Abstract
Aims: Bathing-related ischemic stroke (BIS) is sometimes fatal. However, its mechanisms and risk factors remain unclear. We aimed to identify the incidence of stroke subtypes in BIS, and clarify the impact of cerebral small vessel disease (CSVD) on BIS.
Methods: Consecutive patients with ischemic stroke between October 2012 and February 2022 were retrospectively screened. The inclusion criteria were: 1) onset-to-door time within 7 days; and 2) availability of the results of MRI evaluation of CSVD markers during hospitalization. BIS was defined as an ischemic stroke that occurred while or shortly after bathing. We investigated the incidence of the stroke subtype and the correlation between CSVD markers and BIS.
Results: 1,753 ischemic stroke patients (1,241 [71%] male, median age 69 years) were included. 57 patients (3%) were included in the BIS group. A higher frequency of large artery atherosclerosis (LAA) (prevalence ratio [PR] 2.069, 95% confidence interval [CI] 1.089 to 3.931, p=0.026) and lower frequency of cardio-embolism (CES) (PR 0.362, 95% CI 0.132 to 0.991, p=0.048) in BIS cases were identified. Moreover, lower periventricular hyperintensity (PVH) Fazekas grade (PR 0.671, 95% CI 0.472 to 0.956, p=0.027) and fewer cerebral microbleeds (CMBs) in deep brain region (PR 0.810, 95%CI 0.657 to 0.999, p=0.049) were associated with BIS cases.
Conclusions: The BIS group was more likely to develop LAA and less likely to develop CES. Lower PVH grade and fewer CMBs in deep brain region were associated with the development of BIS.
Introduction
Bathing is an indispensable part of our daily routine. However, bathing imposes considerable cardiovascular strain1, 2), which might induce the development of ischemic stroke. Given the dangerous environments of bathrooms, where it is often slippery and one is usually alone, with the potential risk of drowning, bathing-related ischemic stroke (BIS) can be fatal, even if its symptoms are mild. Indeed, bathing-related deaths account for approximately 8% to 9% of all sudden deaths in Japan, and a certain proportion of these deaths are caused by strokes3, 4). Therefore, elucidating the physiological mechanisms of BIS and establishing a strategy for its prevention is much required.
The correlation between bathing and the development of ischemic stroke has rarely been investigated5, 6). A physiological response to bathing, such as hemodynamic instability in association with autonomic nervous function1) and transient alterations in the fibrinolytic system and platelet function have been suspected as possible mechanisms of BIS5, 6), indicating that vascular atherosclerotic lesions and a certain kinds of embolism might be involved in BIS. However, plausible preexisting internal risk factors for BIS and the precise mechanisms of BIS remain unknown. Further, since the pathophysiological etiology of thrombus formation differs according to the ischemic stroke subtype7), the impact of bathing might be different according to stroke subtype. Elucidation of the preexisting internal risk factors and stroke subtypes in BIS would be helpful in establishing a preventive strategy for BIS. This information would also allow provision of intensive education and alerts for those who are at a high risk for BIS before its onset.
In this study, we focused on cerebral small vessel disease (CSVD). CSVD, which is represented by white matter hyperintensity, cerebral microbleeds (CMBs), enlarged perivascular spaces (EPVS) and old lacunes, is usually asymptomatic, but steadily increases in severity with progressive atherosclerotic risk factors and higher age8). CSVD is reportedly related to various conditions, such as dementia, vascular parkinsonism, and behavioral symptoms9). Additionally, since the presence of CSVD can be conveniently confirmed using magnetic resonance imaging (MRI)10), it can be non-invasively quantified. If CSVD is related to BIS, CSVD might be a good indicator of the risk of BIS, enabling education of persons at a high risk for BIS before its onset. Interestingly, in recent studies, the severity of CSVD, especially white matter hyperintensity, was associated with autonomic nervous function11, 12). Since, in previous studies, the autonomic nervous response to the stimulus of bathing was considered one of the main physiological mechanisms of BIS1, 6), it suggests that CSVD might underlie the pathophysiological mechanism of BIS. Hence, we hypothesized that the severity of CSVD might be related to the development of BIS and might influence differences in the incidence of stroke subtypes in BIS.
Aim
We aimed to elucidate the incidence of stroke subtypes in BIS, along with evaluation of the correlation between CSVD and BIS in an acute ischemic stroke population.
Methods
Patient Selection and Outcomes
Patients were selected from the Jikei University School of Medicine Stroke Registry (Jikei Stroke Registry), a prospective database of patients with cerebrovascular disease admitted to Jikei University Hospital, which is a tertiary medical center located in the center of the Tokyo metropolitan area. Consecutive patients with ischemic stroke were retrospectively screened between October 2012 and February 2022.
The participants included consecutive patients with ischemic stroke with an onset-to-door time of ≤ 7 days13), in whom CSVD markers could be evaluated on MRI during hospitalization. We defined BIS as an ischemic stroke that occurred while or shortly after bathing6). We reviewed all the patients’ medical records and determined whether the patients fulfilled the above BIS criteria.
We divided patients into BIS and non-BIS groups for comparison of background factors between the two groups, with the aim to elucidate the stroke subtype related to BIS, and determine the relationship between BIS and CSVD markers, as evaluated by MRI.
