Association Between Adverse Clinical Outcomes After Coronary Artery Bypass Grafting and Perioperative Blood Transfusions

Effect of perioperative erythropoietin on postoperative morbidity and mortality after cardiac surgery: a meta-analysis of randomized controlled trials

OBJECTIVE

Cardiac surgery is known to have high rates of perioperative red blood cell (RBC) transfusions which are associated with increased postoperative mortality and morbidity. Perioperative erythropoietin (EPO) has been suggested to lower perioperative RBC transfusions, and the effect on postoperative morbidity or mortality is unknown.

METHODS

The registered study protocol is available on PROSPERO (CRD42022314538). We searched the Pubmed, EMbase, and Cochrane CENTRAL databases for randomized controlled trials (RCT) of EPO in cardiac surgery. Outcomes were short-term mortality, acute kidney injury (AKI), re-operation, cerebrovascular accident (CVA), perioperative myocardial infarction (MI), infectious complications, and RBC transfusions. RCT studies of perioperative EPO that reported at least one prespecified outcome of interest were included.

RESULTS

A total of 21 RCT's (n = 2,763 patients) were included. Mortality analysis included 17 studies (EPO 1,272 patients, control 1,235) and showed no significant difference (risk difference (RD) 0.0004, 95%CI: -0.016, 0.009). EPO did not reduce the incidence of AKI (RD -0.006, 95% CI: -0.038, 0.026) and reoperation (RD 0.001, 95% CI: -0.013, 0.015). The incidence of CVA (RD -0.004, 95% CI: -0.015, 0.007) and perioperative MI (RD -0.008, 95% CI: -0.021, 0.005) was similar between the groups.

CONCLUSIONS

Although EPO had been proven to reduce perioperative RBC transfusions, we did not find that it reduces the incidence of postoperative short-term mortality, AKI, and reoperation. The study results support that perioperative EPO is also safe, with no rise in thrombotic events, including CVA and perioperative MI.

Liver Status Assessment After Coronary Artery Bypass Grafting

Background Coronary artery bypass grafting (CABG) is a common surgical intervention used to treat severe coronary artery disease. The Model for End-Stage Liver Disease (MELD) score has become a widely used prognostic index for assessing the severity of liver disease and prioritizing liver transplantation. However, its utility in predicting outcomes in cardiac surgery procedures has not been extensively evaluated. Methods This retrospective study gathered data on patients who underwent CABG or CABG combined with other concomitant surgical interventions, such as carotid common or external carotid artery endarterectomy, thoracic aortic aneurysm repair, and aortic or mitral valve replacement or decalcification procedures, at a single tertiary care facility from January 2011 to December 2020. Researchers collected demographic, clinical, and laboratory information, including MELD score and European System for Cardiac Operative Risk Evaluation (EuroSCORE) data. The patients were divided into two groups: the first group included only those who underwent CABG, while the second group comprised patients who underwent CABG along with other concomitant cardiac interventions. Results The MELD score at discharge was significantly higher in the CABG and other interventions group compared to the CABG-only group (median = 14.09, IQR = 7.41-18.7 vs. median = 6.41, IQR = 4.61-9.44, p < 0.001). However, the difference in MELD score at admission between the two groups was not statistically significant (p = 0.328). A p-value < 0.05 was considered statistically relevant, indicating that liver function worsened postoperatively in the patients with additional interventions. The EuroSCORE was also significantly higher in the CABG and other interventions group, suggesting a higher surgical risk as expected (median = 5.74, IQR = 3.54-11.47 vs. median = 3.34, IQR = 1.97-5.66, p < 0.001). Additionally, differences in laboratory parameters, especially coagulation and hemostasis indicators throughout the postoperative period, including the ICU stay (divided into four equal periods based on each patient's total ICU length of stay) and at discharge, indicate a more complex biological state in patients with additional interventions. These findings may have implications for perioperative management and long-term outcomes. Conclusions The elevated MELD score in patients undergoing CABG with additional interventions emphasizes the need for close monitoring of liver function and coagulation status. Evaluating hepatic status preoperatively would be beneficial, and incorporating liver-protective strategies could help mitigate postoperative repercussions. It may also be useful to include liver function parameters in existing cardiovascular risk scores to improve risk assessment.

The Relationship Between Transfusion in Cardiac Surgery Patients and Adverse Outcomes

OBJECTIVES

To understand if red blood cell (RBC) transfusions are independently associated with a risk of mortality, prolonged intubation, or infectious, cardiac, or renal morbid outcomes.

DESIGN

A retrospective review.

SETTING

A single-institution university hospital.

PARTICIPANTS

A total of 2,458 patients undergoing coronary bypass artery graft and/or valvular surgery from July 2014 through January 2018.

INTERVENTIONS

No interventions were done.

