Original Articles
 
Preoperative Serum High Mobility Group Box 1 Protein (HMGB1) Level for Predicting Mortality of Secondary Peritonitis
 
Vineeta Kumari1, Robin Kaushik1, Shivani Jaswal2, Simrandeep Singh1, Sushma Bhardwaj1
Department of 1Surgery and 2Biochemistry, Government Medical College and Hospital, Chandigarh-160030, India. 


Corresponding Author:
Dr. Robin Kaushik
Email: robinkaushik@yahoo.com

Abstract

Aims: Secondary peritonitis is a common surgical emergency worldwide, but there is limited information on whether pre-operative serum High Mobility Group Box Protein 1 (HMGB1) can help in identification of patients at risk for mortality.
Methodology: The in-patient record of 140 patients of secondary peritonitis was prospectively maintained to identify the role of pre-operative serum HMGB1 in predicting mortality.
Results: There were 123 (87.86%) male and 17 (12.14%) female patients with an average age of 38.81 years; the commonest site of perforation was gastroduodenal area (64 patients; 45.71%). Mortality in the present series was 15% (21 patients). Univariate analysis revealed higher mortality in patients above 50 years of age, those with pre-existing comorbidity, and in those patients having on admission APACHE II score greater than 6, respiratory rate greater than 23 per minute, need for vasopressor support, deranged pre-operative serum creatinine (greater than 1.4 mg/dl), and bicarbonate (less than 18 mmol/L); HMGB1 levels higher than 130 ng/ml was associated with significantly higher mortality.
Conclusion: HMGB1 levels (greater than 130 ng/ml) are associated with significantly higher mortality, especially in association with age above 50 years and the need for inotropic support on admission. Understanding the role of HMGB1 may open further avenues for therapy in secondary peritonitis.

Introduction

‘High Mobility Group Box Protein 1’ (HMGB1; also known as alarmin / amphoterin) is an essential mediator in the pathogenesis of multiple organ failure in sepsis, as a damage-associated molecular patterns (DAMP) that initiates and maintains the inflammatory response.1-3 When it is released extracellularly by immune and necrotic / damaged cells, it acts as a cytokine, signalling for and activating pro-inflammatory cells. High levels of HMGB1 disrupt the endothelial barrier, with consequent vascular leakage and tissue hypoperfusion.4 It is considered an early mediator in ischemic (sterile) injury and a late mediator in sepsis, rising after 8 to 32 hours of endotoxin administration.5,6 Serum HMGB1 levels rise higher in sepsis than in non-infectious SIRS3 and correlate well with the severity of septic shock; survivors show a progressive decline inits levels7 but this has not shown any correlation with mortality.8
Although HMGB1 is overexpressed in many cancers (melanoma, pancreas, colon, mesothelioma and glioma),9-12 chronic inflammatory diseases (rheumatoid arthritis, systemic lupus erythematosus and atheromatous plaques),13-15 sterile injury conditions (myocardial infarction, stroke and hemorrhagic shock),16,17 neurological conditions (inflammation, epilepsy and cognitive dysfunction)18 and sepsis,3 there is limited data about HMGB1 in the setting of secondary peritonitis, which is a common surgical emergency worldwide. Therefore, the present study was planned to assess the role of preoperative HMGB1 as a prognostic marker in secondary peritonitis.

Material and Methods

After Institutional Ethics Committee clearance conveyed vide GMC/IEC/2018/273 dated 31st December 2018, this prospective observational, longitudinal study was carried out on 140 patients of secondary peritonitis over a period of one year to study the relationship of HMGB1 with mortality in patients presenting with secondary peritonitis and to establish whether it could be used as a predictor of mortality in this group of patients. The sample size was calculated as 138 patients based on 15% mortality and 90% confidence level with 5% absolute precision.
All adult patients of either sex above 18 years of age who presented with peritonitis were included after due informed consent. Patients operated elsewhere before presentation, unwilling to participate, or those with stroke or myocardial infarction within 1 year of presentation, autism, inflammatory bowel disease, colon or cervical cancer, or those with hyperlipidemia taking statins were excluded from the study. Patients were managed as per standard protocol for peritonitis and operated in the emergency at the earliest possible time.
In addition to routine investigations, 2 ml of venous blood was collected in plain vial and sent to laboratory, where it was centrifuged, serum separated and stored at -800C until analysis. Analysis of serum HMGB1 level was done by ELISA using commercially available kit with a detection range of 31.25-2000 pg/ml.
All the patient particulars, history and clinical findings on admission, pre-operative investigations, details of surgery, intra-operative findings and post-operative course were recorded in the study proforma.
Microsoft Excel and SPSS version 20 and the ‘R Project for Statistical Computing’ software were used for data tabulation and analysis. Mean and standard deviation were used to represent continuous quantitative data, while qualitative data were described in terms of percentages. Chi-square test, Fisher’s exact test and Yates’ corrected test were the different statistical tests used to compare various parameters. Receivers operating characteristic (ROC) curves were used to find out best cut-off values/ critical values of pre-operative serum HMGB1 and other significant parameters. Logistic regression analysis was used to determine independent predictors of mortality.Mortality was adjusted for collinearity using the variance inflation factor. HMGB1 protein levels of patients were described according to patient characteristics and significance of association of these levels was tested using normal tests of proportions. Predictive value of HMGB1 protein was assessed in terms of sensitivity, specificity and predictive values.
The study was registered with the Clinical Trials Registry of India (www.ctri.nic.in)vide CTRI/2019/02/017874 and patients were enrolled only after receiving trial approval.

