG) and aortic valve replacement (AVR).
Methods:
A Statewide Society of Thoracic Surgeons Adult Cardiac Surgery Database was linked to Medicare claims data to identify 8,887 beneficiaries undergoing CABG and AVR (surgical or transcatheter) between 2017 and 2021. The primary outcome was the incidence of 180-day infection. Secondary outcomes included ten infection subtypes. Multivariable logistic regression was used to evaluate the relationship between sex and race (Black versus non-Black) and infections. Two secondary analyses were conducted: (1) robustness of the primary analysis after excluding urinary tract infections “UTIs” given established sex-related differences and (2) testing a sex*race interaction.
Results:
The mean (SD) age of the cohort was 74.5 (8.9) years, with 36.9% female and 4.2% Black. The infection rate was 19.6%, although varied by patient sex (female versus male: 23.7% versus 17.1%) and race (Black versus non-Black: 28.0% versus 19.2%), both p<0.0001. Differences in infection rates for females were driven by UTI, while pneumonia for Black patients. Risk-adjusted odds of infection were 1.6-fold significantly higher among female while non-significant for among Black patients. A sex*race interaction was present, with non-Black females versus non-Black males having a 1.63 higher odds of infection.
Conclusion:
This multi-center study identified a 1.6-fold higher odds of infection among female patients. Non-Black female versus male patients had a 63% higher odds of infection. Transdisciplinary collaborative learning interventions should be considered to address these known disparities in infection rates.
Keywords: cardiac surgery, infections, race, sex
Graphical Abstract
Introduction
Postoperative infections are a source of major morbidity and mortality after cardiac surgery.1–3 Reducing infections has been the focus of many national and regional policies and quality improvement initiatives4–8, as many infections are potentially preventable. Considerable focus has surrounded postoperative infections occurring within the index admission because of their associated financial sequelae.9,10 Indeed, the median incremental expenditures increased $33,318 among patients developing deep sternal wound infections ($57,744 versus $24,426) and $38,295 ($62,352 versus $24,057) for pneumonia among patients undergoing isolated coronary artery bypass grafting (CABG) compared to those without infection.10
Efforts to address infections after cardiac surgery have predominantly evaluated pre- and intraoperative risk factors, with discrepant findings associated with the role of patient sex and race. Strobel et al. evaluated preoperative risk factors associated with pneumonia among a large cohort study of 16,084 patients undergoing isolated CABG surgery across 33 hospitals.6 Pneumonia occurred among 3.3% of patients, with non-White race (AOR 1.66, p<0.001) being one of 17 significant risk factors. Abukhodair et al. evaluated risk factors for infections (surgical site infections of the chest and leg, pneumonia, bloodstream infection, urinary tract infection, Clostridioides difficile, endocarditis) within a six-week postoperative period among 364 patients undergoing cardiac surgery11 and identified an incidence of infection of 32.7%. The adjusted odds of infection were non-significantly lower among males versus females (OR 0.96, CI 95%: 0.84–1.10). In a recent study of 1,042,506 patients undergoing CABG, Enumah et al. evaluated the associated impact of racial disparities following CABG.12 In particular, Black versus White patients had higher adjusted odds of surgical site infections (OR 1.28, CI [1.19–1.38]) and sepsis (OR 1.60, CI [1.48–1.73]). Although the aforementioned studies highlight possible interactions of sex and race on infection rates, they are limited in generalizability across multiple cardiac surgeries as the majority were limited to CABG surgery and a short follow-up period after discharge.
To advance health equity across traditionally minoritized populations, further evaluation is warranted to identify the independent effect of sex and race on infections and associated adverse sequelae across CABG and AVR and over a longer follow-up period. In that context, this statewide study of Medicare beneficiaries evaluated the associated impact of patient sex and race on infection rates after CABG and AVR.
Methods
The University of Michigan Medical School’s Institutional Review Board approved this study on August 24, 2022 (HUM00214501), with a waiver of informed consent granted for the observational analyses associated with this manuscript.
Data Source and Study Population
This study leveraged The Society of Thoracic Surgeons Adult Cardiac Surgical Database (STS-ACSD) clinical registry data from 33 non-federal hospitals participating in the Michigan Society of Thoracic and Cardiovascular Surgeons Quality Collaborative (MSTCVS-QC) linked to Medicare fee-for-service claims data from the Michigan Value Collaborative (MVC). Medicare beneficiaries undergoing CABG, surgical aortic valve replacement (SAVR), transcatheter aortic valve replacement (TAVR), or combined SAVR and CABG from July 1, 2017, to December 31, 2021, were included in the study. Beneficiaries must have enrolled in the traditional Medicare program (i.e., with both Part A and Part B coverages) during and six months after CABG and AVR. In accordance with the Center for Medicare and Medicaid (CMS), the DUA #26142 between the University of Michigan and the CMS requires suppression of data with less than 11 observations. Cells with less than 11 observations are noted with a “*”. While the University of Michigan’s existing agreements restrict the distribution of raw study-related data files, requests for summary statistics will be reviewed and may be approved by the study team.
Several data sources were used for the present study. Beneficiary characteristics, comorbidities, and intra- and early postoperative data were obtained through the STS-ACSD, except for Medicaid and Medicare dual eligibility, which was based on the Medicare Beneficiary Summary Files. A combination of data variables from the STS-ACSD and Medicare claims was used to identify the date and type of infections, discharge disposition, and vital status within 180 days after CABG and AVR (including surgical and transcatheter). Beneficiary sex and race were the primary exposures of interest. Beneficiary race, identified through the STS-ACSD, was classified as Black versus non-Black (White, Asian, and other).
Clinical Outcomes
The primary outcome was the incidence of infection at 180 days. Secondary outcomes included the cumulative incidences of respiratory failure (at 90- and 180-day), stroke (at 90- and 180-day), mortality (at discharge, 30-, 90-, and 180-day), and component infection at 180-day. Infections included urinary tract infection (UTI), pneumonia, sepsis, gastrointestinal infection, bloodstream infection, endocarditis, sternal wound infection, cellulitis, and cannulation site or conduit harvest site infection. Diagnosis codes from the International Statistical Classification of Diseases and Related Health Problems, Tenth Revision (ICD-10), were used to identify infections, respiratory failure, and stroke from Medicare inpatient and skilled nursing home claims13. Death was identified from the Medicare Beneficiary Summary Files.
Statistical Analysis
Baseline characteristics and clinical outcomes were compared separately by sex (male versus female) and race (Black versus non-Black). Continuous variables were presented as mean (SD) and compared with a two-sample t-test. Categorical variables were presented as frequency (percentage) and compared using the Chi-Squared test of association. The risk-adjusted incidence of 180-day infection was calculated using a multivariable logistic regression model, adjusting for baseline characteristics, operation type, and weighting for survival time. Effect modification between infections and mortality by sex and race was tested with an interaction term. Statistical significance was defined as a p-value less than 0.05.
Two sensitivity analyses were performed. First, the robustness of findings between beneficiary sex and race on the incidence of infections was evaluated. Second, given established higher incidence of UTI among females14, UTI was excluded from list of component infections in sensitivity analysis. Statistical analyses were performed with SAS version 9.4 (SAS Institute, Cary, NC), with figures generated using Prism.
Results
A total of 8,887 cardiac operations were identified, with a mean (SD) age of 74.5 (8.9) years, with 36.9% female and 4.2% Black (Table 1). Most operations were performed electively (69.1%). Female versus male beneficiaries were older (76.1 versus 73.6 years, p<.0001), more likely to have dual eligible Medicaid and Medicare (14.2% versus 9.1%, p<.0001), and have a higher classification of New York Heart Association Class (Class III: 32.1% versus 22.3%; Class IV: 6.7% versus 5.1%, p<.0001). Black versus non-Black beneficiaries were younger (69.7 versus 74.7 years, p<.0001), more likely current smokers (20.3% versus 11.4%, p<.0001), have diabetes mellitus (56.3% versus 44.1%, p<.0001), peripheral arterial disease (34.9% versus 23.1%, p<.0001), renal failure (14.7% versus 3.0%, p<.0001), although were less likely to have elective procedures (57.6% versus 69.6%, p<.0001). More extensive comparisons of baseline demographics, comorbidities, and outcomes across beneficiary sex and race are displayed in Supplemental Table 1 and Supplemental Table 2, respectively.