Patients’ Clinical Characteristics
Baseline data included sex, age, body mass index, modified Rankin scale (mRS) score before symptom onset, onset-to-door time, National Institutes of Health Stroke Scale (NIHSS) score at admission, systolic and diastolic blood pressure at admission, use of antiplatelet drugs and anticoagulant drugs before admission, comorbidities and risk factors for stroke, past history of any stroke/transient ischemic attack, laboratory data at admission (hemoglobin, hematocrit, blood urea nitrogen, creatinine, uric acid, low-density lipoprotein cholesterol, glycosylated hemoglobin level [National Glycohemoglobin Standardization Program], brain natriuretic peptide, and D-dimer), ischemic stroke subtype according to Trial of Org 10172 in Acute Stroke Treatment (TOAST) (large artery atherosclerosis [LAA], cardio-embolism [CES], small vessel occlusion [SVO], stroke of other determined etiology and stroke of undetermined etiology)14) and transient ischemic attack, MRI findings (ischemic lesion involving the left hemisphere, large vessel occlusion, periventricular hyperintensity [PVH], deep and subcortical white matter hyperintensity [DSWMH], CMBs, EPVS in the basal ganglia, and old lacunes), and mRS score 3 months after the onset.
Comorbidities and risk factors for stroke included: 1) hypertension, defined as use of antihypertensive agents before admission or a history of diagnosis of hypertension before admission, but not on any medication, 2) diabetes mellitus, defined as use of oral hypoglycemic agents or insulin, diet therapy, glycosylated hemoglobin level (National Glycohemoglobin Standardization Program) ≥ 6.5%, or a history of diagnosis of diabetes before admission, but not on any medication; 3) dyslipidemia, defined as use of cholesterol-lowering drugs before admission, or serum LDL-cholesterol ＞140 mg/dL or a history of diagnosis of hypercholesterolemia before admission, but not on any medication; 4) atrial fibrillation, defined as a history of diagnosis of atrial fibrillation prior to the ischemic stroke or found on examination during hospitalization; 5) smoking, defined as the current use of cigarettes; and 6) excessive alcohol intake, defined as daily alcohol consumption of more than 60 g15). Information on use of antiplatelet and anticoagulant drugs before admission, smoking status, alcohol consumption and past history of any stroke/transient ischemic attack were obtained from medical interviews with the participant or substitute guardians by physicians.
MRI Findings, Vessel Examinations and Imaging Data Collection/Evaluation of Cerebral Small Vessel Disease Markers
In our hospital, all patients suspected to have had a stroke undergo MRI at admission. Intracranial large vessel occlusion is identified by brain magnetic resonance angiography (MRA), and is defined as occlusion of the following intracranial arteries: internal carotid artery, vertebral artery, basilar artery, anterior cerebral artery (A1 and A2), middle cerebral artery (M1 and M2), and posterior cerebral artery (P1 and P2). In order to diagnose the stroke subtype, we also conducted other vessel examinations for evaluating the extracranial lesions (carotid ultrasound for all patients and computed tomography-angiography and/or digital subtraction angiography for patients who were needed). The CSVD markers evaluated in this study included white matter hyperintensities scale scores (scored separately for PVH and DSWMH)16), the number and location of CMBs, EPVS in the basal ganglia, and the presence of old lacunes. PVH and DSWMH were evaluated using Fazekas grading scale (with separate rating of the deep and periventricular regions on a scale from 0 to 3) 16), defined as hyperintensity on fluid attenuated inversion recovery (FLAIR) sequences17). A CMB lesion was defined as a small void with a hypointense area 2 to 10 mm in diameter, that was visible on susceptibility-weighted imaging (SWI)18). We manually counted the number of CMBs and classified their locations as deep brain region, lobar, and infratentorial19). EPVS were defined as high-linear intensities on T2WI or low-linear intensities in the basal ganglia on a FLAIR sequence, with a rating scale of 0 to 4 17, 20, 21). EPVS in the basal ganglia with a score of ≥ 2 were defined as severe EPVS22). An old lacune of presumed vascular origin was defined as a round or ovoid hypointense area with a hyperintense rim between 3 -15 mm in size on a FLAIR sequence or a hyperintense area on a T2-weighted image (T2WI)17) at the following sites23): deep subcortical white matter including the corona radiata, caudate head, lentiform, posterior limb and genu of the internal capsule, thalamus, and brainstem (including midbrain, pons and medulla).
The MRI platforms used in this study were MAGNETOM Avanto and MAGNETOM Symphony (Siemens, Erlangen, Germany) at 1.5 T. Sequence parameters for diffusion weighted images (DWI) were: repetition time (TR), 2700 ms; echo time (TE), 90 ms; matrix, 128×128; and field of view (FOV), 21 cm. Sequence parameters for T2 were: TR, 4070 ms; echo time (TE), 92 ms; matrix, 232×320; and field of view (FOV), 22 cm. Sequence parameters for FLAIR were: TR, 8000 ms; TE, 103 ms; matrix, 256×256; and FOV, 21 cm. Sequence parameters for SWI were: TR, 49 ms; TE, 40 ms; matrix, 256×230; and FOV, 23 cm. Sequence parameters for MRA were: TR, 23 ms; TE, 7.15 ms; matrix, 320×288; and FOV, 18 cm. We also used a MAGNETOM Skyra system (Siemens, Forchheim, Germany) at 3.0 T. With this machine, the sequence parameters for DWI were: TR, 5000 ms; TE, 65 ms; matrix, 160×160; and FOV, 22 cm. Sequence parameters for T2 imaging were: TR, 4500 ms; TE, 87 ms; matrix, 256×256; and FOV, 22 cm. Sequence parameters for FLAIR imaging were: TR, 9000 ms; TE, 103 ms; matrix, 192×384; and FOV, 21 cm. Sequence parameters for SWI imaging were: TR, 28 ms; TE, 20 ms; matrix, 200×320; and FOV, 24 cm. Sequence parameters for MRA were: TR, 21 ms; TE, 3.69 ms; matrix, 202×320; and FOV, 18 cm.
Statistical Analysis
First, baseline characteristics were compared between BIS and non-BIS groups using the χ2 test, Fisher’s exact test and Mann-Whitney U test, as appropriate.
Next, we performed Poisson regression analysis with a robust variance estimator24) to elucidate the stroke subtype related to BIS. This multivariable analysis was adjusted for stroke subtypes with p＜0.05 on the above univariable analysis, sex, age, and onset-to-door time25, 26).
