Experimental investigation in rats to identify the cause of sudden death during bathing in Japan

Abstract

Aim

Approximately 14,000 cases of sudden death during bathing are reported annually in Japan. The cause of the deaths is still unclear. Because Japanese‐style bathing is characterized by whole‐body bathing in high temperature water (41–45°C), we hypothesized that heat exposure causes sudden death as a result of hyperthermia.

Methods

The aim was to clarify the relationship between heat exposure and mortality in an animal experiment. In the experiment, 44 male Wistar rats weighing 180–255 g were anesthetized with sodium pentobarbital, and their body temperature was measured with a thermometer inserted into the esophagus. The animals were immersed up to their shoulder in 40–45°C water until respiratory arrest occurred or until they awoke from the anesthesia.

Results

All animals immersed in water heated to ≥41°C died, whereas all animals immersed in 40°C water awakened from the anesthetic. The mean interval between the start of immersion and respiratory arrest in 41°C, 42°C, 43°C, 44°C, and 45°C water was 105 min, 54 min, 34 min, 27 min, and 22 min, respectively. Regression analysis revealed a significant relationship between the interval and water temperature (r² = 0.96, P < 0.01), and a Cox proportional hazard analysis revealed that water temperature was a significant predictor (adjusted odds ratio, 23.9; 95% confidence interval, 7.4–77.2). The body temperature of the animals that died during hot water immersion was above 41°C (median, 42.9°C).

Conclusions

Water temperature was correlated with the interval between the start of immersion and respiratory arrest. The results suggest that hyperthermia contributes to the cause of sudden death during hot water bathing.

Introduction

There are 14,000 deaths in Japan annually that occur during bathing,1, 2 and they account for more than 15% of all out‐of‐hospital sudden deaths.3 According to the death certificates and results of forensic autopsies, the reported causes of death include heart attack, stroke, or drowning of unknown cause.4, 5 Consequently, many researchers, including forensic pathologists, have attributed the major cause of the sudden deaths to cardiovascular events.4, 6, 7, 8

However, their view is inconsistent with observations by other researchers including Japanese emergency physicians.2 If myocardial ischemia or stroke caused the sudden deaths while bathing, they would encounter many survivors of myocardial ischemia or stroke while bathing, whereas they actually frequently encounter patients who developed a consciousness disturbance in the bathtub. When such patients are helped out of the bathtub by paramedics, they gradually recover consciousness without any treatment, and they have no evidence of cardiovascular disease.2 Hence, the actual cause of the sudden death during bathing is still unclear.

The Japanese bathing style is characterized by immersion in hot water. Many Japanese immerse themselves into deep hot water at temperatures above 41°C. Japanese bathtubs are generally deeper than the bathtubs in other countries, and people sit in hot water up to shoulder level. They also like to take long baths. Thus, heat exposure may cause hyperthermia and heat stroke during Japanese‐style bathing. We therefore hypothesized that heat exposure during bathing causes sudden death as a result of hyperthermia and heat stroke, but as it is impossible to test this hypothesis by clinical investigation, the present study was designed to clarify the relationship between heat exposure and mortality during bathing using animal experiments.

Methods

The protocol of this study was approved by the Institutional Animal Care and Use Committee of Keio University (Tokyo, Japan). Experiments were carried out on 44 male Wistar rats (Sankyo Labo Service Co., Ltd, Tokyo, Japan) weighing 180–255 g (mean ± SD, 207 ± 20 g). Their environment was controlled at 20–24°C with a 12:12‐h light : dark cycle. The animals were given free access to food and water.

The rats were anesthetized with sodium pentobarbital (50 mg/kg, i.p.), and their body temperature was measured by inserting a thermometer (Physitemp Instruments, Inc., Clifton, NJ, USA) midway down the esophagus. The temperature data were continuously stored on a personal computer with the PowerLab system (ADInstruments Pty Ltd, Castle Hill, NSW, Australia). Animals were then immersed in 40–45°C water in a bath incubator (Fig. 1) until respiratory arrest occurred, and the interval between the start of immersion and respiratory arrest was measured as survival time. If an animal woke up while in the bath, the water immersion was discontinued.

Figure 1.

Schematic diagram of the experimental setup. Animals were immersed in the water‐bath incubator. Water temperatures of 40–45°C were controlled automatically. A thermometer was inserted into the esophagus to measure the animals' core temperature. The data were continuously stored in a personal computer.

The statistical analysis, including calculation of mean values and their 95% confidence interval, was carried out using SPSS 21 J software (SPSS Japan Co., Ltd, Tokyo, Japan). The relationship between water temperature and survival time was assessed by a regression analysis. The mortality data were analyzed by the Kaplan–Meier method. To adjust for the effect of body weight, a Cox proportional hazards model was used to calculate the impact of heat exposure on mortality during hot water immersion.

Results

As shown in the Table 1, all animals immersed in water at temperatures of 41°C or higher died while in the water, whereas all animals immersed in 40°C water awakened during the bath, and the experiment was discontinued. The body temperature of all animals that experienced respiratory arrest was above 41°C (median, 42.9°C; range, 41.4–44.5°C). The Kaplan–Meier curve and the regression analysis showed a close correlation between survival time and water temperature (Figs. 2, 3). The Cox proportional hazards analysis revealed water temperature to be a significant independent predictor of mortality during bathing (adjusted odds ratio, 23.9; 95% confidence interval, 7.4–77.2).

