Circulating white blood cell traits and prolonged night shifts: a cross-sectional study based on nurses in Guangxi

This manuscript focuses on leukocytes (i.e., total leukocytes, neutrophils, monocytes, lymphocytes, eosinophils, and basophils). Our study was a cohort study of 1540 healthcare workers, in which all participants were divided into two groups based on the presence or absence of long-term shift work experience, and associations between markers of peripheral blood leukocytes, lymphocytes, platelets, and erythrocytes were collected and investigated. In univariate analyses, MPV, PLCR, WBC, BASO%, ELR, MCHC, PLR, RDW-CV, and RDW-SD significantly differed between the long-term shift population and the control population. In subsequent machine learning and multivariate analyses, only WBC and ELR were considered meaningful indicators of immunity. One study found that decreased robustness of circadian rhythms was associated with increased leukocyte counts, increased neutrophil counts, and increased SII levels, which is consistent with our results and suggests that mild disruption of circadian rhythms is associated with an increased inflammatory state. Chronic, low-level pro-inflammatory processes are present in many chronic diseases, and disrupted rest-rest rhythms are associated with increased risk of cardiovascular disease, cancer, and diabetes; we excluded nurses with inflammatory chronic diseases such as cardiovascular disease, cancer, and diabetes at the initial stage of our study to avoid spurious associations from adjusting for multiple confounding factors. While the study¹⁹ showed increased white blood cell counts in nurses working long night shifts, the increase was not pathologic, suggesting that our bodies may have some adaptive or resilient capacity. However, the disadvantages of rhythmic disorders are unquestionable. Still, there may also be some benefits for disease treatment, and experiments have shown that time disorders have beneficial effects on mice carrying lymphomas, which could be used to develop new treatments²⁰.

The current less extensive study of leukocyte levels in a night-shift population showed that total leukocyte, neutrophil, monocyte, and lymphocyte counts were 7.5%, 7.4%, 11.9%, and 7.0% higher, respectively, in a population of 1654 iron and steelworkers in Taiwan who worked shifts (both afternoon and night shifts). In the present study, WBC increased from 6.34 to 7.00, but no significant increase in neutrophil, monocyte, and lymphocyte counts was seen. Although some results differed, both studies suggested an increase in WBC. Also, several studies of airline, male manual, and male factory workers have shown higher total leukocyte counts in night shift workers than in non-shift workers. These cohorts, including our nurse cohort, in several countries (Finland, South Korea, China, and Argentina) and corresponding to different occupations, have amply demonstrated the potential link between leukocyte levels and prolonged night shifts and the possible damage to health.

The eosinophil-to-lymphocyte ratio (ELR) is a biomarker of inflammation. One study found significant decreases in eosinophils and ELR in both deceased and surviving COVID-19 patients^21,22 and substantial reductions in the mean values of both biomarkers in patients with moderate-to-severe disease who were hospitalized for more or fewer than 15 days, suggesting that lower eosinophil counts and ELR may be predictive of a poorer prognosis in COVID-19 patients^23,24. In this study, ELR was also one of the indicators significantly associated with prolonged night shifts. Our results showed that the log (OR) of ELR =  − 3.9 (95% CI: − 5.8– − 2.0) and that prolonged night shift nurses showed a significant decrease in ELR. Also, the machine learning results showed that ELR (Fig. 4) was the most critical indicator affected by night shifts. In today’s neocoronavirus pandemic, the nursing population is an occupation that is exposed to or chronically exposed to neocoronavirus patients, and long-term night-shift nurses also showed a low ELR relative to the low ELR of neocoronavirus-infected patients with a poor prognosis, and potentially suggesting a health risk for night-shift nurses or populations with circadian disorders in a neocoronavirus pandemic. However, the data of our study population came from before the pandemic. We will also collect data on night shift nurses after the pandemic in a follow-up study and compare the hematologic changes associated with the new coronavirus epidemic²⁵.