Table 1.
Demographics by Patient Sex and Race
| Variable | Overall | Male | Female | P value | Non-Black | Black | P value | Non-Black Male | Non-Black Female | Black Male | Black Female | P value | |||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Total | 8,887 | 5,611 (63.1) | 3,276 (36.9) | 8,512 (95.8) | 375 (4.2) | 5,411 (60.9) | 3,101 (34.9) | 200 (2.3) | 175 (2.0) | ||||||
| Year | 0.1016 | <0.0001 | 0.1681 | ||||||||||||
| 2017 | 1,096 (12.3) | 727 (13.0) | 369 (11.3) | 1,055 (12.4) | 41 (10.9) | 710 (13.1) | 345 (11.1) | 17 (8.5) | 24 (13.7) | ||||||
| 2018 | 2,278 (25.6) | 1,417 (25.3) | 861 (26.3) | 2,166 (25.4) | 112 (29.9) | 1,357 (25.1) | 809 (26.1) | 60 (30.0) | 52 (29.7) | ||||||
| 2019 | 2,116 (23.8) | 1,346 (24.0) | 770 (23.5) | 1,991 (23.4) | 125 (33.3) | 1,274 (23.5) | 717 (23.1) | 72 (36.0) | 53 (30.3) | ||||||
| 2020 | 1,777 (20.0) | 1,122 (20.0) | 655 (20.0) | 1,698 (19.9) | 79 (21.1) | 1,079 (19.9) | 619 (20.0) | * | * | ||||||
| 2021 | 1,620 (18.2) | 999 (17.8) | 621 (19.0) | 1,602 (18.8) | 18 (4.8) | 991 (18.3) | 611 (19.7) | * | * | ||||||
| Age at surgery | |||||||||||||||
| Mean (STDEV) | 74.5 (8.9) | 73.6 (8.7) | 76.1 (9.2) | <0.0001 | 74.7 (8.8) | 69.7 (10.6) | <0.0001 | 73.8 (8.6) | 76.3 (9.0) | 68.2 (10.4) | 71.4 (10.5) | <0.0001 | |||
| <0.0001 | <0.0001 | <0.0001 | |||||||||||||
| <65 | 768 (8.6) | 537 (9.6) | 231 (7.1) | 671 (7.9) | 97 (25.9) | 477 (8.8) | 194 (6.3) | 60 (30.0) | 37 (21.1) | ||||||
| 65–69 | 1,809 (20.4) | 1,259 (22.4) | 550 (16.8) | 1,716 (20.2) | 93 (24.8) | 1,204 (22.3) | 512 (16.5) | 55 (27.5) | 38 (21.7) | ||||||
| 70–74 | 1,924 (21.6) | 1,309 (23.3) | 615 (18.8) | 1,863 (21.9) | 61 (16.3) | 1,276 (23.6) | 587 (18.9) | 33 (16.5) | 28 (16.0) | ||||||
| 75–79 | 1,769 (19.9) | 1,121 (20.0) | 648 (19.8) | 1,717 (20.2) | 52 (13.9) | 1,100 (20.3) | 617 (19.9) | 21 (10.5) | 31 (17.7) | ||||||
| 80–84 | 1,400 (15.8) | 791 (14.1) | 609 (18.6) | 1,355 (15.9) | 45 (12.0) | 771 (14.2) | 584 (18.8) | 20 (10.0) | 25 (14.3) | ||||||
| >=85 | 1,217 (13.7) | 594 (10.6) | 623 (19.0) | 1,190 (14.0) | 27 (7.2) | 583 (10.8) | 607 (19.6) | 11 (5.5) | 16 (9.1) | ||||||
| Race | <0.0001 | ||||||||||||||
| Black | 375 (4.2) | 200 (3.6) | 175 (5.3) | N/A | N/A | N/A | N/A | N/A | N/A | ||||||
| Non-Black | 8,512 (95.8) | 5,411 (96.4) | 3,101 (94.7) | N/A | N/A | N/A | N/A | N/A | N/A | ||||||
| Hispanic | 0.6313 | 0.17 | 0.3578 | ||||||||||||
| Yes | 118 (1.3) | 77 (1.4) | 41 (1.3) | * | * | 77 (1.4) | * | * | * | ||||||
| No | 8,769 (98.7) | 5,534 (98.6) | 3,235 (98.7) | * | * | 5,334 (98.6) | * | * | * | ||||||
| Sex | <0.0001 | ||||||||||||||
| Female | 3,276 (36.9) | N/A | N/A | 3,101 (36.4) | 175 (46.7) | N/A | N/A | N/A | N/A | ||||||
| Male | 5,611 (63.1) | N/A | N/A | 5,411 (63.6) | 200 (53.3) | N/A | N/A | N/A | N/A | ||||||
| Dual eligibility | <0.0001 | <0.0001 | <0.0001 | ||||||||||||
| Yes | 977 (11.0) | 511 (9.1) | 466 (14.2) | 850 (10.0) | 127 (33.9) | 454 (8.4) | 396 (12.8) | 57 (28.5) | 70 (40.0) | ||||||
| Body Mass Index | |||||||||||||||
| Mean (STDEV) | 30.0 (6.3) | 29.8 (5.7) | 30.5 (7.1) | <0.0001 | 30.1 (6.3) | 29.6 (6.3) | 0.188 | 29.8 (5.7) | 30.5 (7.1) | 29.0 (6.0) | 30.3 (6.5) | <0.0001 | |||
| <0.0001 | 0.13 | 0.009 | |||||||||||||
| 18.5–24.9 (normal) | 1,698 (19.1) | 991 (17.7) | 707 (21.6) | 1,614 (19.0) | 84 (22.4) | 943 (17.4) | 671 (21.6) | 48 (24.0) | 36 (20.6) | ||||||
| 25.0–29.9 (pre_obesity) | 3,119 (35.1) | 2,151 (38.3) | 968 (29.5) | 2,995 (35.2) | 124 (33.1) | 2,081 (38.5) | 914 (29.5) | 70 (35.0) | 54 (30.9) | ||||||
| 30.0–34.9 (obesity I) | 2,329 (26.2) | 1,526 (27.2) | 803 (24.5) | 2,237 (26.3) | 92 (24.5) | 1,477 (27.3) | 760 (24.5) | 49 (24.5) | 43 (24.6) | ||||||
| Dyslipidemia | 5,815 (65.4) | 3,753 (66.9) | 2,062 (62.9) | 0.0002 | 5,517 (64.8) | 298 (79.5) | <0.0001 | 3,594 (66.4) | 1,923 (62.0) | 159 (79.5) | 139 (79.4) | <0.0001 | |||
| Hypertension | 8,211 (92.4) | 5,183 (92.4) | 3,028 (92.4) | 0.9212 | * | * | <0.0001 | 4,988 (92.2) | 2,854 (92.0) | * | * | 0.0001 | |||
| Smoker | 1,049 (11.8) | 734 (13.1) | 315 (9.6) | <0.0001 | 973 (11.4) | 76 (20.3) | <0.0001 | 692 (12.8) | 281 (9.1) | 42 (21.0) | 34 (19.4) | <0.0001 | |||
