In this study, we found that the HRG levels on POD 1 were significantly lower in the complication group than in the no-complication group. Furthermore, the ability of HRG to predict postoperative complications was superior to that of WBC and CRP, and similar to that of PCT and P-SEP. However, the difference in HRG levels on POD 1 between the no-complication and complication groups was not significant for the patients of the respiratory and hepato–biliary–pancreatic surgery departments, and there were no significant differences in HRG levels on POD 1 between the infectious- and non-infectious-complication groups and mild- and severe-complication groups.
Previous studies have shown that HRG levels negatively correlate with CRP levels in patients with acute inflammation; therefore, HRG may function as a negative acute phase reactant.[16,29] Previous clinical studies have reported that a decrease in HRG levels may be a biomarker for sepsis,[16,20] ventilator-associated pneumonia, preeclampsia, and coronavirus disease 2019. We found that a decrease in HRG levels on POD 1 might predict postoperative complications, and the ability of HRG to predict postoperative complications was more strongly associated with postoperative complications than that of WBC and CRP and had an association strength similar to that of PCT and P-SEP. Furthermore, the HRG levels could independently predict postoperative complications in the multivariate analyses. Thus, our study showed that the HRG level may be a novel and independent biomarker for predicting postoperative complications.
However, there were no significant differences in HRG levels on POD 1 between patients who developed and did not develop postoperative complications in the respiratory and hepato–biliary–pancreatic surgery departments. It could be that decrease in HRG levels on POD 1 might not predict postoperative complications in some clinical departments; however, the detection power may have been insufficient because of the low number of samples. Further studies could confirm this hypothesis. Furthermore, there were no significant differences in HRG levels on POD 1 between patients who developed infectious complications and those who developed non-infectious complications and between the mild- and severe-complication groups. In contrast, in our previous study with patients who had systemic inflammatory response syndrome (SIRS), we concluded that HRG may be a biomarker for detecting infection and may be useful for evaluating severity. This contradictory result may be attributable to differences in the studied populations, i.e., patients who had undergone surgery and patients with SIRS.
The main strength of this study is that it showed the association between HRG levels and postoperative complications. We believe that patients with very low postoperative HRG levels are at high risk of developing postoperative complications, and clinicians need to follow them more closely.
This study had several limitations. First, it was a single-center study. Therefore, it is unclear whether our findings can be applied to other populations. Second, many of the included patients had cancer. It has been reported that HRG may prevent the development of tumors, and the levels vary across breast, ovarian, and lung cancers.[31–33] The variations in HRG levels in other cancers are unknown. These patient characteristics may have influenced the results. Third, the time from the end of surgery to specimen collection varied across patients because we used residual blood samples collected for routine blood tests in the morning of POD 1. This difference in time may have influenced the postoperative changes in HRG levels. Fourth, HRG levels were measured only on POD 1. HRG levels before surgery and the time course of HRG levels after surgery are unknown. Further studies are needed to examine these issues. Fifth, the postoperative complications in this study include surgical complications, which are unrelated to the intrinsic physiology of the patients. Further study limited to medical complications related to the bioactivity of HRG is needed. Sixth, the study included postoperative patients from a variety of clinical departments, with differences in the originating departments between the two groups. This background may have influenced the results. Further studies limited to specific departments are needed. Seventh, the differences in HRG levels between patients with and without postoperative complications were smaller than expected. We expected a difference of approximately 20 μg/mL between the two groups, in reference to the results of our previous study. However, accurate prediction of differences in HRG levels on POD 1 between the two groups was challenging because of the paucity of studies in which postoperative HRG levels have been measured. Hence, our power calculation may have been ineffective.
BMC Anesthesiol. 2022;22(232) © 2022 BioMed Central, Ltd.