Third, Poisson regression analysis with a robust variance estimator24) was performed to identify the relationship between CSVD markers and BIS. This multivariable analysis was adjusted for each CSVD marker with a p value of ＜0.05 on the above univariable analysis, sex, and the following pre-specified risk factors for CSVD and ischemic stroke: age, hypertension, diabetes mellitus, dyslipidemia and atrial fibrillation27-29).
To avoid multicollinearity among the independent variables, we checked variable inflation factors (VIF) among the variables. For variables with a high correlation (VIF ≥ 10), only one variable was entered into the multivariable regression modeling procedure. Values of p＜0.05 were considered significant for all results. All statistical analyses were performed using IBM SPSS Statistics version 25 software (IBM-Armonk, New York, NY).
Standard Protocol Approvals and Registrations
The study protocol conformed to the ethical guidelines of the 1975 Declaration of Helsinki. The Regional Ethics and Hospital Management Committee of Jikei University School of Medicine approved the use of data from the Jikei Stroke Registry (approval number 29-195 (8811), 29-196 (8812), and 29-197 (8813)). The board waived the need for patient consent, and instead patients from whom data had been collected were given the opportunity to opt out from this research.
Results
We screened 2,112 consecutive ischemic stroke patients, among whom 1,753 patients were included (1,241 [71%] males, median age 69 years). Of these patients, 57 patients (3%) were included in the BIS group (41 [72%] males, median age 68 years, Fig.1). Comparison of the characteristics of patients in the two groups are shown in Table 1. The BIS group showed lower NIHSS scores at admission (p=0.007). Regarding stroke subtype, the BIS group showed a higher frequency of LAA (p=0.008) and lower frequency of CES (p=0.016, Fig.2). Regarding CSVD markers, the BIS group showed lower PVH Fazekas grade (p=0.014), a smaller number of total CMBs (p=0.039), and a smaller number of CMBs in deep brain region (p=0.020).
Table 1.Clinical characteristics of the total cohort of patients with acute ischemic stroke and following their subdivision into with and without bathing-related ischemic stroke groups
| Variables |
Total cohort
(n= 1,753)
|
BIS group
(n= 57)
|
Non-BIS group
(n= 1,696)
|
p Value
|
| Males |
1,241 (71) |
41 (72) |
1,200 (71) |
0.848 |
| Age (years) |
69 (58-79) |
68 (55-77) |
69 (58-79) |
0.381 |
| Body mass index (kg/m2)
|
23.4 (21.1-25.8) |
23.2 (20.9-24.6) |
23.5 (21.1-25.9) |
0.140 |
| mRS score before onset |
0 (0-0) |
0 (0-0) |
0 (0-0) |
0.260 |
| Onset-to-door time (hours) |
8.6 (1.8-28) |
6.9 (2.0-21) |
8.7 (1.8-28) |
0.709 |
| NIHSS score at admission |
2 (1-5) |
1 (0-3) |
2 (1-5) |
0.007 |
| Systolic blood pressure at admission (mmHg) |
160 (140-182) |
158 (139-176) |
160 (140-182) |
0.733 |
| Diastolic blood pressure at admission (mmHg) |
88 (76-101) |
88 (73-102) |
88 (76-101) |
0.503 |
| Use of antiplatelet drugs before admission |
414 (24) |
15 (26) |
399 (24) |
0.627 |
| Use of anticoagulants before admission |
197 (11) |
2 (4) |
195 (11) |
0.060 |
| Risk factors for stroke |
| Hypertension |
1,193 (68) |
39 (68) |
1,154 (68) |
0.957 |
| Diabetes mellitus |
504 (29) |
18 (32) |
486 (29) |
0.634 |
| Dyslipidemia |
1,017 (58) |
33 (58) |
984 (58) |
0.981 |
| Atrial fibrillation |
364 (21) |
4 (7) |
360 (21) |
0.009 |
| Current smoking |
468 (27) |
17 (30) |
451 (27) |
0.556 |
| Excessive alcohol intake |
87 (5) |
1 (2) |
86 (5) |
0.218 |
| History of previous stroke/transient ischemic attack |
414 (24) |
15 (26) |
399 (24) |
0.626 |
| Laboratory data |
| Hemoglobin (g/dL) |
14.1 (12.6-15.3) |
14.2 (12.8-15.1) |
14.1 (12.6-15.3) |
0.836 |
| Hematocrit (%) |
41.9 (37.9-45.0) |
42.0 (38.4-44.5) |
41.9 (37.9-45.1) |
0.627 |
| Blood urea nitrogen (mg/dL) |
16 (13-20) |
15 (13-19) |
16 (13-20) |
0.342 |
| Creatinine (mg/dL) |
0.83 (0.69-1.0) |
0.82 (0.64-0.95) |
0.83 (0.69-1.0) |
0.204 |
| Uric Acid (mg/dL) |
5.6 (4.7-6.7) |
5.3 (4.4-6.3) |
5.6 (4.7-6.7) |
0.268 |
| Low density lipoprotein cholesterol (mg/dL) |
118 (95-143) |
112 (87-136) |
119 (95-143) |
0.074 |
| Glycosylated hemoglobin (%) |
5.8 (5.5-6.3) |
5.8 (5.5-6.4) |
5.8 (5.5-6.3) |
0.770 |
| Brain natriuretic peptide (pg/mL) |
35 (13-100) |
32 (9-55) |
35 (13-102) |
0.100 |
| D-dimer (μg/mL) |
0.80 (0.60-1.5) |
0.70 (0.50-1.7) |
0.80 (0.60-1.5) |
0.188 |
| Stroke subtypes / Transient ischemic attack |
| Large artery atherosclerosis |
183 (10) |
12 (21) |
