Abstract
The existence of various coagulation and/or fibrinolytic system disorders (such as inherited thrombophilia) in patients with sepsis could possibly modify host response to infection as well as patient outcome. The aim of the study is to investigate inherited thrombophilic profile in patients with sepsis. Eighty-three patients with sepsis admitted at the Department of Internal Medicine of the University General Hospital of Patras, Greece were included. Thrombophilic profile (factor V G1691A (Leiden), factor V H1299R (R2), prothrombin G20210A, MTHFR C677T, MTHFR A1298C, factor XIII V34L, β-fibrinogen-455 G-A and plasminogen activator inhibitor (PAI)-1 4G/5G) was evaluated using the cardiovascular diseases (CVD) StripAssay based on DNA isolation, PCR and reverse hybridisation. Data were collected from patients’ chart reviews. Seventy patients (84.3%) of the 83 enrolled had at least one thrombophilic mutation. The most common mutations were heterozygous for β-fibrinogen-455 G-A (43.4%), heterozygous for factor XIII V34L (32.5%), PAI-1 4G/4G (26.5%), homozygous MTHFR C677T (22.9%), heterozygous factor V H1299R (R2) (13.3%) and homozygous MTHFR A1298C (12.0%). A 30-day mortality was 14.5%. Multivariate analysis revealed that mortality was independently associated with Simplified Acute Physiology Score II score on admission, pneumonia and fibrinogen on admission. Nine patients (10.8%) developed septic shock. Coagulation disorders on admission, bacteraemia and PAI-1 genotype 5G/5G were independently associated with development of septic shock. The presence of thrombophilic mutations in patients with sepsis may affect their clinical response, and future studies are needed in order to elucidate the role of isolated thrombophilic mutations in patients with sepsis or septic shock.
Significance of this study
What is already known about this subject?
Coagulation disorders are important causes of sepsis morbidity and mortality.
Haemostatic gene polymorphisms’ role on sepsis morbidity have been studied with conflicting results.
4G/4G genotype of plasminogen activator inhibitor 1 was associated with increased risk for complications during meningococcal disease.
What are the new findings?
Low fibrinogen levels were independently associated with mortality.
Mortality was not influenced by thrombophilic profile.
5G/5G genotype of the plasminogen activator inhibitor 1 was independently associated with development of septic shock.
How might these results change the focus of research or clinical practice?
Studies with large number of participants are needed in order to clarify the impact of haemostatic gene polymorphisms and especially plasminogen activator inhibitor 1 on septic shock development and mortality among patients with sepsis.
Introduction
Sepsis remains a major public health issue, which is characterised by a dysregulated host response to infection and leads to increased morbidity and mortality.1 2 It is well known that sepsis is a life-threatening organ dysfunction caused by a dysregulated host response to infection.1 Particularly in sepsis, the coagulation cascade is activated, while the anticoagulant mechanisms and fibrinolysis are downregulated.3–6 In turn, the dysregulation of coagulation system and subsequent thrombin deposition represent a crucial part of the host response to infection that considerably affects severity and clinical course of sepsis.3 6–9 The clinical expression of such dysregulation is deep vein thrombosis and pulmonary embolism10; therefore, it is strongly recommended to use unfractionated heparin or low-molecular-weight heparin in order to prevent aforementioned complications.11
The role of haemostatic gene polymorphisms in sepsis-induced coagulopathy disorders and subsequently to mortality has been studied especially in animal models, with sometimes conflicting results.8 12 The most commonly recognised mutation in haemostatic gene is the factor V Leiden (fVL) mutation and it is thoroughly studied in meningococcal disease. In a previous study, patients with heterozygous fVL mutation had increased risk for complications of purpura fulminans during meningococcal disease, while no association with mortality was found.13 Another gene polymorphism associated with increased risk for complications of purpura fulminans during meningococcal disease and also mortality is the 4G/4G genotype of the plasminogen activator inhibitor (PAI)-1.14
Existence of inherited thrombophilia in patients with sepsis could possibly affect coagulation system, host response to infection and subsequently clinical course of sepsis. Data on the role of haemostatic gene polymorphisms among patients with sepsis are scarce. The aim of the present study is to investigate the role of such polymorphisms on mortality and development of septic shock.
