To our knowledge, this is the largest cohort study assessing the prognostic role of fibrinogen in adult sHLH. In addition, our study first indicated that plasma fibrinogen level is nonlinearly and inversely associated with poor survival in adult sHLH patients regardless of conventional confounding factors, including age, PLT, EBV infection, ferritin, etiologies and treatment strategies. Further threshold effect analysis showed that the inflection point was 1.76 g/L, and the risk of mortality was inconsistent on the left and right sides of the inflection point.
A few previous studies regarding the association of FIB and its threshold with the risk of mortality defined plasma FIB levels using different categories or cut-off values. In pediatric patients with with HLH [8, 14]. Signoff demonstrated that hypofibrinogenemia (FIB༜1.5 g/L) was independently associated with higher mortality rates (adjusted odds ratio, 6.0; 95% CI, 2.0–18.1), while in adult sHLH studies, Sandrine et al. used 117 ICU patients and identified that coagulation disorders were associated with higher mortality, especially FIB < 2 g/L (adjusted OR = 2.42, 95% CI 1.08–5.41; P = 0.04) [4]. The large sample size of this study, combined with the application of RCS, afforded us greater statistical power and model flexibility to characterize the associations of plasma FIB concentrations with mortality.
The increased risk of mortality associated with low fibrinogen might be ascribed to the activation of lymphocyte and macrophage immune cells, creating an uncontrolled loop of inflammation that is responsible for liver dysfunction, DIC and fibrinolysis. Stimulated macrophages secrete proinflammatory cytokines, including TNF-\(\alpha\), IL-1\(\beta\), and IL-6, which in turn, can release tissue plasminogen activator in excess; correspondingly, an increase in plasmin responsible for fibrinolysis leads to low fibrinogen levels [15, 16]. In addition, high IFN-γ, presented in sHLH, can induce the production of tissue factor expression in activated macrophages, and activated histiocytes can activate factor X through Mac-1 receptors, eventually initiating blood coagulation and aberrantly provoking an overconsumption of fibrinogen [17, 18]. Another mechanism could be caused by diffuse liver infiltration with activated T lymphocytes and macrophages and hypercytokinemia [19]. All of the above mechanisms, alone or together, may be attributed to low fibrinogen levels, which reflect low in vivo disease activity. Hypofibrinogenemia was also indicative of adverse outcomes beyond coagulopathy, as indicated by its persistent association with complicated course and sHLH/MAS with hyperinflammation.
Our study explored the nonlinear association between plasma FIB levels and the risk of mortality and indicated that FIB displayed an L-shaped relationship with the risk of mortality. One hypothesis is that the lower fibrinogen level (≤ 1.76 g/L) is only the reflection of a more severe form of HLH (with more intense cytokine storm and intense hemophagocytic activity). These lower fibrinogen patients who have higher mortality may reach a certain threshold of excessive activated immune cells and an overwhelming systemic inflammation. Early identification of these patients may prompt earlier consideration of alternative therapeutic strategies, including intensive immunotherapies.
Our study has several strengths. This is the first study that addressed the nonlinearity between FIB levels at the diagnosis of adult sHLH and the risk of mortality and further explained this nonlinearity with a threshold effect analysis. However, several limitations of our study were noted. First, due to the single-center retrospective study, the results may not be generalizable to a broader population. Second, although we had fully adjusted a broad set of covariates available to influence prognosis, we could not rule out the possibility that our findings were biased by unmeasured or unrecognized confounders. Finally, the observational study could only demonstrate the association between FIB levels and the prognosis of patients with sHLH but could not provide conclusions for causality.