Supplementary Information

Effects of cold and hot temperature on dehydration: A mechanism of cardiovascular burden

Authors:Youn-Hee Lim, PhD1,2*, Min-Seon Park, MD3*, Yoonhee Kim, PhD4, Ho Kim, PhD4,5,, Yun-Chul Hong, MD, PhD1,6§

*These authors contributed equally to this work.

1Institute of Environmental Medicine, Seoul National University Medical Research Center, Seoul, Korea

2Environmental Health Center, Seoul National University College of Medicine, Seoul, Korea

3Department of Family Medicine, Seoul National University Hospital, Seoul, Korea

4Department of Biostatistics and Epidemiology, School of Public Health, Seoul National University, Seoul, Korea

5Institute of Health and Environment, Seoul National University, Seoul, Korea

6Institute of Medical Research, Seoul National University Bundang Hospital, Seongnam, Republic of Korea

§Corresponding author: Y-C Hong, Institute of Environmental Medicine, Seoul National University Medical Research Center, 103 Daehakro, Jongro-gu, Seoul 110-799, Republic of Korea. Telephone: +82-2-740-8394; Fax: +82-2-747-4830; E-mail:

1

Tables of Contents

Supplementary Tables

Supplementary Table S1. Monthly average of four markers (Mean (P-valuea))

Supplementary Table S2. Effects of daily mean temperature on dehydration indices per 1°C decrease/ increase at below or above the flexion temperature.

Supplementary Table S3. Amount of variance (%) explained by apparent temperature (AT): Using individual patients’ dehydration makers

Supplementary Table S4. Amount of variance (%) explained by apparent temperature (AT): Using daily average of dehydration markers

Supplementary Figures

Supplementary Figure S1. Monthly patterns of four markers and apparent temperature (AT)

Supplementary Figure S2. Estimated effects per 1°C decrease/increase of mean temperature below /above temperature flexion points (stratified by hypertension group).

Supplementary Figure S3. Plots of changes in dehydration markers according to daily mean temperature change.

Supplementary Table S1. Monthly average of four markers (Mean (P-valuea))

BUN/Creatinine / Urine specific gravityb / Plasma Tonicity (mOsm/l) / Haematocrit (%)
January / 16.11 / 1022.12 / 42.55 / 297.05
February / 16.30 (0.1079) / 1022.18 (0.6765) / 42.45 (0.3116) / 296.11 (<0.0001)
March / 16.33 (0.0514) / 1022.15 (0.8351) / 42.19 (0.0002) / 296.52 (<0.0001)
April / 16.28 (0.1404) / 1021.45 (<0.0001) / 42.13 (<0.0001) / 297.01 (0.7500)
May / 16.01 (0.3877) / 1020.92 (<0.0001) / 42.06 (<0.0001) / 296.26 (<0.0001)
June / 15.94 (0.1342) / 1019.94 (<0.0001) / 42.09 (<0.0001) / 296.19 (<0.0001)
July / 16.23 (0.2725) / 1019.75 (<0.0001) / 41.69 (<0.0001) / 296.20 (<0.0001)
August / 15.63 (<0.0001) / 1020.17 (<0.0001) / 41.69 (<0.0001) / 296.09 (<0.0001)
September / 15.61 (<0.0001) / 1020.18 (<0.0001) / 41.82 (<0.0001) / 296.01 (<0.0001)
October / 15.60 (<0.0001) / 1020.80 (<0.0001) / 42.07 (<0.0001) / 296.96 (0.4323)
November / 16.02 (0.4486) / 1020.93 (<0.0001) / 42.45 (0.3171) / 296.81 (0.0402)
December / 16.27 (0.1610) / 1021.51 (<0.0001) / 42.24 (0.0009) / 296.96 (0.4411)

aP-value for Student’s t-test for month difference (Reference: January)

bUrine specific gravity was multiplied by 1,000.

Supplementary FigureS1. Monthly patterns of four markers and apparent temperature (AT)

BUN/CR: BUN/Creatinine ratio; Usg: Urine specific gravity; Tonicity: Plasma tonicity; HCT: Haematocrit; AT: apparent temperature

1

Supplementary Figure S2. Estimated effects per 1°C decrease/increase of mean temperature below /above temperature flexion points (stratified by hypertension group).

Flexion temperature points and lag days were 25.0°C (lag 0–5), 22.8°C (lag 0), and 26.8°C (lag 0–5) for (a) urine specific gravity, (b) plasma tonicity, and (c) haematocrit, respectively.

1

Supplementary Figure S3.Plots of changes in dehydration markersaccording todaily mean temperature change.

(A), (B), (C), and (D) are Bun/Creatinine ratio, urine specific gravity, plasma tonicity, and haematocrit, respectively.All models were adjustedfor sex, age, body mass index, alcohol consumption, household income, smoking status, haemoglobin (for haematocrit only), daily mean relative humidity, season, and year.

.

Supplementary Table S2.Effects of daily mean temperature on dehydration indices per 1°C decrease/ increase at below or above the flexion temperature.

Dehydration
Index / Temperature (°C, Lag days) / Below flexion / Above flexion
Betaa (SE) / P-value / Betaa (SE) / P-value
Bun/Creatinine ratio / 22.4 (lag 0-1) / 0.036 (0.005) / <0.0001 / 0.083 (0.021) / 0.0001
Urine specific gravityb / 24.0 (lag 0-5) / 0.073 (0.007) / <0.0001 / 0.292 (0.041) / <0.0001
Plasma tonicity / 22.2 (lag 0) / 0.045 (0.005) / <0.0001 / 0.169 (0.022) / <0.0001
Haematocrit / 24.8 (lag0-5) / 0.009 (0.001) / <0.0001 / 0.038 (0.009) / <0.0001

aAll models were adjustedfor sex, age, body mass index, alcohol consumption, household income, smoking status, haemoglobin (for haematocrit only), daily mean relative humidity, season, and year.

bUrine specific gravity was multiplied by 1,000.

1

Supplementary Table S3. Amount of variance (%) explained by apparent temperature (AT): Using individual patients’ dehydration makers

Model / Hydration markers
BUN/CR ratio / Urine specific gravity / Plasma tonicity / Haematocrit
All dataa / 0.3 / 2.2 / 0.3 / 0.3
Cold tails (< flexion point)a / 0.3 / 1.9 / 0.3 / 0.2
Warm tails (> flexion point)a / 0.6 / 0.4 / 0.3 / 0.1

aGeneralized linear modeling (GLM) with natural cubic splines of AT in 5 degrees of freedom

Supplementary Table S4. Amount of variance (%) explained by apparent temperature (AT): Using daily average of dehydration markers

Model / Hydration markers
BUN/CR ratio / Urine specific gravity / Plasma tonicity / Haematocrit
All dataa / 1.4 / 5.8 / 0.5 / 1.5
Cold tails (< flexion point)a / 1.4 / 5.8 / 0.5 / 1.5
Warm tails (> flexion point)a / 2.7 / 3.0 / 0.7 / 1.4

aGeneralized linear modeling (GLM) with natural cubic splines of AT in 5 degrees of freedom.

1