4. Discussion
In this study, we investigated the incidence, outcome, and risk factors of cardiovascular surgery-associated DIC in patients admitted to our institute. At POD7, 66 patients were diagnosed with DIC, which accounted for roughly 11.9% of all patients included in this study. Patients with DIC had a significantly higher SOFA score, longer anesthesia time, and longer ICU length of stay than those without DIC. However, the incidence of DIC did not affect 90-day mortality. Female sex, pre-operative DIC score, and anesthesia time were found to be risk factors for DIC.
DIC results from tissue factor-mediated initiation of systemic coagulation activation, decreased physiologic anticoagulant activity, and impaired endogenous fibrinolysis. A variety of underlying diseases, such as infections and malignant tumors, can lead to DIC, which may also contribute to mortality [
5]. Trauma and surgery can also cause DIC. In particular, tissue injury and shock during cardiovascular surgery, as well as the use of CPB and anticoagulation, can damage the coagulation system. All patients who undergo cardiovascular surgery will have a coagulation profile of DIC for some time after surgery, but most of them will recover from hemostatic disorder immediately in 3–5 days after surgery. However, we experienced prolonged coagulopathy and DIC occurred in some patients after surgery, and no study has reported on the incidence of cardiovascular surgery-associated DIC. To the best of our knowledge, this is the first study to retrospectively examine the incidence, mortality, and risk factors of cardiovascular surgery-associated DIC.
With regard to the mechanism of post-trauma coagulopathy, Spahn et al., reported that inflammation, loss of coagulation factors, and activation of fibrinolysis are related to the progression of coagulopathy [
6]. Kornblith et al., found tissue injury to be a cause of protein C activation, impaired thrombin formation, and dysregulated fibrinolysis [
7]. The mechanism of trauma-induced coagulopathy may be partly applicable to surgery-induced coagulopathy and DIC. In the case of surgery with CPB, an association between CPB and postoperative coagulopathy has been suggested [
8].
While trauma-induced coagulopathy occurs in the early stage of trauma injury, probably within a few days, our patients with DIC were diagnosed at 7 days postoperatively (POD7). Therefore, the status of coagulation and fibrinolysis likely differs between the two conditions (trauma vs. surgery). The trauma-induced coagulopathy is driven by hemorrhagic shock and extensive tissue disruption. On the other hand, cardiovascular surgery-induced coagulopathy is mainly driven by surgical tissue injury and CPB-related coagulopathy. These differences partly explain the results.
In the present study, DIC scores of ≥4 at POD1 and POD3 were significantly associated with intraoperative blood loss. However, DIC scores of ≥4 at POD7 were not, suggesting that some patients might have had prolonged coagulopathy that took a longer time to recover. We thought that postoperative infection or sepsis is a reason for the delayed recovery from coagulopathy. However, in our institute, all patients who underwent cardiovascular surgery were administered antibiotics postoperatively and serum procalcitonin levels were routinely measured in all patients. The patients with DIC in our study did not have clinical manifestations of infection. It remains unclear what factors contribute to prolonged coagulopathy and DIC.
Zhang et al., conducted a study on chronic DIC and aneurysm and reported that the proportion of females was significantly higher in the DIC group than in the no-DIC group, although the reason for this was unclear [
9]. Estrogen may play a role in thrombosis formation, as some studies have reported that estrogen increases the risk of both arterial and venous thrombosis [
10]. However, female patients enrolled in this study were older and possibly had decreased hormone secretion. Therefore, the reason females are a risk factor for DIC still remains unclear.
We found that patients with a higher preoperative DIC score are more likely to develop DIC at POD7. This means that preoperative patients with coagulopathy may still suffer from coagulopathy one week after surgery, with subsequent progression to DIC. These findings suggest the need to closely monitor the time course of coagulation and fibrinolysis parameters in preoperative patients with coagulopathy.
While anesthesia time was also a significant risk factor for DIC, its impact on DIC may be limited given the very small odds ratio (1.016). Operation time did not have a significant impact on DIC. A large-scale study will be needed to further investigate the impact of anesthesia time on DIC.
This study used the JAAM DIC scoring system to diagnose DIC, because it is a commonly used system in clinical settings in Japan. Gando et al., reported that the JAAM DIC scoring system is comparable to the ISTH DIC scoring system; the most frequently noted underlying diseases in their study were trauma, burn, and surgery [
1,
2]. Demma et al., reported that the JAAM DIC scoring system is more useful than the ISTH DIC scoring system in predicting mortality in patients with DIC after surgery with CPB [
11]. Meanwhile, Grafender et al., reported that ISTH DIC scores predict outcomes in non-septic patients admitted to a cardiovascular ICU [
12]. Based on these reports, we considered the JAAM DIC scoring system to be a well-validated tool for evaluating cardiovascular surgery-associated DIC.
In our institute, anticoagulants are often used to treat DIC. Some patients in the present study were administered recombinant human thrombomodulin and antithrombin concentrates, which are usually used to treat septic DIC in clinical settings and have been reported to be effective [
13,
14]. In a previous case report, recombinant thrombomodulin was useful for treating chronic DIC in a patient with dissecting aortic aneurysm [
15]. In the present study, the proportion of patients with anticoagulation therapy was too small to evaluate the efficacy of anticoagulants. Further investigation will be necessary to examine the effectiveness of anticoagulants in the treatment of cardiovascular surgery-associated DIC.
This study has many limitations. First, since this study was a retrospective study, various confounders might have affected the results. Second, we could not follow patients after they were discharged from the ICU. Thus, it is unknown whether any of our patients developed DIC after ICU discharge. Third, we examined intraoperative blood loss but not blood loss after ICU admission, which could have affected the incidence of DIC (although we excluded patients who underwent re-thoracotomy). Fourth, we did not examine the volume of transfusion during ICU stay. Some patients supplemented cryoprecipitate due to the decreased fibrinogen levels. This may also have influenced the results. Fifth, we did not collect the data regarding anticoagulation therapy and antiplatelet therapy before surgery, which may have had an impact on the results, though the impact would have been minimal considering the half-time of those drugs. Sixth, our results suggest that the use of CPB did not influence the incidence of DIC. However, because of the small sample size, we still cannot exclude the possibility that use of CPB affects postoperative coagulopathy and DIC. Future prospective, multi-center studies are warranted to further evaluate coagulopathy and DIC after cardiovascular surgery.