Results

A total of 140 patients of secondary peritonitis were enrolled in the present series. There were 123 (87.86%) male and 17 (12.14%) female patients with an average age of 38.81 years (range 14 - 82 years). The commonest perforation was gastroduodenal ulcer perforation (64 patients; 45.71%) followed by enteric (jejunoileal, 54 patients; 38.57%). Overall, 121 morbidity events were seen in 78 patients (55.7%), with respiratory (38 patients; 29%) and abdominal wound dehiscence (31 patients; 23.7%) being the most common. Mortality in the present series was 15% (21 patients).
Univariate analysis revealed a higher mortality in patients above 50 years of age, those with pre-existing comorbidities (such as hypertension, gout, diabetes mellitus, chronic obstructive pulmonary disease, hepatitis C, congenital heart disease, seizure disorder), APACHE II score on admission greater than 6, respiratory rate (on admission) greater than 23 per minute, in those needing vasopressor support, mechanical ventilation, and in those with deranged values of pre-operative serum creatinine (greater than 1.4 mg/dl) and bicarbonate (less than 18 mmol/L) ) - Table 1. Patients who developed complications in the post-operative period also had a significantly higher mortality. 




On multivariate analysis, it was found that age (above 50 years), HMGB1 levels greater than 130 ng/ml and the need for vasopressors were significant risk factors for mortality; when all three were present, the mortality was 90% (Table 2). The Variance Inflation Factors (VIF) for the significant variables on multivariate analysis was 1.0261 for Age, 1.0301 for vasoactive support and 1.0050 for HMGB1. The VIF being close to 1 suggests minimal correlation between the predictor variables therefore the p values of the multivariate analysis stay significant.




HMGB1 levels ranged from 21.1 ng/ml to 200.0 ng/ml. Receiver operating curve was constructed and a value of 130 ng/ml was obtained for pre-operative HMGB1 above which mortality was significantly higher (Figure 1). This ‘critical’ value had a sensitivity of 95.24%, specificity of 31.09%, positive predictive value of 19.61%, negative predictive value of 97.37% and accuracy of 40.71% (Table 3).