Materials and methods
Study population
The present study is a prospective one and was performed at the Department of Internal Medicine of the University General Hospital, Patras during a 12-month period. Inclusion criteria were age ≥18 years and admission at the Department of Internal Medicine with sepsis. Exclusion criteria were hospitalisation for less than 24 hours, HIV infection, neutropaenia, moderate to severe cirrhosis, as well as patients receiving immunosuppressive agents or anticoagulant therapy.
Patients were enrolled in the study, within 24 hours of admission. Primary outcome was a 30-day mortality. A secondary analysis was performed in order to assess the role of thrombophilic mutation in the development of septic shock. Data (epidemiological data, comorbidities, severity on admission, laboratory and radiological results) were obtained by patients’ chart reviews. Type of infection was defined according to Centers for Disease Control and Prevention (CDC) definition.2 According to the Society of Critical Care Medicine (SCCM)/ European Society of Intensive Care Medicine (ESICM)/ American College of Chest Physicians (ACCP)/ American Thoracic Society (ATS)/ Surgical Infection Society (SIS) consensus conference, sepsis was defined as infection with systemic inflammatory response syndrome, and septic shock as sepsis with arterial hypotension despite adequate fluid resuscitation.15
Genotypic analysis
Blood samples in ethylenediaminetetraacetic acid (EDTA)-containing tubes for DNA studies were collected on enrolment. The following thrombophilic mutations were evaluated using the cardiovascular diseases (CVD) StripAssay T 4–360 (ViennaLab Labordiagnostika): factor V G1691A, factor V H1299R (R2), prothrombin G20210A, MTHFR C677T, MTHFR A1298C, factor XIII V34L, β-fibrinogen-455 G-A and PAI-1 4G/4G. The procedure was based on DNA isolation, PCR and reverse hybridization.
Statistical analysis
Statistical analysis was performed using SPSS V.23.0 (SPSS). The Fisher’s exact test or X2 test was employed for categorical variables, while, Mann-Whitney U test or t-test for continuous ones, as appropriate. Multiple logistic regression analysis was used to calculate unadjusted ORs and 95% CIs. Factors contributing to multicollinearity were excluded from the multivariate analysis. Statistical significance was established at p<0.05.
Results
Overall 83 patients were enrolled in our study. The most common types of infections were urinary tract infection (36 patients, 43.4%), pneumonia (26 patients, 31.3%) and abdominal infections (13 patients, 15.7%), while the remaining eight infections were skin and soft-tissue infections and endocarditis. Fifteen patients (18.1%) developed bacteraemia. Only 34 infections (41.0%) had microbiological confirmation of the pathogen; 29 were caused by gram-negative bacteria and 5 were gram-positive.
Seventy patients (84.3%) had at least one thrombophilic mutation. Three patients (3.6%) had four mutations, 19 (22.9%) had three mutations and 25 (30.1%) had two. The most common mutations were heterozygous for β-fibrinogen-455 G-A (36; 43.4%), heterozygous for factor XIII V34L (27; 32.5%), PAI-1 4G/4G (22; 26.5%), homozygous MTHFR C677T (19; 22.9%), heterozygous factor V H1299R (R2) (11; 13.3%) and homozygous MTHFR A1298C (10; 12.0%).
A 30-day mortality was 14.5% (12 patients). Univariate analysis of risk factors for mortality is depicted in tables 1 and 2. No significant differences were found in a 30-day mortality among patients with and without thrombophilic mutations. Multivariate analysis revealed that Simplified Acute Physiology Score II on admission (p=0.014; OR 1.2, 95% CI 1.0 to 1.4), pneumonia (p=0.016; OR 33.5, 95% CI 1.9 to 580.6) and fibrinogen (p=0.032; OR 0.988, 95% CI 0.977 to 0.999) were associated with a 30-day mortality.
Nine patients (10.8%) developed septic shock necessitating admission at the intensive care unit. Univariate analysis of risk factors for development of septic shock is depicted in tables 2 and 3. Coagulation disorders on admission (p=0.015; OR 10.4, 95% CI 1.6 to 68.4), bacteraemia (p=0.026; OR 15.3, 95% CI 1.7 to 136.1) and PAI-1 genotype 5G/5G (p=0.019; OR 15.1, 95% CI 1.6 to 145.4) were independently associated with the development of septic shock.