| Ejection fraction | |||||||||||||||
| Mean (STDEV) | 54.9 (11.8) | 53.6 (12.0) | 57.3 (11.2) | <0.0001 | 55.0 (11.8) | 53.6 (12.9) | 0.026 | 53.7 (11.9) | 57.4 (11.1) | 51.3 (13,2) | 56.2 (12.0) | <0.0001 | |||
| <0.0001 | 0.008 | <0.0001 | |||||||||||||
| 35–50 | 1,608 (18.1) | 1,163 (20.7) | 445 (13.6) | 1,521 (17.9) | 87 (23.2) | 1,114 (20.6) | 407 (13.1) | 49 (24.5) | 38 (21.7) | ||||||
| >=50 | 6,346 (71.4) | 3,779 (67.3) | 2,567 (78.4) | 6,104 (71.7) | 242 (64.5) | 3,660 (67.6) | 2,444 (78.8) | 119 (59.5) | 123 (70.3) | ||||||
| Diabetes | 3,969 (44.7) | 2,524 (45.0) | 1,445 (44.1) | 0.4238 | 3,758 (44.1) | 211 (56.3) | <0.0001 | 2,412 (44.6) | 1,346 (43.4) | 112 (56.0) | 99 (56.6) | <0.0001 | |||
| Chronic lung disease | 0.5383 | 0.1 | 0.5699 | ||||||||||||
| None | 6,306 (71.0) | 4,001 (71.3) | 2,305 (70.4) | 6,060 (71.2) | 246 (65.6) | 3,879 (71.7) | 2,181 (70.3) | 122 (61.0) | 124 (70.9) | ||||||
| Mild | 1,553 (17.5) | 944 (16.8) | 609 (18.6) | 1,470 (17.3) | 83 (22.1) | 895 (16.5) | 575 (18.5) | 49 (24.5) | 34 (19.4) | ||||||
| Moderate | 520 (5.9) | 319 (5.7) | 201 (6.1) | 499 (5.9) | 21 (5.6) | * | * | * | * | ||||||
| Severe | 508 (5.7) | 347 (6.2) | 161 (4.9) | 483 (5.7) | 25 (6.7) | * | * | * | * | ||||||
| Immunosuppressed | 533 (6.0) | 328 (5.8) | 205 (6.3) | 0.4301 | 507 (6.0) | 26 (6.9) | 0.436 | 313 (5.8) | 194 (6.3) | 15 (7.5) | 11 (6.3) | 0.6527 | |||
| Peripheral arterial disease | 2,098 (23.6) | 1,339 (23.9) | 759 (23.2) | 0.4565 | 1,967 (23.1) | 131 (34.9) | 1,269 (23.5) | 698 (22.5) | 70 (35.0) | 61 (34.9) | <0.0001 | ||||
| Cerebrovascular disease | 2,838 (31.9) | 1,730 (30.8) | 1,108 (33.8) | 0.0035 | 2,728 (32.0) | 110 (29.3) | 0.27 | 1,678 (31.0) | 1,050 (33.9) | 52 (26.0) | 58 (33.1) | 0.013 | |||
| Renal failure | 311 (3.5) | 215 (3.8) | 96 (2.9) | 0.0257 | 256 (3.0) | 55 (14.7) | <0.0001 | 180 (3.3) | 76 (2.5) | 35 (17.5) | 20 (11.4) | <0.0001 | |||
| New York Heart Association Class | <0.0001 | 0.002 | <0.0001 | ||||||||||||
| I&II | 1,429 (16.1) | 913 (16.3) | 516 (15.8) | 1,396 (16.4) | 33 (8.8) | 896 (16.6) | 500 (16.1) | * | * | ||||||
| III | 2,303 (25.9) | 1,250 (22.3) | 1,053 (32.1) | 2,207 (25.9) | 96 (25.6) | 1,202 (22.2) | 1,005 (32.4) | 48 (24.0) | 48 (27.4) | ||||||
| IV | 504 (5.7) | 285 (5.1) | 219 (6.7) | 479 (5.6) | 25 (6.7) | 277 (5.1) | 202 (6.5) | * | * | ||||||
| Not Documented | 1,481 (16.7) | 989 (17.6) | 492 (15.0) | 1,408 (16.5) | 73 (19.5) | 952 (17.6) | 456 (14.7) | 37 (18.5) | 36 (20.6) | ||||||
| Missing | 3,170 (35.7) | 2,174 (38.7) | 996 (30.4) | 3,022 (35.5) | 148 (39.5) | 2,084 (38.5) | 938 (30.2) | 90 (45.0) | 58 (33.1) | ||||||
| Elective Surgery | 6,142 (69.1) | 3,769 (67.2) | 2,373 (72.4) | <0.0001 | 5,926 (69.6) | 216 (57.6) | <0.0001 | 3,655 (67.5) | 2,271 (73.2) | 114 (57.0) | 102 (58.3) | 0.3257 | |||
| First cardiovascular surgery | 8,328 (93.7) | 5,220 (93.0) | 3,108 (94.9) | 0.0003 | 7,970 (93.6) | 358 (95.5) | 0.342 | 5,026 (92.9) | 2,944 (94.9) | 194 (97.0) | 164 (93.7) | 0.0008 |
P-value between sex or race, test for trend; t-test for mean, otherwise Chi-Square test
Categorical data are expressed as counts (percentages), while continuous data are presented as mean (SD).
Cells with less than 11 observations are noted with a
The most common performed operation was isolated CABG (47.8%), followed by TAVR (38.5%), Table 2. Overall, among operations that used a cardiopulmonary bypass, the mean (SD) bypass time was 109.3 (46.1) minutes. Among operations that used the aortic cross-clamp, the mean (SD) cross-clamp time was 84.6(39.0) minutes. The mean (SD) length of stay after operations was 5.7 (4.9) days overall; however, Black beneficiaries had a 1.1-day greater length of stay than non-Blacks (Blacks: 6.7 days versus 5.6 days, p<.0001). The mean length of stay is minimally shorter among females (5.4 days versus 5.8 days, p<.0001). Over 80% of all patients were discharged home. However, Black compared to non-black beneficiaries were less likely to be discharged home (64.2% vs. 81.0%, p<.0001)), and females compared to males were less likely to be discharged home (77.2% vs 82.1%, p<.0001).