Discussion

Secondary peritonitis is a common surgical emergency; and at the cellular level, it provides a continuous source of PAMPs and DAMPs, which keep adding fuel (cytokines) to fire (ongoing sepsis) that keeps the “motor of multisystem organ failure” running.19  Despite an exhaustive search of available English language literature, we could come across only a few studies dealing with HMGB1 as a potential marker for mortality in peritonitis3,20 even though it has been extensively evaluated in other conditions such as pneumonia21 and sepsis.22 With this in mind, this observational study was planned with the sole aim of assessing whether a single point measurement of preoperative serum HMGB1 could act as a prognostic marker for mortality in secondary peritonitis, and it was observed that patients with serum HMGB1 value more than 130 ng/ml had an almost seven and half times mortality than in those with value less than 130 ng/ml (19.6% versus 2.6%) - this probably reflects severe sepsis at a cellular level, where sepsis-induced injury causes over expression of HMGB1 and subsequent effects.
Milic et al3 evaluated 100 patients with diffuse secondary peritonitis and reported that preoperative serum levels of HMGB1 (and other inflammatory markers) were statistically highly increased (p<0.001) in patients with peritonitis compared to their control group (patients with hernia), decreased over the next few days, and again rose up, possibly due to release from activated mononuclearcells or infective or persisting necrotic foci. However, they did not correlate elevated HMGB1 levels with mortality. Our results reinforce their findings that HMGB1 is elevated in secondary peritonitis and add further information that it can also be also used to prognosticate mortality in this group of patients. 
Given our findings that HMGB1 is significantly elevated in patients who die in secondary peritonitis, this may suggest a potential for therapeutic manipulation to increase survival. Direct administration of HMGB-1 was reported to be lethal in mice and administration of anti-HMG-1 antibodies, even if delayed, reduced endotoxin induced deaths. The improvement in renal functions in cases of established sepsis showed beneficial effects of anti-HMGB1 antibodies in animal models and possibly supports its use therapeutically.2,5,6,23 It has also been reported that natural anti-HMGB1 antibodies formed during septic shock are protective and lead to a better outcome.24 Our study probably establishes a relationship between high levels of serum HMGB1 and mortality in peritonitis, and may open the field for additional pharmacological interventions targeted to HMGB1 to improve outcomes.
Although the present series shows a role of preoperative serum HMGB1 in risk assessment in secondary peritonitis, a major lacuna is that we did not establish a temporal correlation of HMGB1 in these patients over their hospitalization. Milic et al3 reported a fall in HMGB1 concentration in the post-operative period in patients with peritonitis; however, since HMGB1 was not estimated after surgery in our patients, we are unable to comment on HMGB1 time trends or their association with morbidity or mortality. Similarly, a comparison with other commonly used inflammatory markers such as C-reactive protein and procalcitonin was not performed due to financial constraints.
Another important fact to be kept in mind is that HMGB1 is elevated in sepsis states only, and does not have any ‘normal’ value, various studies done so far have compared their own median/mean values in cases and controls.3,7,21 This is an important factor that needs to be considered in any patient population being studied, since it becomes necessary to establish a critical or watershed value in the population under consideration before using it as a marker for outcome. In the present series, we had to construct an Receiver Operating Characteristic (ROC) curve from the values obtained from all patients to establish a critical value of 130 ng/dl above which there was a significant mortality; although HBMG-1 was significantly higher in patients with secondary peritonitis, the area under the curve was only 0.587. Even at levels higher than 130 ng/dl, HMGB1 was associated with a low specificity (Table 3), suggesting that there might have been other, unidentified confounders in the present series. When analysed solely, HMGB1 had a poor positive predictive value for mortality, but was most useful when combined with age (greater than 50 years) and need for inotropic support on admission, showing 90% mortality in such patients (Table 2). 
Despite these lacunae, the present series showed a correlation between high pre-operative serum HMGB1 levels and mortality in patients with secondary peritonitis. Further studies are needed to establish a linear correlation with morbidity and mortality as well as any possible benefit of interventions to modulate HMGB1 release in this group of patients.

