IL-6 rs1800795 polymorphism is associated with septic shock-related death in patients who underwent major surgery: a preliminary retrospective study

To our knowledge, we described for the first time of the relationship between IL-6 rs1800795 polymorphism and risk of septic shock-related death in European patients who underwent major cardiac or abdominal surgery.

The genetic variation at cytokine genes may influence the risk of sepsis and death [15]. Among these genetic variations, the SNPs at IL-6 promoter are important members which might be associated with sepsis risk and death [17]. In our study, patients with IL-6 rs1800795 CC genotype had higher risk of septic shock-related death, suggesting that IL-6 rs1800795 polymorphism may play a major role in pathogenesis of septic shock. We analyzed allelic and genotypic frequencies of IL-6 rs1800795 polymorphism between groups of patients: septic shock patients, SIRS patients, and healthy people (IBS). No significant differences were found, indicating that our cohort did not have any significant bias regarding the distribution of IL-6 polymorphism.

The IL-6 rs1800795 SNP has been previously associated with sepsis and death, but with discrepant results [17‐21]. In European population, the rs1800795 GG genotype was associated with protection against the development of septic shock in patients with pneumococcal community-acquired pneumonia [17] and lower mortality [17, 21]. In Greek population, the rs1800795 GG genotype did not show any association with severe sepsis and mortality [18]. In Asian population, the carriers of rs1800795 C allele had higher risk of septic shock, but not higher risk of death [19]. In a recent meta-analysis, Gao et al. [20] found a statistically significant association between rs1800795 CC genotype and sepsis-related mortality (CC vs. GC/GG: OR 1.92, p = 0.03), which disappeared after Bonferroni correction. In this regard, it should be taken into account that only six studies, four of them with limited sample size (n < 60), were included for meta-analysis about sepsis-related mortality. They concluded that current evidence does not support a direct effect of rs1800795 polymorphism on the risk of sepsis and more investigations would be needed to evaluate the effect of this polymorphism on sepsis mortality.

IL-6 is a proinflammatory cytokine which plays a vital role in the regulation of host immune response in sepsis, and elevated expression of IL-6 is associated with the development of severe sepsis and mortality [11‐14]. This association may be mediated through a variable IL-6 expression encoded by the host. The G to C polymorphism at position-174 of the IL-6 gene (rs1800795) causes differential activity in promoter constructs which up-regulates IL-6 gene transcription and promotes higher circulating levels of IL-6 in individuals that carry rs1800795 G allele [16]. However, Tischendorf et al. observed a quantitative trait locus effect, IL-6 serum concentrations were highest in patients with the rs1800795 GG genotype, followed by CG genotype, and lowest in individuals with CC genotype, but a high ex vivo secretion after LPS stimulation in rs1800795 C carriers was found [26]. These discrepancies among studies (including our study) might be due to the fact that rs1800795 polymorphism is not the causal mutation, or it is not uniquely responsible. Thus, it is possible that the effect rs1800795 SNP may also be due to other SNPs at IL-6 promoter, which are in high linkage disequilibrium with rs1800795 polymorphism forming a haplotype [27, 28]. In our study, plasma IL-6 values could be very helpful and convincing to reinforce the hypothesis that the observed effect of the IL-6 rs1800795 polymorphism is a real biological result. However, IL-6 measurements were not available in this study. As an alternative, a downstream biological measurement such as the white blood cell count, C reactive protein, and procalcitonin was analyzed, but we did not find any significant relationship of IL-6 rs1800795 polymorphism with these markers (data not shown).

We also performed an in silico analysis for evaluating the possible functional implication of rs1800795 polymorphism by using rSNABase (http://​rsnp.​psych.​ac.​cn/​) [29]. This type of analysis allows to study whether the variant is located within regulatory regions and has a possible transcriptional regulatory effect. We found that IL-6 rs1800795 polymorphism, located within the promoter of IL-6, could be part of RNA-binding protein site and could be involved in RNA-binding protein-mediated post-transcriptional regulation. Moreover, etiology and pathology of sepsis are complex, and the rs1800795 polymorphism may act synergistically with other genetic factors [15], which could be contributing to the risk of death in septic shock. These factors may be SNPs in different type of interleukins, which could play different roles in development of sepsis and clinical outcome [15].

In our study, the association between rs1800795 polymorphism and septic shock-related death was also found when the population was stratified by the type of surgery. However, patients with cardiac surgery showed a significant risk of septic shock-related death in the first 7 and 28 days, whereas patients who underwent abdominal surgery had a significant risk during all follow-up (90 days). These different patterns may be intrinsic to each type of surgery, but we should not rule out the impact of the low sample size of surgery groups. Regarding the association found for cardiac surgery subgroup, the extracorporeal circulation and the surgical injury itself produce complex inflammatory responses which can lead to varying degrees of ischemia–reperfusion injury and/or systemic inflammatory response [30]. Under these circumstances, the role of IL-6 rs1800795 polymorphism might be enhanced and this could justify the fact that we found a higher risk of death during the first 4 weeks of follow-up. Moreover, the association found for abdominal surgery subgroup may be due to a faster development of sepsis than in patients who underwent cardiac surgery, because an intra-abdominal infection usually occurs in the first days of post-surgical intervention [4]. In this case, the influence of the IL-6 rs1800795 polymorphism on the development of septic shock-related death seems to be maintained over time, but the effect is significant when the follow-up of patients reaches 90 days.

The attributable fatality rate of septic shock is high but has significantly dropped in last two decades due to the combination of anti-infective treatments and aggressive organ failure supports [4]. However, patients become exposed to ICU-acquired complications that significantly impact on their prognosis [31, 32]. In a recent article, Daviaud et al. [33] reported that early deaths are mainly attributable to intractable multiple organ failure related to the primary infection and late deaths are related to ICU-acquired complications such as nosocomial infections and mesenteric ischemia. In our study, besides the IL-6 rs1800795 polymorphism, we found a number of factors that were significantly associated with death. APACHE-II score was the most relevant factor associated with the very early mortality, but this factor disappears in favor of others like emergency surgery, peritonitis, and heart disease in the late mortality. Moreover, we included in the analysis a high number of known prognostic factors, but we cannot exclude that other variables not recorded in our study could be influencing the clinical outcome.

Finally, several limitations should be taken into account for the correct interpretation of the results. Firstly, this report has a retrospective design and the sample size was relatively small, which could limit the achievement of statistically significant values between rs1800795 polymorphism and death, especially when we performed the analysis stratified by the type of surgery (abdominal or cardiac surgery). Besides, the limited sample size might increase the risk of false positive. However, we control homogeneity by only including patients with septic shock, without mixing different stages of disease. Secondly, differences in COPD were found when patients were stratified by rs1800795 genotype, but COPD was taken into account in the multivariate analysis. Thirdly, we used multiple points of censoring for death, which may cause problems of multiple comparisons. However, here is a considerable controversy about adjusting the “p value” after multiple tests on clinical-orientated studies [34, 35]. In our study, we had a hypothesis supported by theory and previous reports in septic patients. Therefore, we were not literally doing a random search of a meaningful result, and our results should not be affected by the fact of carrying out a high number of statistical tests.