Table 2:
Operative Events and Outcomes by sex and race
| Variable | Overall | Male | Female | P value | Non Black | Black | P value | Non-Black Male | Non-Black Female | Black Male | Black Female | P value | |||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Total | 8,887 | 5,611 (63.1) | 3,276 (36.9) | 8,512 (95.8) | 375 (4.2) | 5,411 (60.9) | 3,101 (34.9) | 200 (2.3) | 175 (2.0) | ||||||
| Procedure | <0.0001 | <0.0001 | <0.0001 | ||||||||||||
| CABG | 4,248 (47.8) | 3,044 (54.3) | 1,204 (36.8) | 4,023 (47.3) | 225 (60.0) | 2,916 (53.9) | 1,107 (35.7) | 128 (64.0) | 97 (55.4) | ||||||
| SAVR | 597 (6.7) | 348 (6.2) | 249 (7.6) | * | * | 333 (6.2) | 239 (7.7) | * | * | ||||||
| SAVR+CABG | 621 (7.0) | 460 (8.2) | 161 (4.9) | * | * | 453 (8.4) | 158 (5.1) | * | * | ||||||
| TAVR | 3,421 (38.5) | 1,759 (31.3) | 1,662 (50.7) | 3,306 (38.8) | 115 (30.7) | 1,709 (31.6) | 1,597 (51.5) | 50 (25.0) | 65 (37.1) | ||||||
| Cardiopulmonary Bypass Duration | 0.0001 | 0.1065 | 0.0002 | ||||||||||||
| N | 5213 | 3,684 | 1,529 | 4,967 | 246 | 3,543 | 1,424 | 141 | 105 | ||||||
| Mean minutes (SD) | 109.3 (46.1) | 110.9 (46.4) | 105.5 (45.3) |
109.1 (46.3) | 114.0 (42.6) |
110.6 (46.5) |
105.5 (45.7) | 119.6 (43.5) |
106.4 (40.4) |
||||||
| Aortic Cross Clamp Duration | 0.0002 | 0.7722 | 0.0021 | ||||||||||||
| N | 5,122 | 3,617 | 1,505 | 4,882 | 240 | 3,479 | 1,403 | 138 | 102 | ||||||
| Mean minutes (SD) | 84.6 (39.0) |
85.9 (39.5) | 81.4 (37.7) | 84.6 (39.3) | 83.9 (32.9) | 85.9 (39.8) | 81.5 (38.1) | 87.1 (32.6) | 79.5 (32.9) |
||||||
| Length of Stay (operation to discharge) | 5.7 (4.9) |
5.8 (4.8) | 5.4 (5.2) | <0.0001 | 5.6 (4.9) | 6.7 (5.1) | <0.0001 | 5.8 (4.7) | 5.3 (5.2) | 7.0 (5.4) | 6.4 (4.7) | <0.0001 | |||
| Discharge Location | <0.0001 | <0.0001 | <0.0001 | ||||||||||||
| Home | 7,032 (80.3) | 4,544 (82.1) | 2,488 (77.2) | 6,797 (81.0) | 235 (64.2) | 4,414 (82.7) | 2,383 (78.1) | 130 (66.3) | 105 (61.8) | ||||||
| Extended Care/Rehab | 1,600 (18.3) | 923 (16.7) | 677 (21.0) | 1,476 (17.6) | 124 (33.9) | 858 (16.1) | 618 (20.3) | 65 (33.2) | 59 (34.7) | ||||||
| Nursing Home | 91 (1.0) | * | * | * | * | * | * | * | * | ||||||
| Other (excluded died in hospital) | 34 (0.4) | * | * | * | * | * | * | * | * | ||||||
| Infection | |||||||||||||||
| 90-day | 1,513 (17.0) | 841 (15.0) | 672 (20.5) | <0.0001 | 1,418 (16.7) |
95 (25.3) | <0.0001 | 790 (14.6) | 628 (20.2) | 51 (25.5) | 44 (25.1) | <0.0001 | |||
| 180-day | 1,740 (19.6) | 962 (17.1) | 778 (23.7) | <0.0001 | 1,635 (19.2) | 105 (28.0) | <0.0001 | 908 (16.8) | 727 (23.4) | 54 (27.0) | 51 (29.1) | <0.0001 | |||
| Respiratory Failure | |||||||||||||||
| 90-day | 1,492 (16.8) | 918 (16.4) | 574 (17.5) | 0.1579 | 1,395 (16.4) | 97 (25.9) | <0.0001 | 873 (16.1) | 522 (16.8) | 45 (22.5) | 52 (29.7) | <0.0001 | |||
| 180-day | 1,624 (18.3) | 990 (17.6) | 634 (19.4) | 0.0443 | 1,516 (17.8) | 108 (28.8) | <0.0001 | 937 (17.3) | 579 (18.7) | 53 (26.5) | 55 (31.4) | <0.0001 | |||
| Stroke | |||||||||||||||
| 90-day | 321 (3.6) | 182 (3.2) | 139 (4.2) | 0.0149 | 305 (3.6) | 16 (4.3) | 0.488 | 173 (3.2) | 132 (4.3) | * | * | 0.075 | |||
| 180-day | 357 (4.0) | 201 (3.6) | 156 (4.8) | 0.0063 | 341 (4.0) | 16 (4.3) | 0.801 | 192 (3.5) | 149 (4.8) | * | * | 0.0421 | |||
| Mortality | |||||||||||||||
| Discharge | 0.1949 | 0.122 | 0.2515 | ||||||||||||
| Dead | 130 (1.5) | 75 (1.3) | 55 (1.7) | * | * | 71 (1.3) | 50 (1.6) | * | * | ||||||
| 180-day | 0.4087 | 0.615 | 0.5174 | ||||||||||||
| Dead | 516 (5.8) | 317 (5.6) | 199 (6.1) | 492 (5.8) | 24 (6.4) | 302 (5.6) | 190 (6.1) | * | * |
P*value between sex or race, test for trend; t*test for mean, otherwise Chi*Square test
Categorical data are expressed as counts (percentages), while continuous data are presented as mean (SD).
Cells with less than 11 observations are noted with a
Abbreviation: CABG -coronary artery bypass grafting; SAVR -surgical aortic valve replacement; TAVR -transcatheter aortic valve replacement
Notable differences existed in outcomes across sex and race. The overall incidence of 180-day infection was 19.6%, although was higher among females (23.7% versus 17.1%, p<.0001) and Black beneficiaries (28.0% versus 19.2%, p<.0001). The overall 180-day mortality rate was 5.8%. Rates of mortality were 13.6% higher among beneficiaries who developed an infection within 180 days (16.7% versus 3.1%). Similarly, beneficiaries who developed an infection had higher mortality rates at 30- and 90-days (Supplementary Table 3). Mortality rates within 180 days of CABG and AVR were non-significantly higher among females (6.1% versus 5.6%, p=0.408) and Black beneficiaries (6.4% versus 5.8%, p=0.615), Table 2. Figure 1 shows infection-specific mortality rates across beneficiary sex and race. Female beneficiaries were more likely to have respiratory failure within 180 days (19.4% versus 17.6%, p=0.044) and a stroke (within 90 days: 4.2% versus 3.2%, p=0.014; 180 days: 4.8% versus 3.6%, p=0.006). Black beneficiaries had higher rates of respiratory failure (within 90 days: 25.9% versus 16.4%, p<0.0001; 180 days: 28.8% versus 17.8%, p<0.0001) (Table 2).
Figure 1. 180-Day Infection-Specific Mortality Rates by Patient Sex and Race.
This figure shows 180-day mortality rates by infection status across patient sex and race. Infection types include urinary tract infection, pneumonia, sepsis, gastrointestinal, bloodstream / central line-associated bloodstream infection, endocarditis, sternal wound infection, conduit harvest site, cannulation site, and cellulitis.
* identifies that the bar is suppressed due to CMS suppression rules.
Female beneficiaries had higher infection rates during both index hospitalization (12.2% versus 8.1%, p<0.0001) and following discharge (15.7% versus 11.7%, p<0.0001), Table 3. Black beneficiaries had higher infection rates during both index hospitalization (15.5% versus 9.4%, p<0.0001) and following discharge (17.6% versus 13.0%, p=0.0095). Female beneficiaries had a higher incidence of 180-day UTI (8.1% versus 2.5%, p<0.0001), although a lower incidence of pneumonia (3.2% versus 4.1%, p=0.0478). Rates of sepsis were similar across sex, p=0.891. Black beneficiaries had a higher incidence of pneumonia (6.4% versus 3.6%, p=0.0060) although similar UTI rates (p=0.469). Overall, following discharge, infection rates varied by discharge location. Beneficiaries discharged home had the lowest associated cumulative 180-day infection rate (8.9%), followed by those discharged to a nursing home (25.6%), and rehabilitation facilities (31.9%) (Supplemental Table 4).
Table 3.
