Surgical patients bleed; this is a reality every cardiac surgeon must confront. The importance of this has led to the rate of reinterventions for bleeding becoming a quality indicator. Yet, the rule has a workaround; sometimes quality can be maintained by avoiding reinterventions at the cost of multiple transfusions. The price paid includes hemolytic reactions and organ damage, such as renal or pulmonary injury secondary to transfusion. Consequently, there is an increase in mechanical ventilation duration, pneumonia incidence, sepsis, renal failure, and postoperative mortality.
When should we intervene? We might consult Kirklin et al., various articles, or even our colleagues; each will likely give different guidelines. Clinical variability reflects a lack of consensus, often seen when attempting to standardize complex processes. In today’s article, Shou et al. aim to identify temporal criteria for indicating reexploration for bleeding.
This retrospective study from the Adult Cardiac Surgery Department at Johns Hopkins Hospital includes all consecutive surgeries from 2010 to 2020, excluding heart transplants, ventricular assist device implantations, and patients with open chest following the initial procedure. The primary aim was to identify the association between reexploration timing and postoperative outcomes. The secondary objective was to evaluate potential relationships between bleeding location and chest tube output. Five categories of bleeding were assessed: suture line, chest wall, mediastinum (perivascular fat or thymus), multiple locations, and “dry reexploration.” Morbidity was analyzed as a composite variable for the following complications: stroke, renal failure, pneumonia, and/or surgical wound infection. Mortality was defined as death within the initial postoperative days.
A total of 10,070 patients were analyzed, of whom 251 (2.5%) required reexploration. Reintervention patients had significantly more comorbidities (p < .05), such as liver disease, dialysis-dependent renal failure, cerebrovascular disease, and endocarditis. This group also had more urgent surgeries (p < .01) and longer cardiopulmonary bypass times (p < .05). A third of these reexplorations were “dry” (n = 75), followed by suture line bleeding (n = 70; 28%). Chest tube output was higher when bleeding involved mediastinal structures, with a median output of 450 mL/h, while dry explorations had a much lower median output of 151 mL/h. Delayed reexploration significantly increased morbidity (0-4 hours 12.3% vs. 25-48 hours 37.5%; p = .001) and mortality (0-4 hours 3.1% vs. 25-48 hours 43.8%; p = .001).
The authors conclude that delayed reexploration for bleeding increased morbidity and mortality. They advocate for early intervention, preferably within the first 4 hours, and highlight biases resulting from using reexploration for bleeding as a quality criterion.
COMMENTARY:
Studies on postoperative bleeding are nothing new, nor is its association with increased morbidity and mortality. It remains unclear whether the risk factor is reintervention itself, blood product transfusion, or a combination of these two with other factors. Few studies have assessed the temporal association between reexploration and clinical outcomes, but all agree that reexploration is preferable when necessary within the first 12 hours.
Although we would like a specific timeframe or output level to indicate reexploration, we should avoid oversimplification. In similar studies, it’s common to cite a postoperative bleeding rate exceeding 1000 mL in the immediate postoperative period as a reexploration criterion, though this remains an arbitrary and lenient threshold. Critical variables, such as active anticoagulation and/or antiplatelet medication, EuroSCORE, and other unweighted risk factors (like liver dysfunction), as well as the characteristics and complexity of the procedure itself (aortic surgery, reoperation) and individual patient conditions (preoperative anemia, female gender, low body surface area, endocarditis/sepsis), were not considered for assessing temporal associations. Additionally, limitations arise from the retrospective, single-center nature of this study. However, the 10-year evaluation period allows time for perioperative improvements, staff experience gains, and protocol changes (e.g., thromboelastography and guided transfusion), introducing new biases in interpreting the results.
In conclusion, in cases of postoperative bleeding, one should not waste time on decision-making. Reexploration should be individually decided based on the patient’s clinical status, the complexity of the intervention, the team’s experience, and hemostasis test results. Some bleeding cases may be managed conservatively. Accepting variability in specific clinical practice aspects is essential; therein lies the art of surgery.
REFERENCE:
Shou BL, Aravind P, Ong CS, Alejo D, Canner JK, Etchill EW, et al. Early Reexploration for Bleeding Is Associated With Improved Outcome in Cardiac Surgery. Ann Thorac Surg. 2023 Jan;115(1):232-239. doi: 10.1016/j.athoracsur.2022.07.037.
