https://orcid.org/0000-0003-0375-250X
TO THE EDITOR:
COVID-19 is associated with high levels of inflammation and high rates of thrombosis.1 We have shown previously that the inflammatory biomarker γ' fibrinogen (GPF) can reach extreme levels in COVID-19 patients.2 In a previous study, we showed that out of 18 randomly-chosen hospitalized patients, 10 had GPF levels that were higher than any of the 10 601 participants in the Atherosclerosis Risk in Communities study.3 One patient had the highest level of GPF ever reported, 260 mg/dL. For comparison, the reference interval for GPF in healthy adults is 8.8 to 55.1 mg/dL, as determined in the Framingham Offspring Study.4 In addition, GPF is significantly associated with COVID-19 severity, as assessed by oxygen saturation (SpO2) levels.5 Furthermore, up to 70% of patients with long COVID-19 have GPF levels above the upper limit of normal, even 6 weeks after infection.6
GPF, as well as the major isoform of fibrinogen, is expressed in liver hepatocytes and secreted into the bloodstream.7 Interleukin-6 upregulates the expression of the γ' chain messenger RNA by 8.3-fold in vitro, whereas the major isoform of the γ chain is upregulated only 3.6-fold, demonstrating a disproportionate upregulation of the γ' chain during inflammatory conditions.8 High levels of GPF are also seen in other inflammatory diseases, including periodontitis.9
GPF, in addition to being an inflammatory biomarker, has unique biological activities that may contribute to thrombosis. GPF forms fibrin clots that are resistant to fibrinolysis,10 and purified recombinant GPF formed clots that were 3-times stiffer than the major isoform of fibrinogen and had a 10-fold slower fibrinolysis rate.11 Unlike the major isoform of fibrinogen, GPF has a high-affinity thrombin binding site that binds thrombin anion-binding exosite II12,13 and renders thrombin resistant to heparin/antithrombin III inhibition.14 In vitro studies under flow conditions have shown that as the concentration of GPF increases, blood clot formation accelerates and fibrin clot size increases, both at arterial and venous shear rates.15 We, therefore, investigated the hypothesis that high GPF levels would be associated with adverse outcomes in COVID-19 patients.
Patients from the IMPACC (Immunophenotyping Assessment in a COVID-19 Cohort)16 and Biobank studies were included in this investigation. Fifty-two COVID-19 patients at a tertiary care medical center in the IMPACC and Biobank studies were retrospectively enrolled. A total of 140 plasma samples were measured, with an average of 2.69 ± 0.24 (standard error) per patient. GPF levels were measured in EDTA-anticoagulated plasma samples using a commercial enzyme-linked immunosorbent assay (Gamma Diagnostics, Inc, Portland, OR). This study was approved by the Oregon Health & Science University (OHSU) Institutional Review Board and was carried out in accordance with The Code of Ethics of the World Medical Association (Declaration of Helsinki).
Statistical analyses were performed using RStudio (2023.06.1 Build 524). Wilcoxon rank-sum tests were carried out comparing the patients’ highest GPF levels to outcomes.
In this cohort of 52 patients, the outcomes of oxygen use, mechanical ventilation (MV), extracorporeal membrane oxygenation (ECMO) use, acute respiratory distress syndrome (ARDS), intensive care unit (ICU) status, venous thromboembolism (VTE) incidence, and death were analyzed for their association with GPF levels. Twenty-one patients required the use of oxygen, but their highest GPF levels, 112 ± 9.4 mg/dL (mean ± standard error), were not significantly different from patients who did not use oxygen, 86 ± 11 mg/dL (P = .35), using a Wilcoxon rank-sum test (Table 1). However, 14 patients required MV, and their highest GPF levels, 131 ± 17 mg/dL, were significantly higher than patients who did not require a ventilator (Figure 1A; Table 1), 85 ± 7.1 mg/dL (P = .0097). Similarly, 7 patients who required ECMO had significantly higher GPF levels, 142 ± 23 mg/dL, compared to patients who did not require ECMO, 89 ± 7.5 mg/dL (P = .046). Overall, 18 patients were diagnosed with ARDS and had significantly higher GPF levels, 131 ± 13 mg/dL, compared to patients without ARDS, 79 ± 7.5 mg/dL (P = .0010).