Rates of infection subtypes stratified by patient sex and race
| By Sex | By Race | ||||||
|---|---|---|---|---|---|---|---|
| All infections | Male (n=5,611) |
Female (n=3,276) | P value | Non-Black (n=8,512) | Black (n=375) | P value | |
| 180-day Infection Rate (%) | 17.1 | 23.7 | <.0001 | 19.2 | 28.0 | <.0001 | |
| Index Infection Rate (%) | 8.1 | 12.2 | <.0001 | 9.4 | 15.5 | <.0001 | |
| Post-Index Infection rate (%) | 11.7 | 15.7 | <.0001 | 13.0 | 17.6 | 0.0095 | |
| Infection subtypes | |||||||
| Urinary Tract Infection + other infections | 2.6 | 5.2 | <.0001 | 3.6 | * | 0.3475 | |
| Urinary Tract Infection only | 2.5 | 8.1 | <.0001 | 4.5 | 5.3 | 0.4686 | |
| Pneumonia + other infections | 3.8 | 4.4 | 0.1730 | 3.9 | 5.9 | 0.0586 | |
| Pneumonia only | 4.1 | 3.2 | 0.0478 | 3.6 | 6.4 | 0.0060 | |
| Sepsis + other infections | 4.1 | 4.8 | 0.1627 | 4.2 | 7.2 | 0.0061 | |
| Sepsis only | 0.7 | 0.7 | 0.8912 | 0.6 | * | 0.0047 | |
| Infections other than Pneumonia or Sepsis or Urinary Tract Infection | 4.1 | 3.9 | 0.5997 | 4.0 | 5.1 | 0.3080 | |
| Secondary Analysis (excluding UTIs) | |||||||
| Male (n=821) | Female (n=513) | P value | Non-Black (n=1,249) | Black (n=85) | P value | ||
| All infections | 821 | 513 | 1249 | 85 | |||
| 180-day Infection Rate (%) | 14.6 | 15.7 | 0.1908 | 14.7 | 22.7 | <0.0001 | |
| Index Infection Rate (%) | 6.4 | 6.1 | 0.6000 | 6.1 | 11.7 | <0.0001 | |
| Post-Index Infection rate (%) | 10.3 | 11.8 | 0.0303 | 10.6 | 15.5 | 0.0033 | |
| Infection subtypes | |||||||
| Pneumonia + other infections | 3.2 | 3.1 | 0.6774 | 3.1 | 5.3 | 0.0155 | |
| Pneumonia only | 4.6 | 4.5 | 0.8610 | 4.5 | 6.9 | 0.0250 | |
| Sepsis + other infections | 3.6 | 3.8 | 0.5824 | 3.6 | 6.7 | 0.0019 | |
| Sepsis only | 1.2 | 1.6 | 0.0950 | 1.3 | * | 0.0719 | |
| Infections other than Pneumonia or Sepsis | 4.5 | 5.0 | 0.2570 | 4.7 | 5.6 | 0.4021 | |
Cells with less than 11 observations are noted with a
Other infections include gastrointestinal infection, bloodstream infection, endocarditis, sternal wound infection, cellulitis, and cannulation site or conduit harvest site infection
In a pre-specified sensitivity analysis that excluded UTI as an infection subtype, the overall incidence of 180-day infection dropped to 15.0%, with rates differing significantly higher among Black beneficiaries (22.7% versus 14.7%, p<0.0001) although non-significantly different across patient sex (p=0.191). The primary findings were otherwise robust, exclusive of significantly higher isolated pneumonia rates among Black beneficiaries (5.3% versus 3.1%, p=0.0155).
Figure 2 displays the risk-adjusted odds of infection by beneficiary sex and race. Female beneficiaries had a 60% increased adjusted odds of infection compared to males (AOR 1.60, 95% CI 1.41–1.80, p<0.0001). The adjusted odds of infection did not significantly differ for Black versus non-Black beneficiaries (p=0.880), Table 4. After including the sex*race interaction term in the multivariable model (non-Black males as reference), non-Black female beneficiaries had increased odds of developing infection (AOR 1.63, 95% CI 1.44–1.85, p<0.0001). There were no significant differences in the odds of developing infection for Black males (p=0.319) or Black females (p=0.196).
Figure 2. Risk-Adjusted Relative Odds of 180-Day Infection by Patient Sex and Race.
The risk-adjusted odds ratio of 180-day infection by patient sex and race. Variables found to have a statistically significant impact on 180-day infection were included in the figure. Infection types include urinary tract infection, pneumonia, sepsis, gastrointestinal, bloodstream / central line-associated bloodstream infection, endocarditis, sternal wound infection, conduit harvest site, cannulation site, and cellulitis. The risk-adjusted model includes patient characteristics, operation type, hospital, and follow-up time covariates.
Table 4.
Multivariable modeling results of 180-Day infection after cardiac surgery (adjusting for survival to 180 days)
| Main effects of sex and race | Interaction of sex and race | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Odds Ratio | 95% Wald Confidence Limits |
P value | Odds Ratio | 95% Wald Confidence Limits |
P value | |||||
| Sex | Sex-Race | |||||||||
| Female | 1.60 | 1.41 | 1.80 | <.0001 | Non-Black Female | 1.63 | 1.44 | 1.85 | <.0001 | |
| Male | Ref | Black Male | 1.21 | 0.83 | 1.76 | 0.3191 | ||||
| Race | Black Female | 1.29 | 0.88 | 1.90 | 0.1962 | |||||
| Black | 0.98 | 0.74 | 1.30 | 0.8804 | Non-Black Male | Ref | ||||
| Non-Black | Ref | |||||||||
| Year | Year | |||||||||
| 2017 | Ref | 2017 | Ref | |||||||
| 2018 | 0.91 | 0.75 | 1.10 | 0.3359 | 2018 | 0.91 | 0.75 | 1.10 | 0.3233 | |
| 2019 | 0.88 | 0.72 | 1.06 | 0.1805 | 2019 | 0.87 | 0.72 | 1.06 | 0.1692 | |
| 2020 | 0.79 | 0.64 | 0.98 | 0.0297 | 2020 | 0.79 | 0.64 | 0.97 | 0.0273 | |
| 2021 | 0.65 | 0.52 | 0.82 | 0.0002 | 2021 | 0.65 | 0.52 | 0.81 | 0.0002 | |
| Procedure | Procedure | |||||||||
| CABG | 1.85 | 1.51 | 2.27 | <.0001 | CABG | 1.86 | 1.52 | 2.28 | <.0001 | |