MEASUREMENTS AND MAIN RESULTS

The primary outcome was the occurrence of an adverse event or prolonged intubation. Infectious, cardiac, and renal composite outcomes were also defined. These composites, along with mortality, were analyzed individually and then combined to form the "any adverse events" composite. Preoperative demographic and intraoperative parameters were analyzed as univariate risk factors for adverse outcomes. Logistic regression was used to screen variables, with a p value criterion of p < 0.05 for entry into the model selection procedure. A backward selection algorithm was used with variable entry and retention criteria of p < 0.05 to select the final multivariate model. Multivariate logistic regression models were used to determine whether there was an association between the volume of RBC transfusion and the defined adverse event after adjusting for covariates. A p value < 0.01 was considered statistically significant in the final model of each aim to adjust for multiple comparisons. The final logistic models for each of the following outcomes indicate an increased risk of that outcome per each additional unit of RBC transfused. For prolonged intubation, the odds ratio (OR) was 1.493 (p < 0.0001), OR = 1.358 (p < 0.0001) for infectious composite outcomes, OR = 1.247 (p < 0.0001) for adverse renal outcomes, and OR = 1.467 (p < 0.0001) for any adverse event.

CONCLUSIONS

The authors demonstrated a strong independent association between RBC transfusion volume and adverse outcomes after cardiac surgery. Efforts should be undertaken, such as preoperative anemia management and control of coagulopathy, in order to minimize the need for RBC transfusion.

Implementation of a Short-term Treatment Protocol in Anemic Patients before Cardiac Surgery

BACKGROUND

 We assessed whether implementation of an immediate preoperative treatment in anemic patients could result in fewer perioperative packed red blood cell (PRBC) transfusions and improved outcomes in a real-world setting.

METHODS

 From January 1, 2020, to November 31, 2022, we implemented a perioperative protocol for anemic patients (hemoglobin (Hb) level in women <11.5 g/dL, men <12.5 g/dL), which included subcutaneous erythropoietin α, intravenous Iron, and intramuscular vitamin B12 (all given preoperatively) and per os iron and folic acid given once a day postoperatively. We retrospectively compared all patients receiving the protocol to all eligible patients who were operated upon in the 4 years prior to implementation of the protocol. Primary outcome was amount of PRBC transfusions during surgery and index admission.

RESULTS

 In the months after protocol implementation, 114 patients who received the treatment protocol were compared with 236 anemic patients in the 4 years prior to who did not receive the protocol. The treatment reduced total PRBC use (control group median 4 [2-7] units vs. treatment 2 [1-3] units, p < 0.0001) and the incidence of postoperative blood products transfusions (treatment group 58 patients, 50.88% vs. control group 177 patients, 75%, p < 0.0001). Hb prior to discharge was higher among the protocol group (treatment median 9 g/dL [8.3-9.5 g/dL] vs. control 8.6 g/dL [8.1-9.1 g/dL], p = 0.0081).

CONCLUSION

 Despite some differences compared with previously described protocols, the implementation of a perioperative treatment protocol for anemic patients was associated with a reduction in PRBC transfusion in a real-world setting.

Clinical impact of red blood cell transfusion on adverse clinical events in acute heart failure patients with anemia

BACKGROUND

Anemia has been recognized as an important comorbidity in patients with acute heart failure (AHF) and is associated with adverse clinical events. However, the clinical impact of red blood cell (RBC) transfusion in such patients is unclear.

METHOD

This study was a retrospective single-center registry including AHF patients admitted to Kyorin University Hospital between 2007 and 2014. Anemia was defined as a hemoglobin level < 130 g/L in males or < 120 g/L in females. Those with major bleeding with a fall in hemoglobin concentration of >20 g/L were excluded. AHF readmission at 3 months and in-hospital and 2-year all-cause mortality were evaluated.

RESULTS

Of 501 AHF patients, 38 were excluded owing to major bleeding; finally, 463 (age, 77 ± 11 years; males, 58%) were evaluated. RBC transfusion during hospitalization was performed in 112 patients (24%). Hemoglobin level on admission was 105 ± 16 g/L (transfusion, 89 ± 17 g/L; no transfusion, 110 ± 12 g/L; p < 0.001). AHF readmission at 3 months and in-hospital and 2-year all-cause mortality were observed in 46 (10%), 16 (3%), and 121 (26%) patients, respectively. Univariate Cox regression analysis demonstrated that RBC transfusion was not associated with AHF readmission at 3 months (hazard ratio: 0.80; 95% confidence interval: 0.39-1.66) The association did not differ at any hemoglobin concentration or left ventricular ejection fraction value. Multivariate Cox regression analysis revealed similar results. Furthermore, RBC transfusion was not correlated with in-hospital and 2-year all-cause mortality.

CONCLUSIONS

RBC transfusion was not associated with AHF readmission or all-cause mortality.