Extreme temperature increases the severity of intracerebral hemorrhage: An analysis based on the cold region of China

1 Introduction

Stroke is one of the most common causes of death and disability. In 2019, there were 6.6 million deaths attributable to stroke worldwide^[1]. In recent years, stroke has become the leading cause of death in China, surpassing cancer and coronary heart disease^[2]. Intracerebral hemorrhage (ICH) accounts for about 26.7% of all stroke cases but causes 50.0% of mortality in China^[3], placing a high burden on the health system.

Environmental factors, including extreme temperature, are considered risk factors for stroke. According to the Global Burden of Disease Study (GBD) 2019, 6.6% of total stroke disability-adjusted life-years (DALYs) were attributable to low or high ambient temperature^[4]. Several reports suggested associations between cold exposure and ICH incidence^[5–6] and mortality^[7–8]. However, very little research has been published on the relationship between high temperature and ICH occurrence. Moreover, few studies have addressed the association between clinicopathologic features of ICH and ambient temperature. Zheng et al. found no evident relationship between ambient temperature, clinical parameters, and hematoma volume in ICH^[9]. However, the sample size in this study is relatively small, and patients are geographically dispersed. A study with a larger sample and stronger regionality is needed.

Heilongjiang province is the northernmost province of China, with a frigid climate and high mortality rate of stroke^[10]. In this study, we analyzed 11 124 admissions of ICH cases in the First Affiliated Hospital of Harbin Medical University from 2015 to 2020. We explored the effects of extreme temperatures on clinicopathologic parameters of ICH patients and the potential correlation with gender and age.

2 Methods

3 Result

A total number of 11 124 ICH patients were admitted to the Stroke Department of the First Affiliated Hospital of Harbin Medical University between 1st January 2015 and 31th December 2020. Dramatic atmospheric temperature changes accompanied by an increase in the number of patients presenting with intracerebral hemorrhage were observed despite that only single-center data were recorded (Fig. S1–S6). Pearson correlation test revealed that the number of patients with ICH was significantly correlated with the temperature change (P < 0.001)^[7]. Table 1 shows the statistical data on the demographic characteristics of the patients included in this study. The majority of the patients were male (66.8%). The mean age of the patients was 58.28 ± 12.76 years, and more than half of the patients had or maintained smoking and drinking habits (50.1% and 50.3%). Most of the patients with ICH included in this study had a history of hypertension (67.9%), and 1 493 patients had a history of ischemic stroke before intracerebral hemorrhage. The mean systolic blood pressure (SBP) of all patients was 169.24 ± 31.13 mmHg, and the median value of hematoma volume was 13.55 mL (4.77–32.36).

The number and proportion of ICH patients and mean outdoor temperature per month are shown in Table 2. January had the lowest average temperature, and patients had the highest systolic blood pressure and the most massive average hematoma volume in the month. There was a significant increase in the number of individuals with ICH onset in the other three seasons compared to summer (June, July, and August), but the degree of midline shift was smaller. We conducted an analysis of meteorological data from 2015 to 2020 and stratification on the data collected from patients with ICH in cold climates and hot climates. At admission, the proportion of male patients in the cold spell group was higher than that in the control group (69.9% vs. 66.7%, P = 0.038), and fewer patients consumed alcohol and smoked (Drinking: 42.1% vs. 50%, P < 0.001; Smoking: 43.2% vs. 49.5%, P < 0.001; Table 3). The occurrence of diabetes mellitus, hypertension, and ischemic stroke was not significantly different between the two groups. In addition, there was no significant statistical difference in vital signs of body temperature, heart rate, and blood pressure between the two groups at admission. However, compared with the control group, the bleeding volume in the cold spell group was significantly increased (14.56 mL, IQR: 5.02–36.03 vs. 2.48 mL, IQR: 0–17.82, P < 0.047), but no significant difference was observed in hematoma location and the degree of midline shift.

Compared with the control group, more patients in the heatwave group continued their drinking and smoking habits (Drinking: 56.9% vs. 50%, P < 0.001; Smoking: 57.5% vs. 49.5%; P < 0.001), and systolic blood pressure at admission was significantly lower than that in the control group (166.76 ± 32.37 vs. 169.36 ± 30.79 g/L, P = 0.019). The imaging data showed no significant difference in hematoma volume and bleeding location between the two groups. However, the probability and degree of midline shift in the heatwave group were significantly higher than those in the control group (P < 0.001).

The results presented above indicate that the change of temperature has a significant influence on the occurrence and severity of ICH. We compared the overall performance of the 18 models to identify the best extreme climate-ICH model for Heilongjiang province. Table 4 presents nine cold spell models and nine heatwave models combining different intensities and frequencies. Apparently, while cold spell and heat wave are defined with a longer duration, the occurrence of extreme weather decreases. In cold spell models, only 21 cold spell days when the temperature dropped to below 3% for more than five days. Also, only 17 heat wave days were recorded in the heatwave model when the temperature stayed above 97% for more than five days. In addition, the AIC values of the two models were relatively smaller in the same group.