| SAVR | 1.79 | 1.35 | 2.36 | <.0001 | SAVR | 1.79 | 1.35 | 2.36 | <.0001 | |
| SAVR+CABG | 2.60 | 2.01 | 3.36 | <.0001 | SAVR+CABG | 2.60 | 2.01 | 3.37 | <.0001 | |
| TAVR | Ref | TAVR | Ref | |||||||
| Age on surgery | Age on surgery | |||||||||
| <65 | Ref | <65 | Ref | |||||||
| 65–69 | 1.07 | 0.85 | 1.35 | 0.5627 | 65–69 | 1.08 | 0.85 | 1.36 | 0.5361 | |
| 70–74 | 1.12 | 0.88 | 1.42 | 0.3491 | 70–74 | 1.13 | 0.89 | 1.43 | 0.3225 | |
| 75–79 | 1.30 | 1.02 | 1.66 | 0.0328 | 75–79 | 1.31 | 1.03 | 1.67 | 0.0286 | |
| 80–84 | 1.44 | 1.11 | 1.88 | 0.0062 | 80–84 | 1.45 | 1.12 | 1.89 | 0.0055 | |
| >=85 | 1.90 | 1.43 | 2.53 | <.0001 | >=85 | 1.91 | 1.43 | 2.54 | <.0001 | |
| Dual Eligibility | Dual Eligibility | |||||||||
| Yes | 1.51 | 1.27 | 1.81 | <.0001 | Yes | 1.52 | 1.28 | 1.82 | <.0001 | |
| No | Ref | No | Ref | |||||||
| Body Mass Index | Body Mass Index | |||||||||
| <18.5 (underweight) | 1.23 | 0.69 | 2.20 | 0.4776 | <18.5 (underweight) | 1.23 | 0.69 | 2.19 | 0.4918 | |
| 18.5–24.9 (normal) | Ref | 18.5–24.9 (normal) | Ref | |||||||
| 25.0–29.9 (pre_obesity) | 1.03 | 0.87 | 1.22 | 0.7282 | 25.0–29.9 (pre_obesity) | 1.03 | 0.87 | 1.22 | 0.7168 | |
| 30.0–34.9 (obesity I) | 1.02 | 0.85 | 1.22 | 0.8360 | 30.0–34.9 (obesity I) | 1.02 | 0.85 | 1.22 | 0.8208 | |
| 35.0–39.9 (obesity II) | 0.94 | 0.75 | 1.18 | 0.5769 | 35.0–39.9 (obesity II) | 0.94 | 0.75 | 1.18 | 0.5963 | |
| >=40.0 (obesity III) | 1.55 | 1.21 | 1.98 | 0.0005 | >=40.0 (obesity III) | 1.55 | 1.21 | 1.98 | 0.0005 | |
| Ejection fraction | Ejection fraction | |||||||||
| Not done | 0.88 | 0.47 | 1.64 | 0.6808 | Not done | 0.87 | 0.47 | 1.63 | 0.6665 | |
| <35 | 1.25 | 1.03 | 1.53 | 0.0265 | <35 | 1.25 | 1.02 | 1.52 | 0.0280 | |
| 35–50 | 1.25 | 1.07 | 1.45 | 0.0038 | 35–50 | 1.25 | 1.07 | 1.45 | 0.0038 | |
| >=50 | Ref | >=50 | Ref | |||||||
| Diabetes | Diabetes | |||||||||
| Yes | 1.26 | 1.11 | 1.42 | 0.0002 | Yes | 1.26 | 1.11 | 1.42 | 0.0002 | |
| No | Ref | No | Ref | |||||||
| Chronic lung disease | Chronic lung disease | |||||||||
| None | Ref | None | Ref | |||||||
| Mild | 1.23 | 1.05 | 1.43 | 0.0090 | Mild | 1.22 | 1.05 | 1.42 | 0.0109 | |
| Moderate | 1.52 | 1.20 | 1.91 | 0.0004 | Moderate | 1.52 | 1.20 | 1.91 | 0.0004 | |
| Severe | 1.98 | 1.58 | 2.49 | <.0001 | Severe | 1.97 | 1.57 | 2.48 | <.0001 | |
| Immunosuppressed | Immunosuppressed | |||||||||
| Yes | 1.62 | 1.30 | 2.01 | <.0001 | Yes | 1.61 | 1.30 | 2.01 | <.0001 | |
| No | Ref | No | Ref | |||||||
| Peripheral arterial disease | Peripheral arterial disease | |||||||||
| Yes | 1.34 | 1.17 | 1.53 | <.0001 | Yes | 1.34 | 1.17 | 1.53 | <.0001 | |
| No | Ref | No | Ref | |||||||
| Cerebrovascular disease | Cerebrovascular disease | |||||||||
| Yes | 1.24 | 1.10 | 1.40 | 0.0006 | Yes | 1.24 | 1.10 | 1.40 | 0.0006 | |
| No | Ref | No | Ref | |||||||
| Renal Failure | Renal Failure | |||||||||
| Yes | 1.91 | 1.45 | 2.53 | <.0001 | Yes | 1.90 | 1.44 | 2.50 | <.0001 | |
| No | Ref | No | Ref | |||||||
| New York Heart Association Class | New York Heart Association Class | |||||||||
| I&II | Ref | I&II | Ref | |||||||
| III | 1.27 | 1.04 | 1.55 | 0.0218 | III | 1.27 | 1.04 | 1.55 | 0.0218 | |
| IV | 1.55 | 1.17 | 2.06 | 0.0023 | IV | 1.56 | 1.18 | 2.07 | 0.0019 | |
| Not Documented | 1.11 | 0.89 | 1.38 | 0.3727 | Not Documented | 1.11 | 0.89 | 1.38 | 0.3686 | |
| Missing | 0.77 | 0.63 | 0.95 | 0.0122 | Missing | 0.77 | 0.63 | 0.94 | 0.0117 | |
| Status | Status | |||||||||
| Elective | Ref | Elective | Ref | |||||||
| Urgent | 1.64 | 1.42 | 1.89 | <.0001 | Urgent | 1.64 | 1.42 | 1.89 | <.0001 | |
| Emergent/Emergent Salvage | 1.62 | 0.89 | 2.94 | 0.1146 | Emergent/Emergent Salvage | 1.62 | 0.89 | 2.94 | 0.1150 | |
Odds ratios for centers are not shown
Only significant covariates are included for risk-adjustment
Discussion
In this multicenter study evaluating Medicare beneficiaries who underwent CABG and AVR, postoperative infection was common and varied by patient sex and race. This study contributes to the literature in two specific ways. First, this study is among the largest series to evaluate the independent effect of patient sex and race on 180-day infections following CABG and AVR. In univariate analyses, infection rates were higher among Black versus non-Black and female versus male patients. Nonetheless, after risk-adjustment (including follow-up time), females had 1.6-fold higher odds of developing an infection, although there were no significant race-based differences in the odds of infection. Importantly, there was significant effect modification by sex and race, with non-Black females having a 63% higher adjusted odds of infection relative to non-Black males. Second, this study is among the first to evaluate differences in rates of infection subtypes across sex and race. While isolated UTI and pneumonia rates significantly differed between females versus males, these differences did not persist after excluding UTIs. Nonetheless, rates of isolated pneumonia were 2.4% higher among Blacks versus non-Black beneficiaries.
