Reoperative Interval and Perioperative Mortality Risk in Cardiac Surgery

Richard Ramsingh; Mariana Kawalet; Joshua E Insler; Jules J Bakhos; Abhishek Sharma; Penny L Houghtaling; Marijan Koprivanac; Patrick R Vargo; Michael Z Tong; Haytham Elgharably; Edward G Soltesz; Nicholas G Smedira; Eric E Roselli; Shinya Unai; Gösta B Pettersson; Eugene H Blackstone; A Marc Gillinov; Lars G Svensson; Faisal G Bakaeen

doi:10.1001/jamasurg.2024.5499

. 2024 Dec 18;160(4):463–465. doi: 10.1001/jamasurg.2024.5499

Reoperative Interval and Perioperative Mortality Risk in Cardiac Surgery

Richard Ramsingh ¹, Mariana Kawalet ¹, Joshua E Insler ¹, Jules J Bakhos ¹, Abhishek Sharma ¹, Penny L Houghtaling ², Marijan Koprivanac ¹, Patrick R Vargo ¹, Michael Z Tong ¹, Haytham Elgharably ¹, Edward G Soltesz ¹, Nicholas G Smedira ¹, Eric E Roselli ¹, Shinya Unai ¹, Gösta B Pettersson ¹, Eugene H Blackstone ¹, A Marc Gillinov ¹, Lars G Svensson ¹, Faisal G Bakaeen ^1,^✉

¹Department of Thoracic and Cardiovascular Surgery, Miller Family Heart, Vascular & Thoracic Institute, Cleveland Clinic, Cleveland, Ohio

²Department of Quantitative Health Sciences, Lerner Research Institute, Cleveland Clinic, Cleveland, Ohio

Accepted for Publication: September 21, 2024.

Published Online: December 18, 2024. doi:10.1001/jamasurg.2024.5499

^✉

Corresponding Author: Faisal G. Bakaeen, MD, Department of Thoracic and Cardiovascular Surgery, Heart, Vascular & Thoracic Institute, Cleveland Clinic, 9500 Euclid Ave, Desk J4-1, Cleveland, OH 44195 (bakaeef@ccf.org).

Author Contributions: Dr Bakaeen had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: Kawalet, Insler, Koprivanac, Tong, Elgharably, Soltesz, Unai, Pettersson, Gillinov, Svensson, Bakaeen.

Acquisition, analysis, or interpretation of data: Ramsingh, Kawalet, Insler, Bakhos, Sharma, Houghtaling, Vargo, Smedira, Roselli, Pettersson, Blackstone, Bakaeen.

Drafting of the manuscript: Ramsingh, Kawalet, Insler, Bakhos, Tong, Bakaeen.

Critical review of the manuscript for important intellectual content: Ramsingh, Insler, Bakhos, Sharma, Houghtaling, Koprivanac, Vargo, Elgharably, Soltesz, Smedira, Roselli, Unai, Pettersson, Blackstone, Gillinov, Svensson, Bakaeen.

Statistical analysis: Insler, Houghtaling, Blackstone.

Administrative, technical, or material support: Ramsingh, Kawalet, Insler, Sharma, Koprivanac, Tong, Blackstone.

Supervision: Bakhos, Sharma, Koprivanac, Vargo, Tong, Soltesz, Roselli, Unai, Pettersson, Blackstone, Gillinov, Svensson, Bakaeen.

Conflict of Interest Disclosures: Dr Vargo reported personal fees from Edwards Lifesciences, Artivion, Abbott Medical, and Cellphire outside the submitted work. Dr Tong reported personal fees from Abbott and Abiomed outside the submitted work. Dr Elgharably reported speaking honorarium from Edwards LifeSciences, LifeNet Health, and Artivion outside the submitted work. Dr Roselli reported consulting, speaking, and license fees from Artivion and consulting and speaking fees from Edwards Lifesciences, WL Gore, TerumoAortic, Cook and Medtronic outside the submitted work. Dr Gillinov reported personal fees from Edwards Lifesciences, Medtronic, Abbott, ClearFlow, AtriCure, Abbott, Johnson & Johnson, CorCym, AtriCure, Baxter, and Artivion outside the submitted work; in addition, Dr Gillinov had a patent for devices for mitral valve repair issued. No other disclosures were reported.