Next, we performed a univariate logistic regression analysis on patients' lifestyle, hemorrhage volume, and midline shifts on the generalized extreme weather model (modle 1 and modle 10). The findings indicated that cold spell significantly increased the risk of larger hematoma volume relative to the control group (OR 1.003; 95% CI: 1.000–1.005, P = 0.047). However, there was no significant association in the heatwave group. The degree of midline shift was more severe with the occurrence of heat wave (OR 2.369; 95% CI: 1.852–3.030, P = 0.001). In addition, temperature changes had a noticeable effect on patients’ habits of (Table 5). We performed a univariate logistic regression analysis again based on the optimal model. The results showed that the occurrence of cold spells and heat waves had a significant impact on the severity of ICH, and the risk of midline shift was significantly increased in both groups (OR 1.067; 95% CI: 1.021–1.115, P = 0.004; OR 1.077; 95% CI: 1.030–1.127, P = 0.001).

4 Discussion

In the present retrospective study, we included 11 124 patients with ICH in a regional center hospital in Heilongjiang province over a 6-year period. This study is the first to explore the association between ambient temperature and ICH in the cold region of northeast China. We also innovatively studied the relationship between pathological parameters such as midline shift or hematoma volume of ICH and cold spell or heat wave, which inspired the further exploration of the pathological mechanism of temperature's influence on ICH and provided a theoretical basis for preventive measures of ICH in cold regions.

In the present study, we analyzed the data information of patients with ICH in cold climates and hot climates. Consistent with other studies, our analysis showed that men are more likely to have ICH during cold spells^[18–19], which may correlate with more outdoor activity of men in winter. Interestingly, we found significantly lower smoking and alcohol consumption rates in the cold wave group than in the control group but significantly higher rates in the heat wave group, which has not been previously reported. The findings suggest that tobacco and alcohol might have certain protective effects on cold-induced ICH, which needs to be verified by more rigorous and specially designed studies.

In addition, by selecting appropriate cold spell and heat wave models, we found a significant association between temperature change and the incidence and severity of ICH. Several studies have found that low temperature significantly affects ICH admission in China and in other countries as well. One research conducted in Beijing between 2013 and 2014 showed that extreme cold temperature (first percentile of temperature) is associated with increasing hemorrhagic stroke admissions^[19]. Similar conclusions were reached in the studies conducted in Nanchang^[20], Jinan^[21], and Guangzhou^[22]. The mortality of ICH has also been shown to increase with temperature reduction^[7–8]. However, few studies investigate the effects of temperature on clinical-pathological parameters of ICH. Cold exposure may induce peripheral vasoconstriction and consequently elevates blood pressure and increases cerebral blood flow^[23–24]. Moreover, low temperature inhibits the activation of the coagulation system^[25]. These facts explain why the bleeding volume in the cold spell group was significantly increased. However, an analysis from INTERACT1 indicates no appreciable association of temperature parameters with clinical severity nor with hematoma volume^[9]. The discrepancy could be explained by the difference in sample size and regional distribution of patients in different studies.

Whether heat wave affects ICH remains controversial among researchers. For example, Takumi et al. found a significant increase in hospital admissions for ICH for every 1°C increase in maximum temperature^[26]. Another study in Nanchang, China showed that exposure to extremely high temperatures within three days before the onset of the condition was associated with an increased risk of admission with ICH^[20]. A pooled analysis of four studies reported a statistically significant association between increased mean ambient temperature and ICH incidence, OR 0.97(0.94–1.00)^[27]. In contrast, a study in Shenzhen, China revealed the protective effect of short-term high temperature on the onset of ICH^[28]. In our research, the probability and degree of midline shift in the heatwave group were significantly higher than those in the control group. Considering that there are few absolute high-temperature days in this region, whether the observed phenomenon is merely a random coincidence or a true connection with underlying mechanisms needs to be clarified in future studies.

One of the strengths of the present study is the large sample size. Another advantage is that our data were collected from a large medical center and have high levels of standard and representativeness. Nonetheless, our study also has some limitations. First, this is a single-center study with a single source of data acquisition, and the increase in hospital admissions was perceived as the increase in disease incidence. Second, the present study did not evaluate the effects of air pollution which has been demonstrated to be associated with the incidence of ICH^[29]. Third, the impact of living environment and conditions, such as indoor heating, on cerebral hemorrhage was not included in experimental design. Moreover, it has been suggested that the suburban residents with lower socioeconomic levels are more vulnerable to the effects of heat and cold on stroke hospitalization^[30]. Furthermore, the lag effects of extreme climate were not integrated into the analyses.

5 Conclusion

Our study findings indicate that seasonal cold spells and heat waves are essential factors affecting the severity of cerebral hemorrhage. There is a higher risk of intracerebral hemorrhage and larger hemorrhage volumes during cold spells in Heilongjiang province, whereas shifts in midline structures are more likely to occur during heatwaves. In addition, the severity and duration of extreme climate can amplify the impact. The best cold spell and heatwave models in Heilongjiang province were defined as the daily average temperature below (or above) the 3rd percentile lasting for more than five days. When extreme weather occurs, targeted preventive measures should be actively taken for people with hypertension and a history of stroke.