171 (10) |
0.008 |
| Cardio-embolism |
341 (19) |
4 (7) |
337 (20) |
0.016 |
| Small vessel occlusion |
238 (14) |
3 (5) |
235 (14) |
0.062 |
| Stroke of other determined etiology |
263 (15) |
11 (19) |
252 (15) |
0.356 |
| Stroke of undetermined etiology |
516 (29) |
17 (30) |
499 (29) |
0.948 |
| Transient ischemic attack |
212 (12) |
10 (18) |
202 (12) |
0.199 |
| MRI findings |
| Ischemic lesion involving left hemisphere |
722 (41) |
19 (33) |
703 (41) |
0.221 |
| Intracranial large vessel occlusion |
396 (23) |
8 (14) |
388 (23) |
0.115 |
| Cerebral small vessel disease markers |
| PVH Fazekas |
1 (0-2) |
1 (0-2) |
1 (0-2) |
0.014 |
| DSWMH Fazekas |
1 (1-2) |
1 (0-2) |
1 (1-2) |
0.078 |
| otal number of CMBs |
1 (1-3) |
0 (0-1) |
1 (0-4) |
0.039 |
| Number of CMBs in deep brain region |
0 (0-1) |
0 (0-0) |
0 (0-1) |
0.020 |
| Number of CMBs in lobar |
0 (0-1) |
0 (0-1) |
0 (0-1) |
0.140 |
| Number of CMBs in infratentorial |
0 (0-0) |
0 (0-0) |
0 (0-0) |
0.220 |
| Severe EPVS in basal ganglia |
917 (52) |
29 (51) |
888 (52) |
0.826 |
| Old lacunes |
1,005 (57) |
28 (49) |
977 (58) |
0.203 |
| mRS score at 3 months after onset |
1 (0-2) |
1 (0-2) |
1 (0-3) |
0.257 |
Data are presented as medians (interquartile range) or numbers (%). BIS, bathing-related ischemic stroke; CMBs, cerebral microbleeds; DSWMH, deep and subcortical white matter hyperintensity; EPVS, enlarged perivascular spaces; MRI, magnetic resonance imaging; mRS, modified Rankin Scale; NIHSS, National Institutes of Health Stroke Scale; PVH, periventricular hyperintensities.
In Poisson regression analysis with a robust variance estimator for stroke subtype and BIS, a higher frequency of LAA (prevalence ratio [PR] 2.069, 95% confidence interval [CI] 1.089 to 3.931, p=0.026) and lower frequency of CES (PR 0.362, 95%CI 0.132 to 0.991, p=0.048) were associated with BIS (Table 2). In Poisson regression analysis with a robust variance estimator for CSVD markers and BIS, a lower PVH Fazekas grade (PR 0.671, 95%CI 0.472 to 0.956, p=0.027) and smaller number of CMBs in deep brain region (PR 0.810, 95%CI 0.657 to 0.999, p=0.049) were associated with BIS (Table 3).
Table 2.Poisson regression analysis with a robust variance estimator: Relationship between stroke subtype and bathing-related ischemic stroke
|
Crude |
Multivariable |
| PR |
95% CI |
p value
|
PR |
95% CI |
p value
|
| Males |
1.057 |
0.599 to 1.867 |
0.848 |
0.994 |
0.553 to 1.787 |
0.984 |
| Age/10 (years) |
0.919 |
0.772 to 1.095 |
0.345 |
0.938 |
0.778 to 1.131 |
0.501 |
| Large artery atherosclerosis |
2.288 |
1.233 to 4.245 |
0.009 |
2.069 |
1.089 to 3.931 |
0.026 |
| Cardioembolism |
0.313 |
0.114 to 0.858 |
0.024 |
0.362 |
0.132 to 0.991 |
0.048 |
| Onset-to-door time (hours) |
0.996 |
0.987 to 1.005 |
0.387 |
0.995 |
0.986 to 1.004 |
0.287 |
CI, confidence interval; PR, prevalence ratio.
Table 3.Poisson regression analysis with a robust variance estimator: Relationship between cerebral small vessel disease markers and bathing-related ischemic stroke
|
Crude |
Model including PVH Fazekas |
Model including total number of CMBs |
Model including number of CMBs in deep brain region |
| PR |
95% CI |
p value
|
PR |
95% CI |
p value
|
PR |
95% CI |
p value
|
PR |
95% CI |
p value
|
| Males |
1.057 |
0.599 to 1.867 |
0.848 |
0.999 |
0.548 to 1.821 |
0.998 |
0.998 |
0.548 to 1.819 |
0.995 |
1.005 |
0.552 to 1.831 |
0.987 |
| Age/10 (years) |
0.919 |
0.772 to 1.095 |
0.345 |
1.117 |
0.888 to 1.405 |
0.346 |
0.993 |
0.813 to 1.213 |
0.947 |
1.010 |
0.841 to 1.214 |
0.914 |
| Hypertension |
1.015 |
0.586 to 1.759 |
0.957 |
1.129 |
0.634 to 2.011 |
0.679 |
1.116 |
0.626 to 1.992 |
0.709 |
1.179 |
0.661 to 2.103 |
0.578 |
| Diabetes mellitus |
1.143 |
0.660 to 1.979 |
0.633 |
1.108 |
0.639 to 1.919 |
0.715 |
1.100 |
0.636 to 1.903 |
0.732 |
1.088 |
0.630 to 1.880 |
0.763 |
| Dyslipidemia |
0.994 |
0.592 to 1.667 |
0.981 |
0.907 |
0.539 to 1.526 |
0.713 |
0.923 |
0.546 to 1.560 |
0.765 |
0.896 |
0.534 to 1.504 |
0.679 |
| Atrial fibrillation |
0.288 |
0.105 to 0.790 |
0.016 |
0.272 |
0.100 to 0.741 |
0.011 |
0.286 |
0.105 to 0.773 |
0.014 |
0.275 |
0.102 to 0.743 |
0.011 |
| PVH Fazekas |
0.727 |
0.545 to 0.968 |
0.029 |
0.671 |
0.472 to 0.956 |
0.027 |
|
|
| Total number of CMBs |
0.965 |
0.893 to 1.044 |
0.377 |
|
0.965 |
0.893 to 1.043 |
0.369 |
|
| Number of CMBs in deep brain region |
0.816 |
0.657 to 1.013 |
0.066 |
|
|
0.810 |
0.657 to 0.999 |
0.049 |
CI, confidence interval; CMBs, cerebral microbleeds; PR, prevalence ratio; PVH, periventricular hyperintensities.
