Skip to main content
Frontiers in Cardiovascular Medicine logoLink to Frontiers in Cardiovascular Medicine
. 2023 Mar 27;10:1084611. doi: 10.3389/fcvm.2023.1084611

A systematic review and meta-analysis of cold exposure and cardiovascular disease outcomes

Jie-Fu Fan 1,, Yu-Chen Xiao 1,, Yi-Fei Feng 1,, Lu-Yu Niu 1, Xing Tan 1, Jia-Cen Sun 1, Yue-Qi Leng 1, Wan-Yang Li 1, Wei-Zhong Wang 1,*, Yang-Kai Wang 1,*
PMCID: PMC10083291  PMID: 37051068

Abstract

Background

Cold exposure has been considered an essential risk factor for the global disease burden, while its role in cardiovascular diseases is still underappreciated. The increase in frequency and duration of extreme cold weather events like cold spells makes it an urgent task to evaluate the effects of ambient cold on different types of cardiovascular disease and to understand the factors contributing to the population's vulnerability.

Methods

In the present systematic review and meta-analysis, we searched PubMed, Scopus, and Cochrane. We included original research that explored the association between cold exposure (low temperature and cold spell) and cardiovascular disease outcomes (mortality and morbidity). We did a random-effects meta-analysis to pool the relative risk (RR) of the association between a 1°C decrease in temperature or cold spells and cardiovascular disease outcomes.

Results

In total, we included 159 studies in the meta-analysis. As a result, every 1°C decrease in temperature increased cardiovascular disease-related mortality by 1.6% (RR 1.016; [95% CI 1.015–1.018]) and morbidity by 1.2% (RR 1.012; [95% CI 1.010–1.014]). The most pronounced effects of low temperatures were observed in the mortality of coronary heart disease (RR 1.015; [95% CI 1.011–1.019]) and the morbidity of aortic aneurysm and dissection (RR 1.026; [95% CI 1.021–1.031]), while the effects were not significant in hypertensive disease outcomes. Notably, we identified climate zone, country income level and age as crucial influential factors in the impact of ambient cold exposure on cardiovascular disease. Moreover, the impact of cold spells on cardiovascular disease outcomes is significant, which increased mortality by 32.4% (RR 1.324; [95% CI 1.2341.421]) and morbidity by 13.8% (RR 1.138; [95% CI 1.015–1.276]).

Conclusion

Cold exposure could be a critical risk factor for cardiovascular diseases, and the cold effect varies between disease types and climate zones.

Systematic Review Registration

https://www.crd.york.ac.uk/PROSPERO, identifier: CRD42022347247.

Keywords: low temperature, cold spell, cardiovascular disease, meta-analysis, climate

1. Introduction

Climate change has a significant impact on human health and has become a global health concern (13). Global disease burden 2019 reported that non-optimal temperatures accounted for 1.01 million deaths in males and 0.946 million in females (1). Despite the long-term warming trends, there is an increase in the number, frequency, and duration of extreme weather events such as cold spells, which makes cold exposure a more significant threat (4, 5). It has been reported that for every 1°C temperature decrease below the reference point, the rate of non-accidental mortality increases by 4% (6). Therefore, it is crucial to clarify the impact of cold exposure on human health outcomes.

Cardiovascular diseases (CVDs) are the leading cause of disease burden, accounting for nearly one-third of total deaths worldwide (1). In many countries, CVD mortality is higher in winter than in summer (7, 8). As reported, sudden exposure to low temperatures could disturb cardiovascular activity (9, 10). Cold exposure induces an increase in blood pressure and changes in blood components, which could induce disease conditions such as hypertension, myocardial infarction (MI), and atherosclerosis (7, 11, 12). This evidence suggests that cold exposure might be an essential risk factor for cardiovascular diseases and increase the health burden. Considering the increasing intensity and frequency of cold surges and cold spells (4), it is vital to demonstrate the impact of cold exposure on cardiovascular diseases.

Previous studies have reported a positive association between cold exposure and cardiovascular mortality and morbidity (6, 11, 13, 14). However, the extent of cold impact on cardiovascular health remains disputable. Specifically, Ren et al. reported a 14.3% increase in cardiovascular mortality followed by every 1°C decrease in temperature (15), while Bai et al. found only a 1.1% increase (16). The wide variation between studies hinders a proper understanding of cold impact. More importantly, the influential factors that cause variations are worth investigating. Previous meta-analyses mainly focused on cold impact on all-cause mortality, in which cardiovascular disease was discussed only as a subgroup. Hence, there is currently no study that systematically analyzes cold impact on different kinds of cardiovascular disease, let alone discusses the influential factors of cold impact such as climate zones. A review that focuses on cold impact on cardiovascular disease is crucial to provide more specific and detailed information on matters such as cold impact on different kinds of cardiovascular diseases, the vulnerabilities of the population, and influential factors.