Table 1.

Effects of heat exposure in rats immersed on hot water while anesthetized

	Temperature of bath water
	40°C	41°C	42°C	43°C	44°C	45°C
	n = 8	n = 13	n = 6	n = 6	n = 6	n = 5
Mean body weight, g (95% CI)	198 (185–211)	220 (208–233)	201 (189–213)	196 (184–207)	200 (187–212)	215 (175–255)
Number of deaths during heat exposure (%)	0 (0)	13 (100)	6 (100)	6 (100)	6 (100)	5 (100)
Mean body temperature at death, °C (95% CI)	n.a.	42.2 (42.0–42.5)	42.8 (42.3–43.3)	43.5 (42.9–44.0)	43.5 (43.0–44.1)	44.0 (43.3–44.6)
Mean survival time, min (95% CI)	39 (20–59)a	105 (94–116)	54 (45–63)	34 (32–36)	27 (22–31)	22 (19–24)

^aInterval between the start and discontinuation of immersion because the animal awoke from the anesthesia. CI, confidence interval; n.a., not applicable.

Figure 2.

Comparison of mortality. Survival curves were drawn by the Kaplan–Meier method. The solid black line indicates the cumulative survival of the animals immersed in 40°C water. Open circles indicate discontinuance of water immersion because the animal awoke from the anesthesia. The black dotted line, dashed line, and dotted and dashed line represent animals exposed to 41°C water, 42°C water, and 43°C water, respectively. The solid gray line and dotted gray line represent animals exposed to 44°C water and 45°C water, respectively.

Figure 3.

Relationship between survival time and water temperature. Regression analysis revealed that the survival time depended on the water temperature (r² = 0.96, P < 0.0001). The regression curve is described by the equation Y = 100/X + 40, where Y represents water temperature and X represents survival time.

Discussion

The results of this study indicated that deaths during bathing are induced by immersion in hot water. As the survival time was closely correlated with water temperature, and the body temperature of all animals that died was over 42°C, hyperthermia may contribute to deaths during bathing.

Approximately 14,000 sudden deaths during bathing are reported annually in Japan,1, 2, 3, 4, 5, 8 far more than in other developed country, and the cause of the deaths is thought to be related to the traditional Japanese style of bathing,5, 6, 7 which consists of immersion in high temperature water (41–45°C). In public spas, some Japanese are fond of bathing in water up to 47°C.9 In the present study the animals were exposed to heat by a hot water immersion protocol simulating the traditional Japanese style of bathing.9

According to the death certificates, deaths during bathing have been attributed to cardiac events, stroke, or drowning of unknown cause.4, 5, 8 However, in more than half of the forensic autopsy reports, the victims did not show any pathological evidence of the cause of sudden death.5 Although most researchers have suspected cardiovascular events and investigated hemodynamic parameters,6, 7 they have found no evidence supportive of a cardiac etiology as the cause of sudden death during bathing. Surprisingly, their observations have never mentioned hyperthermia.

This is the first study to suggest that hyperthermia due to hot water immersion is related to the etiology of sudden death during Japanese‐style bathing. None of the animals exposed to water temperatures of 41°C or higher awakened from the anesthesia, whereas the animals exposed to a water temperature of 40°C awakened during the bath. The results suggested that water temperature affected the animal's consciousness, because seven control rats given the same dose of anesthetic woke up within 60 min of administration (data not shown). Moreover, the animals' body temperature at the time of respiratory arrest was over 42°C, with higher body temperatures than the water temperature (Table 1). It suggested that lethal body temperature is set above 42°C. When animals are immersed in hot water, thermal conduction from water and thermal production from metabolism may make the body temperature increase to above the water temperature, provided that the water temperature is below 42°C. This phenomenon may indicate the breakdown of thermoregulatory mechanisms in the animals. These findings, therefore, suggested that heat stroke contributed to their death.

The plasma opioid concentration has been reported to rise during Japanese‐style bathing,9 and the rise has been thought to cause alteration of consciousness and hyperthermia during bathing. It has been also reported that the hemodynamic alterations in heat stroke are caused by endogenous opioids.10 Opioids have been proposed to play a role in hyperthermia‐induced vasoconstriction and the pathogenesis of shock, and they have been found to lower the temperature sensitivity of warm‐sensitive hypothalamic neurons.11 These findings suggested that opioid‐mediated hyperthermic effects may be related to the induction of heat stroke and death while bathing.

Consciousness disturbances induced by hyperthermia may contribute to the occurrence of sudden death during bathing. If a consciousness disturbance develops while bathing, it could make it impossible to get out of the bathtub. Hyperthermia may become increasingly severe and death may ensue. If hyperthermia is a major etiology of sudden death, it would be difficult to identify it based on the pathological findings.

There were several limitations in this study. First, there are differences between humans and rodents in terms of their physiological mechanisms of thermal control. Second, no hemodynamic parameters, including heart rate, arterial blood pressure, or cardiac output, were measured in this study. Finally, blood and pathological examinations were not carried out. Although the results of this study do not make it possible to reject the hypothesis that the deaths were caused by cardiovascular events, the results suggest that hyperthermia plays a key role in sudden deaths during bathing.

In conclusion, animals were killed by hot water immersion, and the water temperature was correlated with survival time, suggesting that hyperthermia contributes to cause sudden death during bathing in Japan.

Conflict of Interest

None.