Discussion
We uncovered the following major findings: 1) with regard to stroke subtype, the BIS group exhibited a higher frequency of LAA and a lower frequency of CES; and 2) lower PVH grade and lower prevalence of CMBs in deep brain region correlated with the development of BIS. Our results should provide a better understanding of the pathophysiological mechanisms of BIS and serve as a key for preventing BIS.
Preserved autonomic nervous function associated with lower PVH grade and fewer CMBs in deep brain region might be a key factor explaining the underlying mechanisms of our findings (Fig.3). Regarding autonomic nervous function and CSVD, an inverse relationship between autonomic nervous function and PVH severity and CMBs in deep brain region has been suggested11, 30, 31). Thus, it could be speculated that patients with lower PVH grade and fewer CMBs in deep brain region might have a preserved autonomic nervous response to the stimulus of bathing compared to those with severe PVH grade and CMBs in deep brain region.
Under the condition of preserved autonomic nervous function mentioned above, the potential mechanism for the higher frequency of LAA in the BIS group could be explained by the following three factors evoked by the stimulus of bathing: 1) increased wall shear stress (WSS) caused by dehydration and an increased heart rate, 2) thrombolytic resistance evoked by elevated plasminogen activator inhibitor-1 (PAI-1), and 3) hemodynamic failure due to dehydration.
Increased WSS caused by the stimulus of bathing likely plays a fundamental role in the development of LAA in BIS. LAA occurs when increased WSS is applied at the site of an existing plaque32, 33), usually the thinnest region of the fibrous cap32, 34), leading to plaque rupture and formation of platelet rich thrombi. This WSS is directly proportional to the product of blood viscosity and the spatial gradient of blood flow velocity32, 35). Regarding blood viscosity, bathing has the potential to increase blood viscosity due to the dehydration resulting from sweating. This effect is enhanced under preserved autonomic function since sweating is mainly controlled by autonomic function36-38). Preserved autonomic nervous function together with fewer PVH and CMBs in deep brain region might also contribute to an increase in blood flow velocity during bathing. Normal autonomic nervous activity in response to the stimulus of bathing has been suggested to cause an increase in heart rate1, 39). Further, dehydration and increased body temperature during bathing would also trigger an increase in heart rate.
Elevated serum PAI-1 levels induced by the stimulus of bathing might facilitate the development of LAA in BIS. PAI-1 is the major physiologic inhibitor of endogenous tissue-type plasminogen activator in plasma, and excess serum PAI-1 promotes the development of atherosclerotic lesions and atherosclerotic thrombosis40). Bathing is known to activate and elevate PAI-1 levels41), and increase thrombolytic resistance42). Additionally, patients with active and preserved autonomic nervous activity are more likely to exhibit higher levels of plasminogen activator inhibitor compared to those with less active autonomic systems43).
Hemodynamic failure due to dehydration induced by bathing might also play an important role in the development of LAA while bathing. Hypoperfusion is one of the main mechanisms of LAA44). Since sweating is mainly controlled by autonomic function, under preserved autonomic function, bathing could easily be a trigger for dehydration36-38).
Contrary to the effect on LAA, under the condition of preserved autonomic nervous function, elevated serum PAI-1 levels with bathing might have the potential to suppress the development of CES while bathing. Cardiac thrombi contain a higher content of fibrin compared to atherosclerotic thrombi which mainly consist of platelets45). Although fibrin itself is the main component of cardiac thrombi, a thrombus with higher fibrin content exhibits greater friction, suggesting that fibrin-rich thrombi are much stickier and are less likely to cause clot migration46-49). Since PAI-1, which is an inhibitor of endogenous tissue-type plasminogen activator, suppresses fibrinolysis of cardiac thrombi40, 42), an increase in serum PAI-1 due to bathing against a background of preserved autonomic nervous function41, 43) might temporarily enhance the adherence of these thrombi to the left atrial/atrial appendage during bathing, which might minimize the incidence of CES as a cause of BIS.
Finally, though the difference was not statistically significant, the BIS group exhibited a lower trend of developing SVO. This trend might be simply influenced by the fact that CSVD and SVO generally share the same pathophysiology. Hence, less CSVD burden should reflect a trend toward less SVO among BIS.
The present study has some limitations and also suggests various directions for future research. First, detailed circumstances related to bathing, such as water temperature, bathroom temperature, duration of bathing, and the method of bathing (i.e., showering or soaking in a bathtub), could not be assessed. Second, we were not able to collect data on physical parameters that could serve as indicators of autonomic nervous function, such as heart rate, blood pressure and body temperature during bathing. Third, to support the suspected mechanism of BIS mentioned above, direct assessment of autonomic nervous function should have been performed. Assessments such as head-up tilt test, analysis of heart rate variability, measurements of neurotransmitter levels should be useful to confirm our results. Further, information on the coagulation and fibrinolytic state during bathing were also unavailable. Hence, we have to admit the pathophysiological mechanism of BIS discussed above still remains within the realm of speculation. In the future, to solve these problems, a large prospective study to clarify the relationship between autonomic nervous function, coagulation and fibrinolytic state, and BIS in relation to CSVD, would be of great interest. Next, the retrospective nature of our study and the fact that it included only east Asian patients from a single hospital center might have affected the results, undermined its internal and external validity and limited the generalizability of the findings. For example, a large number of the stroke subtype ‘stroke of other determined etiology’ or ‘stroke of undetermined etiology’ has been seen in this study. The reason may be because, in our hospital, we strictly evaluate the possible embolic sources and classify stroke subtypes according to the TOAST classification. Even if the patient was suspected as SVO from the initial MRI assessment, we routinely perform a number of tests including examinations for detecting right-to-left shunt and Holter electrocardiographic monitoring etc. to exclude the possibility of embolic stroke as much as possible. As a result, the number of patients classified into ‘stroke of other determined etiology’ or ‘stroke of undetermined etiology’ would naturally increase. The accuracy of TOAST classification should directly affect the reliability of the result. To solve this problem, we need to conduct a large and multi-center study with unified examinations for diagnosing stroke etiology. In addition, the relatively low NIHSS scores at admission in our cohort should also influence the generalizability of our results. The reason could be explained by the following that the age of the patients admitted to our hospital tends to be younger compared to other hospitals because of the geographical location of our hospital
50).