Discussion
The majority of patients with sepsis have relevant coagulation abnormalities, ranging from mild decrease in platelet count or weak prolongation of clotting times, to more severe coagulation disorders and sometimes disseminated intravascular coagulation.6 16 17 These coagulation abnormalities could be explained by an extensive bidirectional interaction between sepsis and coagulation.6 9 18 19 Moreover, proinflammatory cytokines and chemokines are the main mediators in this interaction between sepsis and coagulation.18
As previously shown, pneumonia as compared with other types of infections was associated with higher mortality.20–22 Even though bacteraemia was not associated with worst outcome, it was more common among patients that developed septic shock. Gram-negative pathogens predominated in our cohort, since urinary tract infections were the most common type of infection for which microbiological documentation is higher as comparison to other types of infections and especially pneumonia.
In the present study, a high percentage of patients (84.3%) carried at least one thrombophilic mutations, which is in accordance with previous studies conducted in general population, as well as in patients with sepsis.23–28 Moreover, percentages of individual mutations are comparable to that reported from studies deriving from Europe. Even though most of patients included carried thrombophilic mutations, mortality was influenced by neither individual mutations nor accumulation of such mutations.
Fibrinogen’s role in the survival from infection and especially sepsis has been previously established, with higher fibrinogen levels being associated with higher survival.29 An association between higher plasma fibrinogen levels and −455 G/A polymorphism in the promoter region of the fibrinogen-beta gene has been shown by previous studies.30 Even though we found that lower levels of fibrinogen led to higher mortality, the presence of heterozygous (n=34 patients) or homozygous (n=6)–455 G/A polymorphism was not associated with better outcome.
PAI-1 is a crucial inhibitor of fibrinolysis by inhibiting plasminogen activator, so high circulating levels lead to disseminated intravascular coagulation and organ dysfunction and are associated with worst outcome in infected patients and development of septic shock.31 The most common studied polymorphism was a deletion/insertion (4G/5G) one in the promoter region of aforementioned gene. The 4G allele (four guanine bases) has been consistently associated with higher levels of PAI-1, thus leading to higher mortality among septic patients.32 33 To the best of our knowledge, this is the first study to report that 5G/5G genotype, as compared with both 4G/4G and 4G/5G, was associated with a higher risk for septic shock development, even though no association with mortality was found. A possible explanation for our results may be that production of PAI-1 is influenced apart from genetic factors by many other determinants including metabolic factors, lifestyle, renin–angiotensin system and other inflammatory mediators.32 34
From the remaining studied mutations (factor V G1691A (Leiden), factor V H1299R (R2), prothrombin G20210A, MTHFR C677T, MTHFR A1298C and factor XIII V34L), fVL mutation was the most thoroughly investigated in experimental murine models and clinical sepsis studies with controversial results. In a Danish population-based study, fVL mutation was associated with an increased risk of a 28-day mortality among septic patients,35 while Kerlin et al 28 showed a survival advantage among heterozygous fVL septic patients as compared with non-carriers. Nevertheless, a recent meta-analysis showed no association between fVL mutation and mortality risk.12 In the present study, fVL mutation had no influence on mortality, but only four patients (5.6%) were heterozygotes.
The study had some limitations. The development of sepsis and organ dysfunction after an infection has to be considered as an individualised process. The sample size in our study was relatively small for a condition that varies substantially in its clinical spectrum and severity. Even though most of the patients included had at least one thrombophilic mutation, the effect of some rare polymorphisms could not be properly evaluated. Second, no measurement of PAI-1 circulating levels was conducted in order to evaluate if the effect of 5G/5G genotype on septic shock development was a random association or was associated with higher circulating PAI-1 levels.
Although coagulopathy disorders, such as low fibrinogen levels, play an important role in sepsis-related mortality, the role of individual or combination of thrombophilic mutations remains unclear. PAI-1 5G/5G genotype, as compared with both 4G/4G and 4G/5G, was associated with higher risk for septic shock development, while none of the studied polymorphisms influenced mortality. Consequently, future studies with large number of participants are needed in order to clarify the impact of haemostatic gene polymorphisms on mortality among the patients with sepsis or septic shock.
Footnotes
Contributors MM and MK conceived the idea and experimental design. AG and MP-O collected the patients’ data. AG collected the blood samples. AG and MK conducted the molecular experiments. MM supervised the project. MP-O and AG performed the analysis and interpreted the results. AG and MP-O wrote the manuscript. All authors contributed to manuscript revision, read and approved the submitted version.
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
Competing interests None declared.
Ethics approval The study was approved by the Hospital Ethics Committee (no. 571).
Provenance and peer review Not commissioned; externally peer reviewed.
Patient consent for publication Not required.