Association of patients' highest GPF levels with outcomes
| Outcome (n) | Median GPF level ± SE (mg/dL) | P value (Wilcoxon) |
|---|---|---|
| Oxygen use | .35 | |
| No oxygen (31) | 86 ± 11 | |
| Oxygen (21) | 112 ± 9.37 | |
| Ventilator use | .0097 | |
| No ventilator (38) | 85 ± 7.1 | |
| Ventilator (14) | 131 ± 17 | |
| ECMO use | .046 | |
| No ECMO (45) | 89 ± 7.5 | |
| ECMO (7) | 142 ± 23 | |
| ARDS diagnosis | .0010 | |
| No ARDS (34) | 79 ± 7.5 | |
| ARDS (18) | 131 ± 13 | |
| ICU status | .014 | |
| No ICU (29) | 112 ± 9.8 | |
| ICU (23) | 57 ± 10 | |
| VTE incidence | .64 | |
| No VTE (46) | 97 ± 7.9 | |
| VTE (6) | 116 ± 21 | |
| Death | .013 | |
| Lived (41) | 86 ± 7.3 | |
| Died (11) | 130 ± 18 |
| Outcome (n) | Median GPF level ± SE (mg/dL) | P value (Wilcoxon) |
|---|---|---|
| Oxygen use | .35 | |
| No oxygen (31) | 86 ± 11 | |
| Oxygen (21) | 112 ± 9.37 | |
| Ventilator use | .0097 | |
| No ventilator (38) | 85 ± 7.1 | |
| Ventilator (14) | 131 ± 17 | |
| ECMO use | .046 | |
| No ECMO (45) | 89 ± 7.5 | |
| ECMO (7) | 142 ± 23 | |
| ARDS diagnosis | .0010 | |
| No ARDS (34) | 79 ± 7.5 | |
| ARDS (18) | 131 ± 13 | |
| ICU status | .014 | |
| No ICU (29) | 112 ± 9.8 | |
| ICU (23) | 57 ± 10 | |
| VTE incidence | .64 | |
| No VTE (46) | 97 ± 7.9 | |
| VTE (6) | 116 ± 21 | |
| Death | .013 | |
| Lived (41) | 86 ± 7.3 | |
| Died (11) | 130 ± 18 |
SE, standard error.
Associations between GPF levels and outcomes. (A) Concentrations of the highest GPF levels in patients requiring MV or ECMO, who developed ARDS, or who died. (B) ROC curve to determine the association of the highest GPF levels with death. The area under the curve of was 0.74, and the optimal cut point for death using the Youden index was 111 mg/dL.
Associations between GPF levels and outcomes. (A) Concentrations of the highest GPF levels in patients requiring MV or ECMO, who developed ARDS, or who died. (B) ROC curve to determine the association of the highest GPF levels with death. The area under the curve of was 0.74, and the optimal cut point for death using the Youden index was 111 mg/dL.
For reasons that are unclear, 23 patients who were in the ICU had significantly lower GPF levels (57 ± 10 mg/dL) than ward patients (112 ± 9.8 mg/dL) (P = .014). This was also found in a previous study.17 This may reflect the more intensive interventions that reduce inflammation in ICU patients. Six patients developed VTE, but their highest GPF levels (116 ± 21 mg/dL) were not significantly different from those of patients who did not develop VTE (97 ± 7.9 mg/dL) (P = .64). However, 11 patients died, and they had significantly higher GPF levels (130 ± 18 mg/dL) than patients who lived (86 ± 7.3 mg/dL) (P = .013). Receiver operating characteristic curve analysis revealed an area under the curve of 0.74 for the highest GPF levels and death (Figure 1B). The optimal cut point for death using the Youden index was 111 mg/dL.
The patients' highest GPF levels were significantly associated with adverse outcomes in COVID-19 patients, including the need for MV, ECMO, the development of ARDS, and death. Limitations of this study include the fact that the blood samples were not taken at identical time points in all patients, there was limited sampling in some cases, and collection was not structured. However, the results confirm and extend previous studies of GPF in COVID-19 patients.2,5,6 In these previous studies, the highest GPF levels were significantly associated with COVID-19 severity, as assessed by SpO2 levels,5 whereas other biomarkers, including total fibrinogen, C-reactive protein, ferritin, lactate dehydrogenase, D-dimer, and interleukin-6, were not significantly associated with disease severity. In addition, GPF levels showed a longitudinal association with worsening pulmonary symptoms in 1 case study and decreased dramatically with the administration of steroids in another case study.2
The biochemical properties of GPF provide a potential mechanistic explanation for the association between the highest GPF levels and adverse outcomes. GPF forms fibrin clots that are resistant to fibrinolysis,10 and GPF has a high-affinity thrombin binding site that renders thrombin resistant to heparin/antithrombin III inhibition.14 In an open-label, randomized clinical trial, therapeutic-dose anticoagulation with heparin had no advantage over usual-care pharmacologic thromboprophylaxis, and the trial was stopped for futility.1 Monitoring GPF levels in COVID-19 patients may, therefore, be warranted. Further studies are necessary to determine if the heparin resistance associated with GPF is involved with adverse outcomes.
Acknowledgment: This study was supported by a National Institutes of Health, National Heart, Lung, and Blood Institute grant (R41HL170874).
Contribution: A.K., W.B.M., and D.H.F. designed research; Q.R., L.B., D.X.L., and M.S. performed research; S.M. analyzed data; and A.K., W.B.M., and D.H.F. wrote the manuscript.
Conflict-of-interest disclosure: OHSU and D.H.F. have a significant interest in Gamma Diagnostics, a company that may have a commercial interest in the results of this research. This potential individual and institutional conflict of interest has been reviewed and managed by OHSU. The remaining authors declare no competing financial interests.
Correspondence: David H. Farrell, Donald D. Trunkey Center for Civilian and Combat Casualty Care, Department of Surgery, Oregon Health & Science University, 3181 SW Sam Jackson Park Rd, Portland, OR 97239-3098; email: farrelld@ohsu.edu.