Therefore, we conducted a wide-ranging search and analysis of the available epidemiological evidence concerning the effects of cold exposure (low temperatures and cold spells) on cardiovascular disease outcomes. We carried out an elaborate stratification on the included literature, examining cold impact on different kinds of diseases and exploring the susceptibility of the population to cardiovascular disease outcomes resulting from cold exposure.

2. Methods

We followed the preferred reporting items for systematic reviews and meta-analysis (PRISMA) guidelines to plan and conduct this review (17) (Supplementary Table S1), and the study protocol was registered with PROSPERO (CRD42022347247).

2.1. Literature search and selection criteria

We searched the databases of PubMed, Scopus, and Cochrane. Keywords such as “temperature”, “weather”, “climate change”, or “cold” were used for exposures. As for health outcomes, we used “cardiovascular disease”, “heart disease”, “vascular disease”, “cerebrovascular disease”, “hypertensive disease”, “myocardial infarction”, “stroke”, “heart failure”, “arrhythmia”, “cardiac arrest”, “rheumatic heart disease”, “thrombotic disease”, “pulmonary heart disease”, and “aortic aneurysm and dissection”. Peer-reviewed studies published in English before February 6, 2023, were identified. Reference lists of all selected articles were independently screened to identify additional studies left out in the initial search. These processes were developed by two investigators (JF and YC), and any differences in investigators’ decisions were discussed. The complete search strategy used for each database is outlined in Supplementary Table S2.

2.2. Eligibility criteria

In the literature search, we included studies that met the following selection criteria: (1) original, peer-reviewed articles with an independent study population; (2) articles that included information on the relationship between cold exposure and cardiovascular-related mortality (death) and morbidity (hospitalization, emergency room visit, ambulance call-out, and out-of-hospital cardiac arrest); and (3) articles categorized as a time-series study or case-crossover study. For this review and meta-analysis, articles were excluded if they reported percentiles for exposure assessment or seasonal effects rather than specific temperatures. Figure 1 presents a flow diagram of the study selection process.

Figure 1.

Figure 1

Flowchart of the assessment of eligible studies.

2.3. Data extraction

An Excel data extraction form was created to record study information on the study period, study population, exposure, outcome, and results on the effects of cold (Table 1). The summary estimates were obtained from the published tables and figures through textual descriptions and Supplementary Material. When information from the figures was imprecise or detailed data seemed available but not provided in the article, we contacted the authors to request further data. When both crude and adjusted estimates were reported, we used the adjusted estimates (18). If multiple studies were using the same data and were conducted by the same research group, we considered the results for the most recent publication. If different research groups conducted the studies, we included all of them in the pooled analysis.

Table 1.

Characteristics of included studies.

