Our study focused on the life-threatening dysregulated host responses to various stressors in patients aged 65 years or older, aiming at correlating host response biomarkers to prognosis. The host response in the special group had its own characteristics. Our study discovered that host response tends to be inhibited with aging, and high mortality rates. This is a supplement to the definition of dysregulated host responses, which clarifies that both hyper and hypohost responses could adversely affect the patient's prognosis. This may be a unique clinical phenotype in this special population.
Aging is a physiological phenomenon accompanied by various changes in the body. With improved living conditions and medical standards, aging has gradually become a social problem. According to the World Health Organization, by the year 2050, the world’s population of people aged 60 years and older will reach 2.1 billion. People aged 80 or older is expected to reach 426 million (https://www.who.int/news-room/fact-sheets/detail/ageing-and-health). An extensive data analysis from Brazilian shows that from 2009 to 2015, the older population increased by 27%, the ICU admission rate increased by 20% and hospital mortality increased from 9.8–11.2% [10]. The same trend has been confirmed by other researchers [5, 11]. Age was a unique risk factor for sepsis independent of disease severity [12]. In our study, we found that the in-hospital mortality rate in patients 85 years or older sharply increased from 3.9–37.8% compared with patients aged 65-74years old, which conforms to the law of aging. However, when stratified by gender, it can be seen that males make up the majority especially in VOPs. A patient's underlying health, their willingness to accept intrusive rescue methods, the financial situation of their family, and other considerations all play a role in whether or not an elderly patient in China gets admitted to an intensive care unit. Generally, when encountering major issues involving ICU admission for the very old patients, family meetings are held to discuss and decide the subsequent therapy. Doctors generally respect the joint decisions of patients and their family members.
Host responses are the weapon for individuals to overcome stress. We have provided a comprehensive explanation of them elsewhere [4]. Compared with young patients, the host response of old patients is unique, especially VOP (over 85 years old) [7]. This may also partly explain why we have found that the in-hospital mortality rate of VOP is significantly higher. Previous studies on how aging affects the host response in critically ill patients are still incomplete. Our study focused on old patients and included more than 300 cases of VOPs in the ICU. The results showed that levels of NLR, PLR, hsCRP, IL-10, WBC, D-dimer, and LDH were relatively lower in VOPs. Meanwhile, levels of L, PLT and FBG were relatively higher. These might be different from other articles. A study based on big data in Henan, China showed that in ostensibly healthy populations, PLT decreased in those 65 years or older compared to those age 18–64, while NLR and PLR increased with age [13]. After sepsis, CRP and L levels were lower in the old patients (age ≥ 65) compared to the younger group, while other inflammation biomarkers, including IL-6, IL-8, IL-10, monocyte chemoattractant protein-1 and N were higher in the old group [14]. A nation-wide study in Netherlands focused on host response biomarkers of critically ill sepsis patients, they found a higher PLT level in those age 70 years or older, while a remarkably lower levels of CRP, soluble E-Selectin, soluble Intercellular Adhesion Molecule 1, and fractalkine were detected compared to those age 50 to 59 years old. The levels of coagulation biomarkers between different age groups were statistically insignificant [7]. One study in VOPs showed that sepsis patients had lower levels of WBC, and higher level of hsCRP [15]. Although the trend of the biomarkers differed, we can conclude from the previous studies that the host response, especially the immune response, became weaker with aging, which is the unique feature for the VOPs. The young body typically have sufficient physiologic and functional reserves. Aging consumes the reserves and hemostasis was easily breaking up [16]. Immune senescence, partly caused by inflammaging (chronic low-grade systemic inflammation), makes the aging body vulnerable to all stressors [17]. Biomarkers activated by pathogen-associated molecular patterns or damage-associated molecular patterns further supress adaptive immunity [18]. As a result, old populations always have a high rate of organ dysfunction, mortality and prolonged length of in-hospital stay and ICU stay.