Acknowledgement

The authors would like to acknowledge the help given by Mr. Uday Pratap Singh Parmar in the statistical revisions of the manuscript.
References
  1. Goodwin GH, Sanders C, Johns EW. A new group of chromatin-associated proteins with a high content of acidic and basic amino acids. Eur J Biochem 1973; 38: 14-19.  
  2. Yang H, Tracey KJ. Targeting HMGB1 in inflammation.  Biochem Biophys Acta  2010; 1799: 149-56. .
  3. Milic L, Grigorov I, Krstic S, Ceranic MS, Jovanovic B, Stevanovic J, et al. Serum level of HMGB1 protein and inflammatory markers in patients with secondary peritonitis: time course and the association with clinical status. J Med Biochem 2017; 36: 44-53. 
  4. Mantell LL, Parrish W, Ulloa L. HMGB-1 as a therapeutic target for infectious and inflammatory disorders. Shock 2006; 25: 4-11.  
  5. Wang H, Bloom O, Zhang M, Vishnubhakat JM, Ombrellino M, Che J, et al. HMG-1 as a late mediator of endotoxin lethality in mice. Science 1999; 285: 248-51. 
  6. Yang H, Ochani M, Li J, Qiang X, Tanovic M, Harris HE, et al. Reversing established sepsis with antagonists of endogenous high-mobility group box 1. Proc Nat Acad Sci  2004; 101: 296-301. 
  7. Gibot S, Massin F, Cravoisy A, Barraud D, Nace L, Levy B, et al. High-mobility group box 1 protein plasma concentrations during septic shock. Intensive Care Med 2007; 33: 1347-53. 
  8. van Zoelen MA, Laterre PF, van Veen SQ, van Till JW, Wittebole X, Bresser P, et al. Systematic and local high mobility group box 1 concentrations during severe infection. Crit Care Med 2007; 35: 2799-804. 
  9. Bald T, Quast T, Landsberg J, Rogava M, Glodde N, Lopez-Ramos D, et al. Ultraviolet-radiation-induced inflammation promotes angiotropism and metastasis in melanoma. Nature 2014; 507: 109-13. 
  10. Kang R, Tang D, Schapiro NE, Loux T, Livesey KM, Billiar TR, et al. The HMGB1/RAGE inflammatory pathway promotes pancreatic tumor growth by regulating mitochondrial bioenergetics. Oncogene 2014; 33: 567-77. 
  11. van Beijnum JR, Dings RP, van der Linden E, Zwaans BM, Ramaekers FC, Mayo KH, et al. Gene expression of tumor angiogenesis dissected: specific targeting of colon cancer angiogenic vasculature. Blood 2006; 108: 2339-48. 
  12. Jube S, Rivera ZS, Bianchi ME, Powers A, Wang E, Pagano I, et al. Cancer cell secretion of the DAMP protein HMGB1 supports progression in malignant mesothelioma. Cancer Res 2012; 72: 3290-3301. 
  13. Taniguchi N, Kawahara K, Yone K, Hashiguchi T, Yamakuchi M, Goto M, et al. High mobility group box chromosomal protein 1 plays a role in the pathogenesis of rheumatoid arthritis as a novel cytokine. Arthritis Rheum 2003; 48: 971-81. 
  14. Abdulahad DA, Westra J, Bijzet J, Limburg PC, Kallenberg CG, Bijl M. High mobility group box 1 (HMGB1) and anti-HMGB1 antibodies and their relation to disease characteristics in systemic lupus erythematosus. Arthritis Res Ther 2011; 13: R71. 
  15. Kalinina N, Agrotis A, Antropova Y, DiVitto G, Kanellakis P, Kostolias G, et al. Increased expression of the DNA-binding cytokine HMGB1 in human atherosclerotic lesions: role of activated macrophages and cytokines. Arterioscler Thromb Vasc Biol 2004; 24: 2320-25. 
  16. Krüger B, Krick S, Dhillon N, Lerner SM, Ames S, Bromberg JS, et al. Donor Toll-like receptor 4 contributes to ischemia and reperfusion injury following human kidney transplantation. Proc Natl Acad Sci 2009; 106: 3390-5. 
  17. Andrassy M, Volz HC, Igwe JC, Funke B, Eichberger SN, Kaya Z, et al. High-mobility group box-1 in ischemia-reperfusion injury of the heart. Circulation 2008; 117: 3216-26. 
  18. Muhammad S, Barakat W, Stoyanov S, Murikinati S, Yang H, Tracey KJ, et al. The HMGB1 receptor RAGE mediates ischemic brain damage. J Neurosci 2008; 28: 12023-31. 
  19. Clements TW, Tolonen M, Ball CG, Kirkpatrick AW. Secondary Peritonitis and Intra-Abdominal Sepsis: An Increasingly Global Disease in Search of Better Systemic Therapies. Scand J Surg 2021; 110: 139-49. 
  20. Stevens NE, Chapman MJ, Fraser CK, Kuchel TR, Hayball JD, Diener KR. Therapeutic targeting of HMGB1 during experimental sepsis modulates the inflammatory cytokine profile to one associated with improved clinical outcomes. Sci Rep 2017; 7: 5850. 
  21. Asavarut P, Zhao H, Gu J, Ma D. The role of HMGB1 in inflammation-mediated organ injury. Acta Anaesthesiol Taiwan 2012; 51: 28-33. 
  22. Lee K, Chang Y, Song K, Park YY, Huh JW, Hong SB, et al. Associations between Single Nucleotide Polymorphisms of High Mobility Group Box 1 Protein and Clinical Outcomes in Korean Sepsis Patients. Yonsei Med J 2016; 57: 111-7. 
  23. Tsung A, Sahai R, Tanaka H, Nakao A, Fink MP, Lotze MT, et al. The nuclear factor HMGB1 mediates hepatic injury after murine liver ischemia–reperfusion. J Exp Med 2005; 201: 1135-43. 
  24. Barnay-Verdier S, Fattoum L, Borde C, Kaveri S, Gibot S, Maréchal V. Emergence of autoantibodies to HMGB1 is associated with survival in patients with septic shock. Intensive Care Med  2011; 37: 957-62.