The prior literature has predominantly focused on a limited set of infection subtypes (e.g., surgical site infection, pneumonia) over a short-term time horizon (in-hospital and/or 30 days).6,15–20 These evaluations, which often have leveraged the standardized STS-ACSD data collection instrument, have highlighted both the rate of infections and their subtypes as well as associated risk factors and their sequelae. Nonetheless, patients undergoing CABG and AVR are at risk for developing a broader set of infection sequelae that are not otherwise contained within the STS-ACSD (e.g., UTIs, gastrointestinal infections) that may, in turn, develop over a longer time horizon. In a multi-institutional prospective cohort study of 5,158 patients across 10 academic hospitals affiliated with the Cardiothoracic Surgical Trials Network, Gelijins et al identified pneumonia, bloodstream infections, and clostridium difficile colitis as the major drivers of infection after cardiac surgery.21 Their study had a follow-up period of 65 days, and 45% of identified infections occurred during the post-discharge period. This present multi-center study adds to the literature by evaluating differences in the rates of infections across patient sex and race within a 180-day time horizon. Females, relative to males, had a 1.6-fold increased adjusted odds (AOR 1.60, 1.41–1.80) of infection, while there were no race-based differences in the odds of infection. Importantly, a pre-specified analysis tested for effect modification by sex and race. Non-Black females versus Non-Black males had a significantly 1.63-fold (1.44–1.85) higher adjusted odds of infections. Further work is warranted to understand the mechanism underlying this finding, as well as accounting for social determinants of health that are often collected outside of the confines of existing clinical registries.22
Efforts to improve benchmarking and quality improvement would benefit from precise risk estimates across important patient subgroups. Accordingly, this study evaluated the rates across patient sex and race. Females had significantly higher rates of 180-day infections relative to men (23.7% versus 17.1%, p<0.0001) both in the index (12.2% versus 8.1%, p<.0001) and post-index (15.7% versus 11.7%, p<.0001) periods. The magnitude and directionality of this effect were similar across most infection subtypes, exclusive of a lower risk among women for isolated pneumonia (3.2% versus 4.1%, p=0.0478). At the same time, women are known to have a higher risk of UTIs than men due in part to shorter urethras among women contributing to growth of bacteria within the bladder.14 After excluding UTIs in a pre-specified sensitivity analysis, this study found equivalent rates of infection subtypes among women and men (Table 3). Blacks, relative to non-Blacks, had higher rates of infections both overall (28.0% versus 19.2%, p<0.0001), within the index (15.5% versus 9.4%, p<0.0001) and post-index (17.6% versus 13.0%, p=0.0095) periods. Rates of isolated pneumonia and sepsis were higher among Blacks relative to non-Blacks. Enumah et al. identified similar findings in a retrospective review of the STS-ACSD of 1,042,506 patients who underwent CABG surgery between 2011 and 2018.12 Enumah et al. reported increased rates of sepsis (1.26% versus 0.79%) and pneumonia (3.23% versus 2.54%) in Black versus White patients.12
Investigators have highlighted persistent disparities in care and outcomes for patients undergoing cardiac surgery. Efforts to address these disparities would benefit from both regional and national initiatives to improve health equity following CABG and AVR. Collaborative learning approaches have successfully advanced the uptake of established quality measures.23–25 Specifically, regional, physician-led collaboratives have successfully leveraged hospital benchmarking and site visiting to address observed interhospital variability in quality. The addition of hospital performance benchmarking across priority populations (e.g., sex, race) should be further explored. Nonetheless, with few exceptions, limited experience exists in designing collaborative learning interventions to specifically address disparities in practice and outcomes across priority populations.26,27 The interpretation and contextualization of these data, as well as the design and undertaking of interventions, will benefit from including additional expertise from community stakeholder groups, health equity experts, and implementation scientists. Examples of potential interventions to address the noted disparities include a higher intensity follow-up schedule for patients at higher risk of infection in the postoperative period, employing increased use of physical therapy among “at risk” groups to promote ambulation and reduce the risk of pneumonia. These interventions could be employed as a bundled infection prevention strategy through quality collaboratives. Equally important, national clinical registries should consider tracking and reporting important social determinants of health (e.g., insurance dual eligibility, neighborhood socioeconomic status indicators) that adversely contribute to observed disparities in care and outcomes.28,29
Several study-related limitations are worthy of consideration. First, while a large, statewide evaluation of the most common performed cardiac surgical operations, the study findings may lack generalizability outside of Michigan and/or among commercial or Medicaid populations. Second, while the broad set of infections included within this study was derived from validated billing (ICD-10) billing codes and standardized STS-ACSD registry fields, the risk of under-reporting or over-reporting of infections may persist although is unlikely to be informative across patient sex and race. Third, while a large, statewide cohort, there was a low proportion of minorities represented in our merged analytical dataset. The low penetration of minority patients may be attributed to the use of Medicare claims files as well as the requirement for continuous fee-for-service coverage during the study period. Fourth, while this study includes both surgical and transcatheter AVR, further research is warranted to understand the unique determinants of infections among those undergoing less invasive procedures. Lastly, despite accounting for patient, procedural, hospital, and follow-up time as covariates in multivariable risk adjustment, other important confounders may remain (e.g., social determinants of health).
In conclusion, this large, multi-center observational cohort study has identified important disparities in the short and long-term infection rates following CABG and AVR. Specifically, observed rates of 180-day infections were 6.6% significantly higher in absolute terms among women versus men and 8.8% higher among Black versus non-Black beneficiaries. While females had 1.6-fold higher risk-adjusted odds of infection, there were no race-based differences in the odds of infection. This study is among the first to identify a significant interaction between patient sex and race on infections, with non-Black females relative to non-Black males having a 63% higher adjusted odds of infection. Rates of sepsis and isolated pneumonia were higher among Black versus non-Black beneficiaries. Transdisciplinary collaborative learning interventions should be considered to address these known disparities in infection rates.
Supplementary Material
Figure 3. Graphical Abstract.
Design and findings from this multicenter evaluation of patient sex and race-based differences of infections after coronary artery bypass grafting and aortic valve replacement.
Central Message
Female Medicare beneficiaries have a 60% higher odds of infections after coronary artery bypass grafting and aortic valve replacement. Targeted interventions and follow-up for females are warranted.
Perspective Statement
Infections vary across patient sex and race after coronary artery bypass grafting (CABG) and aortic valve replacement (AVR). Black versus White beneficiaries have higher rates of pneumonia and sepsis. Female versus male beneficiaries have a 1.6 higher adjusted odds of postoperative infections. Targeted interventions and follow-up are warranted to improve equitable outcomes after CABG and AVR.
Central Picture
Rates of infection subtypes across patient sex and race.
Acknowledgments:
The Michigan Society of Thoracic and Cardiovascular Surgeons Quality Collaborative (MSTCVS-QC) and the Michigan Value Collaborative (MVC) supported this project.
The Michigan Value Collaborative (MVC) is a Blue Cross Blue Shield of Michigan-funded collaborative quality initiative that includes over 100 hospitals and 40 physician organizations across the State of Michigan. MVC provides Michigan hospitals and POs with payment and utilization data for an episode of care from paid, adjudicated claims. For this analysis, MVC provided us with insurance claims data from the following payer source: Medicare Fee-For-Service (FFS).
Support for the MSTCVS-QC and MVC is provided by Blue Cross Blue Shield of Michigan (BCBSM) as part of its Value Partnerships program. Although BCBSM and MSTCVS-QC work collaboratively, the opinions, beliefs, and viewpoints expressed by the author do not necessarily reflect the opinions, beliefs, and viewpoints of BCBSM or any of its employees. The content of this work is solely the responsibility of the authors and does not represent the official views of the Agency for Healthcare Research and Quality.
Funding Statement:
This project was supported by the National Heart, Lung, And Blood Institute of the National Institutes of Health (NHLBI/NIH, T32HL166113) and the Agency for Healthcare Research and Quality (AHRQ, R01HS029026). The NHLBI supports Dr. Pegues under Award Number T32HL166113.
Disclosure Statement: Outside of this work, Dr. Likosky receives research funding from the AHRQ and the National Institutes of Health and serves as a consultant for the American Society of Extracorporeal Technology. Drs. Likosky, Pagani, Thompson and Alnajjar receive partial salary support from BCBSM to advance quality in Michigan. Dr. Thompson receives research funding from the AHRQ. Dr. Pagani is a non-compensated ad hoc scientific advisor for Abbott, Medtronic, FineHeart and BrioHealth Solutions and non-compensated medical monitor for Abiomed, and a member of the Data Safety Monitoring Board for Carmat and Chair, The Society of Thoracic Surgeons Intermacs Task Force. Dr. Alnajjar serves on the Advisory Board for Ethicon, is a proctor and consultant for Edwards Lifesciences, Medtronic, Boston Scientific and Intuitive Surgical. Dr. Chang is a member of the Data Safety and Monitoring Board for the “Developing a Program to Educate and Sensitize Caregivers to Reduce the Inappropriate Prescription Burden in Elderly with Alzheimer’s Disease Study (D-PRESCRIBE-AD)” for the National Institute on Aging. Dr. Barnes received research funding from Boston Scientific, AHRQ, NHLBI, and Blue Cross Blue Shield of Michigan. Dr. Barnes also receives consulting fees from Pfizer, Bristol-Myers Squibb, Janssen, Bayer, AstraZeneca, Sanofi, Anthos, Abbott Vascular, Boston Scientific. Dr. Hawkins received honoraria from Medtronic paid to his institution.
Glossary of Abbreviations
- AVR
Aortic valve replacement
- CABG
Coronary artery bypass grafting
- STS-ASCD
Society of Thoracic Surgeons Adult Cardiac Surgical Database
- MSTCVS-QC
Michigan Society of Thoracic and Cardiovascular Surgeons Quality Collaborative
- CMS
Centers for Medicare and Medicaid Services
- SAVR
Surgical aortic valve replacement
- TAVR
Transcatheter aortic valve replacement
- MVC
Michigan Value Collaborative
Footnotes
IRB Approval: This study was approved by the University of Michigan Medical School’s Institutional Review Board on August 24, 2022 (HUM00214501).
