We investigated the age-related differences in the survival benefit of anticoagulant therapy in sepsis in accordance with the JAAM DIC diagnostic criteria. The results of this study indicated that anticoagulant therapy was associated with better outcomes consistent with a higher DIC score in patients aged around 60 to 70 years. Moreover, anticoagulant therapy in septic patients with low DIC scores may be associated with the deterioration of organ dysfunction and poor outcomes.
Previous studies demonstrated that septic patients with DIC were associated with an increased prevalence of multiple organ dysfunction and that the mortality rate of such patients was significantly higher than that of non-DIC patients [16, 27], suggesting that the development of DIC in sepsis is a poor prognosis factor. A recent nationwide multicenter retrospective cohort study suggested that screening for DIC was associated with a survival benefit in patients with sepsis [28]. These results imply that the evaluation and the subsequent intervention for coagulofibrinolytic changes related to sepsis may contribute to the improvement of outcomes in patients with sepsis. To date, however, there has not been clear evidence of treating sepsis-induced coagulopathy with improved outcomes of patients with sepsis.
The important point is that most of the RCTs that evaluated the effect of anticoagulant therapy have targeted patients with “sepsis.” Although the KyberSept trial, a mega-RCT, demonstrated that antithrombin treatment did not affect the 28-day mortality in adult patients with sepsis and septic shock [5], its post-hoc analysis suggested that antithrombin therapy was associated with a significant reduction in mortality only in the patients with DIC but not in the non-DIC patients [29]. These results indicate the importance of selecting the target population for anticoagulant therapy against sepsis. Studies showing that the optimal target for anticoagulant therapy is the patient population fulfilling three factors of sepsis, DIC and high disease severity have recently been published [13, 14, 30–32].
A recent global analysis demonstrated that 1 in 5 deaths around the world is caused by sepsis leading to 11 million deaths annually worldwide, which is more than those killed by cancer [33]. Elderly people make up the major population of patients with sepsis and mortality increases at more advanced ages [34, 35]. Although it is well known that advancing age is associated with increased coagulation, this propensity of deteriorating coagulation disorder during sepsis remains unknown [34]. The innate immune cells such as neutrophils and macrophages show age-related functional alteration, resulting in a diminished ability to rapidly respond to foreign pathogens [36]. In addition to the innate immune system, an attenuation of the adaptive immune functions in old age has been confirmed [37]. A prolonged inflammatory response in aged patients has been noted, as high levels of inflammatory cytokines have been found [38–41]. The interaction between the effect of anticoagulant therapy and the DIC score in patients aged 80 years, which were found in our study may support these previous studies. In other words, the onset of sepsis itself might determine the outcome regardless of the severity of DIC in the elderly. In contrast, patients aged 50 years showed a different interaction between the anticoagulant therapy and the DIC score. The relatively young septic patients are expected to respond well to the treatment of underlying infection, followed by the modulation of dysregulated inflammatory and couagulofibrinolytic reactions. Therefore, it is presumed that the development of DIC in the relatively young septic patients may have little effect on the outcome, while the detailed mechanisms remain unclear.
The present study failed to show the beneficial effect of anticoagulant therapy on the SOFA score of the population of any age. However, it does not mean that organ dysfunction cannot be improved by anticoagulant therapy because the SOFA scores were evaluated relatively early (at 72 h after admission). As anticoagulant therapy improved hospital mortality in certain ages, it may improve the SOFA scores later than 72 h after admission. Our results should be considered when setting up the study designs of future RCTs to validate the effects of anticoagulant therapy against sepsis.
The main pathophysiology of DIC is uncontrolled thrombin generation, leading to ischemic organ dysfunction due to microvascular thrombosis, which is detrimental in the context of pathology. In contrast, the concept that microvascular thrombosis produced by innate immunity is a physiological process to maintain body homeostasis has come to be known as immunothrombosis [3, 42]. From these perspectives, those who could be at an advantage when immunothrombosis is considered, namely, non-DIC patients, and they should not be treated by anticoagulant therapy. The harmful effects of anticoagulant therapy against non-DIC patients were confirmed in the present study (Figs. 3 and 4).
The current study is associated with several limitations. First, although the present data set was prospectively collected, causal relationships could not be defined because of the study’s retrospective design. Second, this study did not assess the dosage and duration of anticoagulant agents. Additionally, this study was unable to evaluate the effects of individual anticoagulant agents or combination therapy because we defined anticoagulant therapy as the administration of antithrombin, recombinant human thrombomodulin, and their combination. Third, the present study data set did not include the information on the adverse effects, including serious bleeding complications associated with anticoagulant therapy, which can cause unfavorable outcomes. Fourth, data elements required to control potential confounders might result in biased effect estimates. Finally, the study being conducted in a single country may limit the generalizability of the obtained results.