Our hospital is located in the center of the Tokyo metropolitan area where a relatively younger population lives in. Again, we need to conduct a large and multi-center study to overcome these problems.
Conclusion
In summary, participants in the BIS group in this study were more likely to present LAA and less likely to have CES. Additionally, lower PVH grade and fewer CMBs in deep brain region were associated with the development of BIS. Our findings indicate that patients with low-grade CSVD, who are typically considered to be at a low risk for stroke10), might be more susceptible to ischemic stroke related to bathing. In other words, we must not undervalue the stroke risk in patients with low-grade CSVD, due to its potential to cause BIS. Further, our findings might be the first step to explaining the pathophysiological mechanisms of BIS and might contribute to establishing strategies for preventing BIS.
Funding Statement
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
Conflicts of Interests
Hidetaka Mitsumura received grants/research funding from JSPS KAKENHI.
Yasuyuki Iguchi reports personal fees from Bayer Healthcare Co. Ltd., grants and personal fees from Pfizer Japan Inc., grants and personal fees from Nippon Boehringer Ingelheim Co. Ltd., grants and personal fees from Takeda Pharmaceutical Co. Ltd., grants and personal fees from Otsuka Pharmaceutical Co. Ltd., and grants and personal fees from Daiichi Sankyo Co. Ltd. outside the submitted work.
Appendix 1. Author Contributions
Takahiro Ishikawa: conceptualization, data curation, formal analysis, methodology, project administration, software, writing - original draft
Takeo Sato: conceptualization, data curation, formal analysis, methodology, project administration, software, writing - review & editing
Motohiro Okumura: conceptualization, data curation, methodology, project administration
Tatsushi Kokuubu: data curation
Junichiro Takahashi: data curation
Tomomichi Kitagawa: data curation
Maki Tanabe: data curation
Hiroki Takatsu: data curation
Asako Onda: data curation
Tappei Komatsu: data curation
Kenichi Sakuta: data curation
Kenichiro Sakai: data curation
Tadashi Umehara: data curation
Hidetaka Mitsumura: data curation
Yasuyuki Iguchi: conceptualization, data curation, formal analysis, methodology, project administration, writing - review & editing
Appendix 2. Coinvestigator
None.
References
- 1) Nagasawa Y, Komori S, Sato M, Tsuboi Y, Umetani K, Watanabe Y, Tamura K. Effects of Hot Bath Immersion on Autonomic Activity and Hemodynamics. Jpn Circ J, 2001; 65: 587-592
- 2) Ojima S, Ohishi M. Effects of hot spring bathing on cardiac and vascular function. Hypertens Res, 2023; 46: 1705-1706
- 3) Katsuyama M, Higo E, Miyamoto M, Nakamae T, Onitsuka D, Fukumoto A, Yatsushiro M, Hayashi T. Development of prevention strategies against bath-related deaths based on epidemiological surveys of inquest records in Kagoshima Prefecture. Sci Rep, 2023; 13: 2277
- 4) Suzuki H, Hikiji W, Tanifuji T, Abe N, Fukunaga T. Characteristics of Sudden Bath-Related Death Investigated by Medical Examiners in Tokyo, Japan. J Epidemiology, 2015; 25: 126-132
- 5) Kubota K, Tamura K, Take H, Kurabayashi H, Shirakura T. Acute Myocardial Infarction and Cerebral Infarction at Kusatsu-spa. Nippon Ronen Igakkai Zasshi Jpn J Geriatr, 1997; 34: 23-29
- 6) Inamasu J, Nakatsukasa M, Oshima T, Tomiyasu K, Mayanagi K, Imai A. Clinical Characteristics of Stroke Occurring while Bathing. J Stroke Cerebrovasc Dis, 2017; 26: 1462-1466
- 7) Santos-Gallego CG, Bayón J, Badimón JJ. Thrombi of Different Pathologies: Implications for Diagnosis and Treatment. Curr Treat Options Cardiovasc Med, 2010; 12: 274-291
- 8) Das AS, Regenhardt RW, Vernooij MW Blacker D, Charidimou A, Viswanathan A. Asymptomatic Cerebral Small Vessel Disease: Insights from Population-Based Studies. J Stroke, 2019; 21: 121-138
- 9) Markus HS, Leeuw FE de. Cerebral small vessel disease: Recent advances and future directions. Int J Stroke, 2023; 18: 4-14
- 10) Chojdak-Łukasiewicz J, Dziadkowiak E, Zimny A, Paradowski B. Cerebral small vessel disease: A review. Adv Clin Exp Med, 2021; 30: 349-356
- 11) Obara T, Nagai K, Shibata S, Hirasawa A, Koshiba H, Hasegawa H, Ebihara T, Kozaki K. Relationship between the severity of cerebral white matter hyperintensities and sympathetic nervous activity in older adults. Geriatr Gerontol Int, 2018; 18: 569-575