Data Sharing Statement: See Supplement 2.

Meeting Presentation: This paper was presented at the Association of VA Surgeons meeting; April 10, 2024; Miami, Florida.

Additional Contributions: We thank Ingrid Sprague, BS, MBA, Cleveland Clinic Communications Manager, for her editorial support in the preparation of this manuscript.

^✉

Corresponding author.

PMCID: PMC11983233 PMID: 39693061

Abstract

This research letter discusses the association between reoperative interval and perioperative mortality risk in cardiac surgery.

Cardiac reoperations pose increased risks due to complex pathologies and more patient comorbidities. The risk of injury during resternotomy is increased by the close proximity of cardiovascular structures to the sternum and the presence of adhesions.^1,2 Early reoperations often involve poorly defined edematous tissue planes and bleeding adhesions due to postsurgical angiogenesis. Over time, these adhesions become less vascular and evolve into fibrous structures, facilitating mediastinal dissection in later reoperations.^3,4

We hypothesized that early cardiac reoperations are associated with increased risk compared with later operations. This study examines the effect of time between previous cardiac surgery and reoperation on operative mortality, and uses operative times and blood transfusions as indicators of surgical difficulty. This builds on prior research on reoperative strategies and outcomes.⁵

Methods

From January 2008 to July 2017, 6021 cardiac reoperations involving resternotomy, with available preoperative computed tomography scans, were performed at the Cleveland Clinic. Reoperations included 597 isolated coronary artery bypass graftings (CABGs), 2557 isolated valve surgeries, 1743 thoracic aorta surgeries, 954 combined valve and CABG, and 170 other surgeries. Patients with endocarditis or reoperation within 30 days of the initial surgery were excluded.

We assessed operative mortality for periods of <1 year (early), 1-5 years, 5-15 years, and >15 years. We investigated operative times and intraoperative blood transfusions as surrogates for surgical difficulty. Random forests were used for classification of imbalanced data, incorporating the interval to reoperation as a continuous variable alongside patient and procedural factors to identify risk factors and estimate risk-adjusted mortality (eMethods in Supplement 1).⁶ Use of these data for research was approved by the Cleveland Clinic institutional review board, with patient consent waived. This study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guidelines.

Results

Operative mortality across the cohort was 3.3%, with a reoperation interval ranging from 31 days to 59 years (median, 10 years). Patient characteristics and operative details by time interval are shown in the Table. Unadjusted operative mortality among early reoperations was 3.6% compared with 3.1% at 1-5 years, 3.4% at 5-15 years, and 3.2% at >15 years (P = .96). Time from previous surgery was not found to be an influential predictor of operative mortality. However, there was an increase in mortality among isolated redo CABG cases within the first year (2 of 29; 95% CI, 1.02-25; P = .04) (Figure, A).

Table. Patient Characteristics and Perioperative Details Stratified by Time From Previous Surgery.