Our study shows that PLR is a protective factor in the study population. Previous studies on the association between PLR and mortality have not yielded consistent conclusions. A retrospective study showed that a PLR > 200 was significantly associated with mortality, with Odds Ratio (OR) 1.0002 (95% CI: 1.0001 to 1.0004) [19]. A study in Italy showed a statistically insignificant association, Hazard Ratio of 1.006 (95% CI 1.000-1.013, P = 0.058) [20]. In patients with acute kidney injury, U-shaped relationships were found between PLR and 30-day and 90-day mortality [21]. However, in a group of newborn patients, the results were different. Lower PLR was associated with higher hospital mortality (OR 0.85, 95% CI 0.75–0.95) and 90-day mortality (OR 0.85, 95% CI 0.76–0.96) [22]. The ratio of platelet count and lymphocyte count calculated PLR. Thrombocytopenia is common among critically ill patients and is often associated with poor outcomes [23–25]. Many underlying mechanisms account for it. Hematopoietic function deteriorates with age. At the same time, procoagulant response during stress consumes a large amount of platelet [26]. So that we can rarely see patients with higher platelet count in our study population. As a result, higher platelet count has the trend to be a protective factor for mortality.
Our study showed that WBC and L are two risk factors for mortality in the study population, while N is a protective factor in VOPs. This effect is amplified in VOPs. A Korean study including 1000 participants aged 65 years or older showed the protective trend of lymphocyte. Higher levels of neutrophils had a trend of higher rates of mortality [27]. Other studies came to the same conclusions [28, 29]. Another study in Leiden proved that not lymphocyte count, but lower levels of CD4 + T lymphocyte contribute to mortality [30]. However, none of the studies reached the same conclusion as ours. A study focusing on lymphocyte subsets concluded that CD4 + CD45RA- T cell, a Regulatory T cell, increased with age [31]. In intense stressed VOPs, a greater number of regulatory T cells may be involved in regulating the body's inflammatory response, which also represents the severity of the host response. However, although the quantity of regulatory T cells was large, their functions were decreased. As a result, decoupling of stress and host response appears. More lymphocytes represent stronger host responses, which might lead to poorer prognosis. This hypothesis requires more research to confirm in the future. Neutrophils have been proven to be a risk factor for mortality in septic patients [32]. However, those who had a neutrophil count under the upper limit of normality in the emergency room had an increased risk of dying [33]. Results from our study were similar in VOPs. That might be an explanation that higher neutrophils in VOPs are helpful for the aging body to fight against stress.
Contrary to the outcome of mortality, the increase in host response biomarkers is associated with shorter length of ICU stay and in-hospital stay. This may be caused by multiple factors. Firstly, it is jointly decided by the family members and the patients. Emotional support plays an important part in the entire therapy process. VOPs needs more family support than medical support. They often accompany each other after consultation between patients and family members instead of staying in the-hospital or ICU. Secondly, the increasing levels of host response biomarkers stand for the higher the risk of mortality. Thus, the length of in-hospital and ICU stay may be correspondingly shortened. Further research is still needed in the VOPs to demonstrate the above speculation.
We further established a predictive model to evaluate the relationship between host response indicators and mortality. It is challenging to assess the host response status of elderly patients using a single indicator, as previous research on the subject of critically ill patients' host response mostly examined plasma cytokines and other inflammatory markers [34, 35]. The final model is the one selected from several models with the highest predictive performance, which includes not only inflammation biomarkers, but also procoagulant indicators. The model helped to combine the inflammation, immune and coagulation conditions of the host together and provide a comprehensive evaluation of the individuals. What is difficult to explain in the model is the proportion and clinical significance of hsCRP. In the past, nomogram models were built only using factors with prognostic significance in the regression analysis [36]. Although hsCRP is of no significance in the present single-factor analysis, it did have a statistical significance in the previous studies [37, 38]. So, we added it to the model, which turned out to have a higher predictive value. This model is a predictive model for the critically ill patients aged 65 years or older. More studies are needed to verify our predictive model. Fibrinogen should not be linear associated with mortality: both higher and lower levels are risk factors [39–41].
Our study has some limitations. First, it is a six-year single-center prospective study. Although some correlations were found, the results still require larger-scale validation. As this study is part of the national multi-center prospective study focusing on host response in intensive care, the results will be verified in a larger population across China, which is more reliable. Second, we collected the host response biomarkers only on admission to the ICU and predicted the outcome with a single measurement. As the biomarkers were tested on clinical demand, only a small number of participants tested IL-6, IL-8, IL-10, and TNF-α on admission. However, as time changes, the biomarkers will also change affecting the primary outcomes. Therefore, repeated measurements may have a better prediction efficiency. Other factors, like education level and family economic conditions will also influence the results. We have included all the data in our ongoing multicenter study and are look forward to results in a larger population. Third, some results have not been explained clearly, especially in VOPs. More clinical and fundamental studies, focusing on VOPs are needed.