It has also been suggested that eosinophils exhibit anti-tumor cytotoxic responses in different tumor microenvironments^26,27. Diseases characterized by elevated peripheral eosinophils are negatively associated with cancer risk^13,28. The ELR can estimate the balance between tumor and immunity²⁷. A negative correlation between ELR and cancer risk was found in a prospective cohort study by BioBank UK²⁹.ELR provides a cost-effective and routinely available biomarker that may impact carcinogenesis and can be used to inform early detection of certain cancers. In the present study, night shift nurses also showed reduced ELR, suggesting that prolonged night shifts may increase the risk of female tumors such as breast cancer in the nurse population.

Shift and night work are increasingly common in modern society³⁰. Studies have shown that animals exposed to light at night show an increase in breast cancer development³¹. Although less evidence has been found in humans, various studies have shown those female night shift workers are at a higher risk of developing breast cancer^32,33. The current International Agency for Research on Cancer (IARC) published a report in 2007 stating that shift work involving disrupted circadian rhythms “may be carcinogenic to humans” (Group 2A)³⁴. Two years after the publication of the IARC report, Denmark was obliged to pay compensation to 38 female permanent night shift workers with breast cancer³⁵. However, there is no conclusive evidence regarding the link between night shift work and cancer risk. Still, this paper finds that, like patients with poorer prognosis breast cancer, we also see increased peripheral blood WBC and decreased ELR in night shift nurses.

To eliminate the effect of age on our results³⁶, we similarly found that prolonged night work increased nurses’ WBC levels and decreased ELR values after adjusting for age. Also, prolonged night shifts were more strongly associated with ELR. However, in this study, while observing shift nurses after matching for age propensity, we only observed an increase in WBC due to night shifts among nurses. The difference between the two groups of ELR was not significant.

This study describes the effect of night shift work on peripheral blood hematocrit-related biomarkers. As this was a cross-sectional study^37,38 and lacked follow-up, we will add follow-up for different diseases in subsequent studies based on the existing cohort. In the initial design of the cohort study, we did not consider the outbreak of new coronaviruses and the importance of hematologic markers in antiviral therapy, so we will continue to collect outcomes after treatment of infections in the same populations in the follow-up, as well as relevant follow-up data.

This study mainly reflects the function of human immunity through peripheral blood cells. In this way, the reaction to the effect of night shifts on the body’s immune function is more comprehensive and intuitive. At the same time, we also found that further changes in these indicators may be due to differences in age or title (age-related) composition between subgroups. In this and some other studies, age was strongly associated with night shift scheduling, and senior nurses tended to schedule no or fewer night shifts, contributing to the study error, as several studies have concluded that immunity is strongly associated with age^36,39,40. In our study, after propensity matching based on age, the ELR difference disappeared, suggesting, on the one hand, that night shift scheduling is related to ELR and WBC and, on the other hand, that age has a more significant effect on ELR and that this difference is reflected in our analysis of the stratification of titles in the group of nurses, where the impact of night shifts on the level of ELR was positive for junior nurses and harmful for middle- and senior-level nurses. It also suggests that younger nurses may tolerate the impairments associated with night shifts and that behavioral interventions for night shifts in younger groups of nurses are feasible. However, the benefits need not be overstated. In contrast, night shift behavioral interventions for the intermediate and senior (middle-aged and elderly) nurse population are highly desirable because of their more severe immune impairments and increased risk of disease.

We have the following limitations. First, our study can only show the association between sleep and immune cell changes. Still, it cannot determine the causal relationship, and deciding on the time sequence between sleep and immune cell changes is challenging. Second, many other factors that have not been considered (confounding factors) may affect the results. In the study, it is difficult to completely rule out the influence of all confounding factors, such as diet, family environment, etc. Third, the individuals involved in our study are nurses, which lack representativeness for other groups and may not be easy to generalize to other populations or environments. At the same time, the individuals we studied all met the inclusion and exclusion criteria, and the data that did not meet the requirements were ignored. A single result may also be accidental.