Variable	No. (%)				P value
	Year ranges
	<1 (n = 359)	1-5 (n = 1116)	5-15 (n = 2851)	>15 (n = 1695)
Demographics
Age, y^a	62 (19, 86)	64 (19, 88)	68 (18, 93)	68 (19, 90)	<.001
Female	126 (35)	417 (37)	939 (33)	511 (30)	.001
Black race	42 (12)	66 (6.0)	110 (3.9)	57 (3.4)	<.001
Body mass index,^b mean (SD)	28 (5.4)	29 (6.4)	29 (6.0)	28 (5.5)	<.001
Operation group
Isolated CABG	29 (8.1)	87 (7.8)	246 (8.6)	235 (13.7)	<.001
Isolated valve	160 (44.6)	493 (44.2)	1279 (44.9)	625 (36.9)
Aorta	113 (31.5)	348 (31.2)	832 (29.2)	450 (26.5)
CABG+valve	33 (9.2)	137 (12.3)	448 (15.7)	336 (19.8)
Other	24 (6.7)	51 (4.6)	46 (1.6)	49 (2.9)
Noncardiac comorbidities
Hypertension	287 (81)	857 (77)	2159 (76)	1211 (72)	.001
COPD	93 (26)	336 (30)	826 (29)	454 (27)	.15
History of smoking	206 (58)	638 (57)	1601 (56)	927 (55)	.52
Prior stroke	56 (16)	162 (15)	352 (12)	213 (13)	.12
Peripheral arterial disease	55 (15)	178 (16)	484 (17)	287 (17)	.76
Kidney dialysis	12 (3.5)	38 (3.6)	39 (1.4)	18 (1.1)	<.001
Creatinine, mg/dL^a^,^c	1.1 (0.80, 1.5)	1.0 (0.80, 1.5)	1.1 (0.80, 1.5)	1.0 (0.78, 1.4)	.18
Hematocrit, %, mean (SD)^d	35 (6.5)	37 (5.9)	38 (5.6)	38 (5.5)	<.001
Cardiac comorbidities
LVEF, %, mean (SD)	52 (14)	55 (11)	53 (12)	52 (12)	<.001
Prior mediastinal radiation	9 (2.5)	27 (2.4)	44 (1.6)	28 (1.7)	.20
Imaging risk
High anatomic risk	44 (12)	108 (9.7)	211 (7.4)	130 (7.7)	.003
Support
CPB time, min^a	130 (78, 201)	124 (75, 197)	121 (79, 183)	118 (77, 180)	.003
Myocardial ischemic time, min^a	84 (43, 145)	84 (46, 143)	86 (50, 138)	86 (49, 137)	.46
Circulatory arrest	58 (16)	159 (14)	266 (9.3)	121 (7.1)	<.001
Circulatory arrest time, min^a	39 (17, 88)	38 (13, 97)	25 (11, 70)	17 (11, 49)	<.001
Operative time, min^a	402 (295, 533)	376 (271, 513)	364 (272, 490)	360 (270, 493)	<.001
Intraoperative
Major reentry injury	14 (3.9)	36 (3.2)	83 (2.9)	41 (2.4)	.38
Blood products given, unit
Any	274 (77)	753 (68)	1905 (67)	1115 (66)	.001
Red blood cells	223 (62)	573 (52)	1385 (49)	781 (46)	<.001
Platelets	221 (62)	579 (52)	1475 (52)	855 (51)	.001
Fresh frozen plasma	183 (51)	462 (42)	1137 (40)	652 (39)	<.001
Cryoprecipitate	93 (26)	212 (19)	488 (17)	279 (16)	<.001
Mortality
Operative mortality	13 (3.6)	35 (3.1)	96 (3.4)	54 (3.2)	.96

Open in a new tab

Abbreviations: CABG, coronary artery bypass grafting; COPD, chronic obstructive pulmonary disease; CPB, cardiopulmonary bypass; LVEF, left ventricular ejection fraction.

^{^a}

Median (15th, 85th percentiles).

^{^b}

Calculated as weight in kilograms divided by height in meters squared.

^{^c}

To convert to μmol/L, multiply by 88.40.

^{^d}

To convert to proportion of 1.0, multiply by .01.

Figure. — A, Risk-adjusted effect of time interval between previous cardiac surgery and reoperation on operative mortality, stratified by reoperative procedure, based on random forest analysis using 5000 bootstrap samples of the data. B, Variable importance plot of operative mortality. Greater VIMP score corresponds to greater contribution of the respective covariate to the model performance. BUN indicates blood urea nitrogen; CABG, coronary artery bypass graft; LVEF, left ventricular ejection fraction; NYHA, New York Heart Association; TV, tricuspid valve.