Informed Consent: A waiver of informed consent was granted for the observational analyses associated with this manuscript.
This study is submitted as a “WTS Conference 2024 Manuscript.”
References
- 1.Mazzeffi M, Gammie J, Taylor B, et al. Healthcare-Associated Infections in Cardiac Surgery Patients With Prolonged Intensive Care Unit Stay. Ann Thorac Surg. 2017;103(4):1165–1170. [DOI] [PubMed] [Google Scholar]
- 2.Massart N, Mansour A, Ross JT, et al. Mortality due to hospital-acquired infection after cardiac surgery. J Thorac Cardiovasc Surg. 2022;163(6):2131–2140.e3. [DOI] [PubMed] [Google Scholar]
- 3.Fowler VG Jr, O’Brien SM, Muhlbaier LH, Corey GR, Ferguson TB, Peterson ED. Clinical predictors of major infections after cardiac surgery. Circulation. 2005;112(9 Suppl):I358–65. [DOI] [PubMed] [Google Scholar]
- 4.Brooks JA. The hospital-acquired condition reduction program. Am J Nurs. 2017;117(10):63–66. [DOI] [PubMed] [Google Scholar]
- 5.AHRQ’s healthcare-associated infections program. Accessed July 29 2024. https://www.ahrq.gov/hai/index.html [Google Scholar]
- 6.Strobel RJ, Liang Q, Zhang M, et al. A Preoperative Risk Model for Postoperative Pneumonia After Coronary Artery Bypass Grafting. Ann Thorac Surg. 2016;102(4):1213–1219. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Edwards FH, Engelman RM, Houck P, Shahian DM, Bridges CR, Society of Thoracic Surgeons. The Society of Thoracic Surgeons Practice Guideline Series: Antibiotic Prophylaxis in Cardiac Surgery, Part I: Duration. Ann Thorac Surg. 2006;81(1):397–404. [DOI] [PubMed] [Google Scholar]
- 8.Engelman R, Shahian D, Shemin R, et al. The Society of Thoracic Surgeons practice guideline series: Antibiotic prophylaxis in cardiac surgery, part II: Antibiotic choice. Ann Thorac Surg. 2007;83(4):1569–1576. [DOI] [PubMed] [Google Scholar]
- 9.Likosky DS, Wallace AS, Prager RL, et al. Sources of Variation in Hospital-Level Infection Rates After Coronary Artery Bypass Grafting: An Analysis of The Society of Thoracic Surgeons Adult Heart Surgery Database. Ann Thorac Surg. 2015;100(5):1570–1575; discussion 1575–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.LaPar DJ, Crosby IK, Rich JB, et al. A contemporary cost analysis of postoperative morbidity after coronary artery bypass grafting with and without concomitant aortic valve replacement to improve patient quality and cost-effective care. Ann Thorac Surg. 2013;96(5):1621–1627. [DOI] [PubMed] [Google Scholar]
- 11.Abukhodair A, Alqarni MS, Alzahrani A, et al. Risk factors for postoperative infections in cardiac surgery patients: A retrospective study. Cureus. 2023;15(8):e43614. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Enumah ZO, Canner JK, Alejo D, et al. Persistent Racial and Sex Disparities in Outcomes After Coronary Artery Bypass Surgery: A Retrospective Clinical Registry Review in the Drug-eluting Stent Era. Ann Surg. 2020;272(4):660–667. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Raza SS, Zhou S, Barnett NM, et al. Interhospital variability in 180-day infections following cardiac surgery. J Thorac Cardiovasc Surg. Published online January 10, 2025. doi: 10.1016/j.jtcvs.2025.01.006 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Foxman B Epidemiology of urinary tract infections: incidence, morbidity, and economic costs. Am J Med. 2002;113 Suppl 1A:5S–13S. [DOI] [PubMed] [Google Scholar]
- 15.Mehaffey JH, Hawkins RB, Byler M, et al. Cost of individual complications following coronary artery bypass grafting. J Thorac Cardiovasc Surg. 2018;155(3):875–882.e1. [DOI] [PubMed] [Google Scholar]
- 16.Brescia AA, Rankin JS, Cyr DD, et al. Determinants of Variation in Pneumonia Rates After Coronary Artery Bypass Grafting. Ann Thorac Surg. Published online 2017. doi: 10.1016/j.athoracsur.2017.08.012 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Shih T, Zhang M, Kommareddi M, et al. Center-level variation in infection rates after coronary artery bypass grafting. Circ Cardiovasc Qual Outcomes. 2014;7(4):567–573. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.O’Brien SM, Feng L, He X, et al. The Society of Thoracic Surgeons 2018 Adult Cardiac Surgery Risk Models: Part 2-Statistical Methods and Results. Ann Thorac Surg. 2018;105(5):1419–1428. [DOI] [PubMed] [Google Scholar]
- 19.Jacobs JP, Shahian DM, Badhwar V, et al. The Society of Thoracic Surgeons 2021 Adult Cardiac Surgery Risk Models for Multiple Valve Operations. Ann Thorac Surg. Published online August 4, 2021. doi: 10.1016/j.athoracsur.2021.03.089 [DOI] [PubMed] [Google Scholar]
- 20.Thourani VH, Bonnell L, Wyler von Ballmoos MC, et al. Outcomes of isolated tricuspid valve surgery: A society of Thoracic Surgeons analysis and risk model. Ann Thorac Surg. Published online May 7, 2024. doi: 10.1016/j.athoracsur.2024.04.014 [DOI] [PubMed] [Google Scholar]
- 21.Gelijns AC, Moskowitz AJ, Acker MA, et al. Management practices and major infections after cardiac surgery. J Am Coll Cardiol. 2014;64(4):372–381. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Agency for Healthcare Research and Quality. Agency for Healthcare Research and Quality Social Determinants of Health Data. SDOH Data and Analytics. July 2022. Accessed August 3, 2024. https://www.ahrq.gov/sdoh/data-analytics.html [Google Scholar]
- 23.Likosky DS, Harrington SD, Cabrera L, et al. Collaborative Quality Improvement Reduces Postoperative Pneumonia After Isolated Coronary Artery Bypass Grafting Surgery. Circulation: Cardiovascular Quality and Outcomes. 2018;11(11):e004756. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Brescia AA, Clark MJ, Theurer PF, et al. Establishment and Implementation of Evidence-Based Opioid Prescribing Guidelines in Cardiac Surgery. Ann Thorac Surg. 2021;112(4):1176–1185. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Stewart JW II, Nieter D, Wu X, et al. Advancing cardiotomy suction practices for coronary surgery via multidisciplinary collaborative learning. JTCVS Open. Published online November 2023. doi: 10.1016/j.xjon.2023.11.005 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Wagner CM, Ibrahim AM. Sex disparities after coronary artery bypass grafting and hospital quality. JAMA Netw Open. 2024;7(6):e2414354. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Wagner CM, Joynt Maddox KE, Ailawadi G, Ibrahim AM. Failure to rescue female patients undergoing high-risk surgery. JAMA Surg. Published online October 16, 2024. doi: 10.1001/jamasurg.2024.4574 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Mehaffey JH, Hawkins RB, Charles EJ, et al. Distressed communities are associated with worse outcomes after coronary artery bypass surgery. J Thorac Cardiovasc Surg. 2020;160(2):425–432.e9. [DOI] [PubMed] [Google Scholar]
- 29.Strobel RJ, Kaplan EF, Young AM, et al. Socioeconomic distress is associated with failure to rescue in cardiac surgery. J Thorac Cardiovasc Surg. Published online July 20, 2022. doi: 10.1016/j.jtcvs.2022.07.013 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.