- 12) Tian Y, Yao D, Pan Y, Wang M, Meng X, Zhao X, Liu L, Wang Y, Wang Y. Implication of heart rate variability on cerebral small vessel disease: A potential therapeutic target. CNS Neurosci Ther, 2023; 29: 1379-1391
- 13) Matsuzono K, Mashiko T, Ozawa T, Miura K, Suzuki M, Furuya K, Ozawa M, Anan Y, Shimazaki H, Koide R, Tanaka R, Kameda T, Fujimoto S. Characteristics of aged ischemic stroke patients indicative of cardioembolism. J Thromb Thrombolysis, 2021; 51: 522-529
- 14) Adams HP Jr, Bendixen BH, Kappelle LJ, Biller J, Love BB, Gordon DL, Marsh EE 3rd. Classification of subtype of acute ischemic stroke. Definitions for use in a multicenter clinical trial. TOAST. Trial of Org 10172 in Acute Stroke Treatment. Stroke, 1993; 24: 35-41
- 15) Reynolds K, Lewis B, Nolen JDL, Kinney GL, Sathya B, He J. Alcohol Consumption and Risk of Stroke: A Meta-analysis. Jama, 2003; 289: 579-588
- 16) Fazekas F, Chawluk J, Alavi A, Hurtig HI, Zimmerman RA. MR signal abnormalities at 1.5 T in Alzheimer’s dementia and normal aging. Am J Roentgenol, 1987; 149: 351-356
- 17) Wardlaw JM, Smith EE, Biessels GJ, Cordonnier C, Fazekas F, Frayne R, Lindley RI, O’Brien JT, Barkhof F, Benavente OR, Black SE, Brayne C, Breteler M, Chabriat H, Decarli C, de Leeuw FE, Doubal F, Duering M, Fox NC, Greenberg S, Hachinski V, Kilimann I, Mok V, Oostenbrugge Rv, Pantoni L, Speck O, Stephan BC, Teipel S, Viswanathan A, Werring D, Chen C, Smith C, van Buchem M, Norrving B, Gorelick PB, Dichgans M; STandards for ReportIng Vascular changes on nEuroimaging (STRIVE v1). Neuroimaging standards for research into small vessel disease and its contribution to ageing and neurodegeneration. Lancet Neurol, 2013; 12: 822-838
- 18) Greenberg SM, Vernooij MW, Cordonnier C, Viswanathan A, Al-Shahi Salman R, Warach S, Launer LJ, Van Buchem MA, Breteler MM; Microbleed Study Group. Cerebral microbleeds: a guide to detection and interpretation. Lancet Neurology, 2009; 8: 165-174
- 19) Yakushiji Y, Wilson D, Ambler G Charidimou A, Beiser A, van Buchem MA, DeCarli C, Ding D, Gudnason V, Hara H, Imaizumi T, Kohara K, Kwon HM, Launer LJ, Mok V, Phan T, Preis SR, Romero JR, Seshadri S, Srikanth V, Takashima Y, Tsushima Y, Wang Z, Wolf PA, Xiong Y, Yamaguchi S, Werring DJ. Distribution of cerebral microbleeds in the East and West: Individual participant meta-analysis. Neurology, 2019; 92: e1086-1097
- 20) Hilal S, Tan CS, Adams HHH, Habes M, Mok V, Venketasubramanian N, Hofer E, Ikram MK, Abrigo J, Vernooij MW, Chen C, Hosten N, Volzke H, Grabe HJ, Schmidt R, Ikram MA. Enlarged perivascular spaces and cognition. Neurology, 2018; 91: e832-842
- 21) Doubal FN, MacLullich AMJ, Ferguson KJ, Dennis MS, Wardlaw JM. Enlarged Perivascular Spaces on MRI Are a Feature of Cerebral Small Vessel Disease. Stroke, 2010; 41: 450-454
- 22) Jian X, Xu F, Yang M, Zhang M, Yun W. Correlation between enlarged perivascular space and brain white matter hyperintensities in patients with recent small subcortical infarct. Brain Behav, 2023; 13: e3168
- 23) Sato S, Delcourt C, Heeley E, Arima H, Zhang S, Al-Shahi Salman R, Stapf C, Woo D, Flaherty ML, Vagal A, Levi C, Davies L, Wang J, Robinson T, Lavados PM, Lindley RI, Chalmers J, Anderson CS; INTERACT2 Investigators. Significance of Cerebral Small-Vessel Disease in Acute Intracerebral Hemorrhage. Stroke, 2016; 47: 701-707
- 24) Zou G. A Modified Poisson Regression Approach to Prospective Studies with Binary Data. Am J Epidemiology, 2004; 159: 702-706
- 25) Alberts MJ, Perry A, Dawson DV, Bertels C. Effects of public and professional education on reducing the delay in presentation and referral of stroke patients. Stroke, 1992; 23: 352-356
- 26) Matsuo R, Yamaguchi Y, Matsushita T, Hata J, Kiyuna F, Fukuda K, Wakisaka Y, Kuroda J, Ago T, Kitazono T, Kamouchi M; Fukuoka Stroke Registry Investigators. Association Between Onset-to-Door Time and Clinical Outcomes After Ischemic Stroke. Stroke, 2017; 48: 3049-3056
- 27) Kleindorfer DO, Towfighi A, Chaturvedi S, Cockroft KM, Gutierrez J, Lombardi-Hill D, Kamel H, Kernan WN, Kittner SJ, Leira EC, Lennon O, Meschia JF, Nguyen TN, Pollak PM, Santangeli P, Sharrief AZ, Smith SC Jr, Turan TN, Williams LS. 2021 Guideline for the Prevention of Stroke in Patients With Stroke and Transient Ischemic Attack: A Guideline From the American Heart Association/American Stroke Association. Stroke, 2021; 52: e483-484