Factors most associated with operative mortality included nonelective surgery, lower hematocrit, higher creatinine, higher New York Heart Association functional class, and lower left ventricular ejection fraction (Figure, B). Early reoperations showed a significantly higher anatomical risk (44 [12%]), defined as proximity of cardiovascular structures to the posterior sternal table, required longer operative times (402 minutes; 15th, 85th percentiles 295, 533), and were more likely to need intraoperative blood transfusions (274 of 359 [77%]) (Table).

Discussion

To our knowledge, this is one of the largest studies to date on the effect of time interval to cardiac reoperations on outcomes. It highlights the increased technical complexity of early cardiac reoperations, evident from the prolonged operative times compared with later reoperations. While early reoperations were associated with high anatomic risk and a greater requirement for intraoperative blood transfusions, these factors were not exclusively tied to the time interval. Interestingly, the time interval between surgeries does not generally influence operative mortality, but it may affect the risk in isolated redo CABG cases. However, due to the small number of deaths in this subgroup, conclusions are not definitive. Notably, other studies have reported a higher risk of injury on sternal reentry and mortality in redo CABG.^2,6

These observations underscore the need for thorough preoperative planning and risk assessment, particularly for early redo of patients who underwent CABG. By understanding these complexities, experienced surgeons tailor management strategies, improving outcomes and neutralizing the high risk.⁵

Supplement 1.

eMethods. Appendix A. Variables Considered in Analyses (included in Random Forests)

jamasurg-e245499-s001.pdf^{(27.1KB, pdf)}

Supplement 2.

Data sharing statement

jamasurg-e245499-s002.pdf^{(15.1KB, pdf)}

References

1.Roselli EE, Pettersson GB, Blackstone EH, et al. Adverse events during reoperative cardiac surgery: frequency, characterization, and rescue. J Thorac Cardiovasc Surg. 2008;135(2):316-323, 323.e1-323.e6. doi: 10.1016/j.jtcvs.2007.08.060 [DOI] [PubMed] [Google Scholar]
2.Park CB, Suri RM, Burkhart HM, et al. Identifying patients at particular risk of injury during repeat sternotomy: analysis of 2555 cardiac reoperations. J Thorac Cardiovasc Surg. 2010;140(5):1028-1035. doi: 10.1016/j.jtcvs.2010.07.086 [DOI] [PubMed] [Google Scholar]
3.Christenson JT, Simonet F, Schmuziger M. The impact of a short interval (< or = 1 year) between primary and reoperative coronary artery bypass grafting procedures. Cardiovasc Surg. 1996;4(6):801-807. doi: 10.1177/096721099600400624 [DOI] [PubMed] [Google Scholar]
4.Cannata A, Petrella D, Russo CF, et al. Postsurgical intrapericardial adhesions: mechanisms of formation and prevention. Ann Thorac Surg. 2013;95(5):1818-1826. doi: 10.1016/j.athoracsur.2012.11.020 [DOI] [PubMed] [Google Scholar]
5.Kindzelski BA, Bakaeen FG, Tong MZ, et al. ; Cleveland Clinic Collaboration on Cardiac Reoperations Working Group Collaborators . Modern practice and outcomes of reoperative cardiac surgery. J Thorac Cardiovasc Surg. 2022;164(6):1755-1766.e16. doi: 10.1016/j.jtcvs.2021.01.028 [DOI] [PubMed] [Google Scholar]
6.randomForestSRC . Fast Unified Forests for Survival, Regression, and Classification (RF-SRC). Accessed November 8, 2024. https://cran.r-project.org/package=randomForestSRC

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplement 1.

eMethods. Appendix A. Variables Considered in Analyses (included in Random Forests)

jamasurg-e245499-s001.pdf^{(27.1KB, pdf)}

Supplement 2.