- 28) Yakushiji Y. Trends and challenges in clinical research on cerebral small vessel disease, with a particular emphasis on type-1 small vessel disease. Clin Neurol, 2020; 60: 743-751
- 29) Yakushiji Y, Charidimou A, Noguchi T, Nishihara M, Eriguchi M, Nanri Y, Kawaguchi A, Hirotsu T, Werring DJ, Hara H. Total Small Vessel Disease Score in Neurologically Healthy Japanese Adults in the Kashima Scan Study. Intern Med, 2018; 57: 189-196
- 30) Del Brutto OH, Mera RM, Zambrano M, Lama J, Del Brutto VJ, Castillo PR. Poor sleep quality and silent markers of cerebral small vessel disease: a population-based study in community-dwelling older adults (The Atahualpa Project). Sleep Med, 2015; 16: 428-431
- 31) Zuurbier LA, Ikram MA, Luik AI, Hofman A, Van Someren EJ, Vernooij MW, Tiemeier H. Cerebral small vessel disease is related to disturbed 24‐h activity rhythms: a population‐based study. Eur J Neurol, 2015; 22: 1482-1487
- 32) Zhou M, Yu Y, Chen R, Liu X, Hu Y, Ma Z, Gao L, Jian W, Wang L. Wall shear stress and its role in atherosclerosis. Front Cardiovasc Med, 2023; 10: 1083547
- 33) Fukumoto Y, Hiro T, Fujii T, Hashimoto G, Fujimura T, Yamada J, Okamura T, Matsuzaki M. Localized Elevation of Shear Stress Is Related to Coronary Plaque Rupture A 3-Dimensional Intravascular Ultrasound Study With In-Vivo Color Mapping of Shear Stress Distribution. J Am Coll Cardiol, 2008; 51: 645-650
- 34) Akyildiz AC, Speelman L, Brummelen H van, Gutiérrez MA, Virmani R, van der Lugt A, van der Steen AF, Wentzel JJ, Gijsen FJ. Effects of intima stiffness and plaque morphology on peak cap stress. Biomed Eng OnLine, 2011; 10: 25
- 35) Hathcock JJ. Flow Effects on Coagulation and Thrombosis. Arter, Thromb, Vasc Biol, 2006; 26: 1729-1737
- 36) Hu Y, Converse C, Lyons MC, Hsu WH. Neural control of sweat secretion: a review. Br J Dermatol, 2018; 178: 1246-1256
- 37) Shibasaki M, Crandall CG. Mechanisms and controllers of eccrine sweating in humans. Front Biosci, 2010; S2: 685-696
- 38) El-Sharkawy AM, Sahota O, Lobo DN. Acute and chronic effects of hydration status on health. Nutr Rev, 2015; 73 Suppl 2: 97-109
- 39) Laukkanen T, Kunutsor SK, Zaccardi F, Lee E, Willeit P, Khan H, Laukkanen JA. Acute effects of sauna bathing on cardiovascular function. J Hum Hypertens, 2018; 32: 129-138
- 40) Vaughan DE. PAI‐1 and atherothrombosis. J Thromb Haemost, 2005; 3: 1879-1883
- 41) Tamura K, Kubota K, Kurabayashi H, Shirakura T. Effects of hyperthermal stress on the fibrinolytic system. Int J Hyperth, 1996; 12: 31-36
- 42) Braaten JV, Handt S, Jerome WG, Kirkpatrick J, Lewis JC, Hantgan RR. Regulation of Fibrinolysis by Platelet-Released Plasminogen Activator Inhibitor 1: Light Scattering and Ultrastructural Examination of Lysis of a Model Platelet-Fibrin Thrombus. Blood, 1993; 81: 1290-1299
- 43) Weinrauch LA, Gleason RE, Keough J, D’Elia JA. Relationship Between Autonomic Function and Plasma Fibrinogen, Viscosity, and Elements of Fibrinolytic Activity in Diabetic Nephropathy. Am J Hypertens, 1997; 10: 454-461
- 44) Banerjee C, Chimowitz MI. Stroke Caused by Atherosclerosis of the Major Intracranial Arteries. Circ Res, 2017; 120: 502-513
- 45) Heo JH, Nam HS, Kim YD, Choi JK, Kim BM, Kim DJ, Kwon I. Pathophysiologic and Therapeutic Perspectives Based on Thrombus Histology in Stroke. J Stroke, 2020; 22: 64-75
- 46) Gunning GM, McArdle K, Mirza M, Duffy S, Gilvarry M, Brouwer PA. Clot friction variation with fibrin content; implications for resistance to thrombectomy. J NeuroInterventional Surg, 2018; 10: 34-38
- 47) Alkarithi G, Duval C, Shi Y, Macrae FL, Ariëns RAS. Thrombus Structural Composition in Cardiovascular Disease. Arter, Thromb, Vasc Biol, 2021; 41: 2370-2383
- 48) Maegerlein C, Friedrich B, Berndt M, Lucia KE, Schirmer L, Poppert H, Zimmer C, Pelisek J, Boeckh-Behrens T, Kaesmacher J. Impact of histological thrombus composition on preinterventional thrombus migration in patients with acute occlusions of the middle cerebral artery. Interv Neuroradiol, 2018; 24: 70-75
- 49) Ni H, Denis CV, Subbarao S, Degen JL, Sato TN, Hynes RO, Wagner DD. Persistence of platelet thrombus formation in arterioles of mice lacking both von Willebrand factor and fibrinogen. J Clin Investig, 2000; 106: 385-392
- 50) Sato T, Sato S, Yamagami H, Komatsu T, Mizoguchi T, Yoshimoto T, Takagi M, Ihara M, Koga M, Iwata H, Matsushima M, Toyoda K, Iguchi Y. D-dimer level and outcome of minor ischemic stroke with large vessel occlusion. J Neurol Sci, 2020; 413: 116814