Data sharing statement

jamasurg-e245499-s002.pdf^{(15.1KB, pdf)}

[sld240030r1] 1.Roselli EE, Pettersson GB, Blackstone EH, et al. Adverse events during reoperative cardiac surgery: frequency, characterization, and rescue. J Thorac Cardiovasc Surg. 2008;135(2):316-323, 323.e1-323.e6. doi: 10.1016/j.jtcvs.2007.08.060 [DOI] [PubMed] [Google Scholar]

[sld240030r2] 2.Park CB, Suri RM, Burkhart HM, et al. Identifying patients at particular risk of injury during repeat sternotomy: analysis of 2555 cardiac reoperations. J Thorac Cardiovasc Surg. 2010;140(5):1028-1035. doi: 10.1016/j.jtcvs.2010.07.086 [DOI] [PubMed] [Google Scholar]

[sld240030r3] 3.Christenson JT, Simonet F, Schmuziger M. The impact of a short interval (< or = 1 year) between primary and reoperative coronary artery bypass grafting procedures. Cardiovasc Surg. 1996;4(6):801-807. doi: 10.1177/096721099600400624 [DOI] [PubMed] [Google Scholar]

[sld240030r4] 4.Cannata A, Petrella D, Russo CF, et al. Postsurgical intrapericardial adhesions: mechanisms of formation and prevention. Ann Thorac Surg. 2013;95(5):1818-1826. doi: 10.1016/j.athoracsur.2012.11.020 [DOI] [PubMed] [Google Scholar]

[sld240030r5] 5.Kindzelski BA, Bakaeen FG, Tong MZ, et al. ; Cleveland Clinic Collaboration on Cardiac Reoperations Working Group Collaborators . Modern practice and outcomes of reoperative cardiac surgery. J Thorac Cardiovasc Surg. 2022;164(6):1755-1766.e16. doi: 10.1016/j.jtcvs.2021.01.028 [DOI] [PubMed] [Google Scholar]

[sld240030r6] 6.randomForestSRC . Fast Unified Forests for Survival, Regression, and Classification (RF-SRC). Accessed November 8, 2024. https://cran.r-project.org/package=randomForestSRC

PERMALINK

Reoperative Interval and Perioperative Mortality Risk in Cardiac Surgery

Richard Ramsingh, MD

Mariana Kawalet, MD

Joshua E Insler, MD

Jules J Bakhos, MD

Abhishek Sharma, MD

Penny L Houghtaling, MS

Marijan Koprivanac, MD

Patrick R Vargo, MD

Michael Z Tong, MD

Haytham Elgharably, MD

Edward G Soltesz, MD

Nicholas G Smedira, MD

Eric E Roselli, MD

Shinya Unai, MD

Gösta B Pettersson, MD, PhD

Eugene H Blackstone, MD

A Marc Gillinov, MD

Lars G Svensson, MD, PhD

Faisal G Bakaeen, MD

Abstract

Methods

Results

Table. Patient Characteristics and Perioperative Details Stratified by Time From Previous Surgery.

Figure. Risk-Adjusted Effect of Time Interval Between Previous Cardiac Surgery and Reoperation on Operative Mortality and Variable Importance (VIMP).

Discussion

References

Associated Data

Supplementary Materials

ACTIONS

PERMALINK

RESOURCES

Cite

Add to Collections

PERMALINK

Reoperative Interval and Perioperative Mortality Risk in Cardiac Surgery

Richard Ramsingh, MD

Mariana Kawalet, MD

Joshua E Insler, MD

Jules J Bakhos, MD

Abhishek Sharma, MD

Penny L Houghtaling, MS

Marijan Koprivanac, MD

Patrick R Vargo, MD

Michael Z Tong, MD

Haytham Elgharably, MD

Edward G Soltesz, MD

Nicholas G Smedira, MD

Eric E Roselli, MD

Shinya Unai, MD

Gösta B Pettersson, MD, PhD

Eugene H Blackstone, MD

A Marc Gillinov, MD

Lars G Svensson, MD, PhD

Faisal G Bakaeen, MD

Abstract

Methods

Results

Table. Patient Characteristics and Perioperative Details Stratified by Time From Previous Surgery.

Figure. Risk-Adjusted Effect of Time Interval Between Previous Cardiac Surgery and Reoperation on Operative Mortality and Variable Importance (VIMP).

Discussion

References

Associated Data

Supplementary Materials

ACTIONS

PERMALINK

RESOURCES

Similar articles

Cited by other articles

Links to NCBI Databases