ID Author Year Location Study period Study design Exposure Mean value (°C), range Mean value (°C) Season Adjusted for air pollution Mortality/morbidity Outcome (ICD) Ages Climate zones Climate zones Income group
1 Ren et al. 2006 Brisbane, Australia 1996–2001 TS Mean 15.42 (1.2 to 26) 15.42 Annual PM10, O3 Both CVD (I00–I99) All ages Cfa C-subtropical H
2 Ferreira et al. 2019 Five cities, Brazil 1996–2013 TS Mean 20.5 to 26.5 (4.9 to 31.9) 20.5 Annual NA Mortality ACS (I21–I22) All ages Af A-tropical UM
3 Analitis et al. 2018 Nine cities, Europe 2004–2010 TS Tapp 18.4 to 30 (NA to NA) 23 Cold (Oct–Mar) PM10, O3, NO2 Mortality CVD (I00–I99) All ages Csb C-mediterranean H
4 Hashizume et al. 2009 Matlab, Bangladesh 1994–2002 TS Mean NA (NA to NA) NA Annual NA Mortality CVD (I00–I99) All ages Aw A-tropical LM
5 Huang et al. 2014 Changsha, China 2008–2011 TS Cold spell NA (−5.3 to 40.7) NA Annual PM10, NO2, SO2 Mortality CVD (I00–I99) All ages Cfa C-subtropical UM
6 Lin et al. 2013 Four metropolitans, Taiwan 1994–2007 TS Mean 24.2 (8.1 to 33) 24.2 Annual PM10, O3, NO2 Mortality CVD (I00–I99) All ages Cfa C-subtropical H
7 Son et al. 2016 São Paulo, Brazil 1996–2010 TS Mean 20.1 (7.5 to 28.7) 20.1 Annual PM2.5, PM10, O3 Mortality CVD (I00–I99) All ages Cfa C-subtropical UM
8 Xiong et al. 2017 Shanghai, China 2011–2013 TS Mean 17.4 (−2 to 36) 17.4 Annual PM10, NO2, SO2 Mortality CVD (I00–I99) All ages Cfa C-subtropical UM
9 Ikefuti et al. 2018 São Paulo, Brazil 2002–2011 TS Mean 19.5 (8.4 to 27.6) 19.5 Annual PM10, O3, NO2,SO2 Mortality Stroke (I60–I69) All ages Cfa C-subtropical UM
10 Gouveia et al. 2003 São Paulo, Brazil 1991–1994 TS Mean 19.3 (7 to 26.3) 19.3 Annual PM10, O3, NO2, CO, SO2 Mortality CVD (I00–I99) All ages Cfa C-subtropical UM
11 Liu et al. 2020 Hong Kong, China 2007–2015 TS Mean 23.5 (8.4 to 32.4) 23.5 Annual PM2.5, O3, NO2, SO2 Mortality CVD (I00–I99) All ages Cfa C-subtropical H
12 Guo et al. 2011 Tianjin, China 2005–2007 CC Mean 13 (−7 to 29) 13 Annual PM10, NO2, SO2 Mortality CVD (I00–I99) All ages Dwa D-continental UM
13 Zhang et al. 2014 Five cities, China 2004–2008 TS Mean 17.12 (−10.5 to 34.2) 17.12 Annual PM10, NO2 Mortality Stroke (I60–I69) All ages Dwa D-continental UM
14 Romani et al. 2020 Two cities, Spain 2005–2017 TS Min NA (−2.8 to 40.8) NA Annual NA Mortality CVD (I00–I99) All ages Csb C-mediterranean H
15 Dai et al. 2015 Shanghai, China 2006–2011 TS Mean 18 (−2 to 34) 18 Annual PM2.5, PM10, O3, NO2, CO, SO2 Mortality CHD (I20–I25) All ages Cfa C-subtropical UM
16 Silveira et al. 2019 27 cities, Brazil 2000–2015 TS Mean 18.9 to 28.9 (3.8 to 36) 23.9 Annual NA Mortality CVD (I00–I99) All ages Multi Multi UM
17 Guo et al. 2012 Chiang Mai, Thailand 1999–2008 TS Mean 26.2 (13.3 to 33.5) 26.2 Annual PM10, O3 Mortality CVD (I00–I99) All ages Aw A-tropical UM
18 Seposo et al. 2015 Manila, Philippines 2006–2010 TS Mean 28.8 (23.5 to 33.3) 18.8 Annual NA Mortality CVD (I00–I99) All ages Aw A-tropical LM
19 Zhang et al. 2016 Wuhan, China 2003–2010 TS Mean 17.9 (−2.7 to 35.8) 17.9 Annual PM10, NO2, SO2 Mortality CVD (I00–I99) All ages Cfa C-subtropical UM
20 Yang et al. 2012 Guangzhou, China 2003–2007 CC Mean 23 (2.1 to 34.2) 23  Annual PM10, NO2, SO2 Mortality CVD (I00–I99) All ages Cfa C-subtropical UM
21 Yu et al. 2011 Brisbane, Australia 1996–2004 TS Mean 20.1 (9.8 to 31.9) 20.1 Summer PM10, O3, NO2 Mortality CVD (I00–I99) All ages Cfa C-subtropical H
22 Yu et al. 2011 Brisbane, Australia 1996–2004 TS Mean 20.1 (15.4 to 25.2) 20.1 Annual PM10, O3, NO2 Mortality CVD (I00–I99) All ages Cfa C-subtropical H
23 Kwon et al. 2015 South Korea 2004–2012 TS Min 24.19 (10.2 to 32.7) 24.19 Cold (Dec–Feb) PM10, O3, NO2, CO, SO2 Mortality CVD (I00–I99) All ages Multi Multi H
24 Ma et al. 2020 Jiangsu, China 2015–2017 TS Mean 13.9 (−11.5 to 30.6) 13.9 Annual PM2.5, O3, NO2, CO, SO2 Both CVD (I00–I99) All ages Cfa C-subtropical UM
25 Ballester et al. 1997 Valencia, Spain 1991–1993 TS Mean 22 (NA to NA) 22 Cold (Nov–Apr) SO2 Mortality CVD (I00–I99) All ages BSk B-dry H
26 Silveira et al. 2021 Rio de Janeiro, Brazil 2001–2018 CC Mean 24.7 (15.5 to 35) 24.7 Annual NA Mortality CVD (I00–I99) All ages Aw A-tropical UM
27 Zhai et al. 2022 Qingdao, China 2009–2017 TS Mean 14.5 (NA to NA) 14.5 Annual NA Mortality CVD (I00–I99) All ages Cwa C-subtropical UM
28 Ma et al. 2014 17 cities, China 1996–2008 TS Mean 15.3 (−23.7 to 36.4) 15.3 Annual PM10, NO2, SO2 Mortality CVD (I00–I99) All ages Multi Multi UM
29 O'Neill et al. 2005 Two cities, Mexico 1996–1998 TS Tapp 19.9 (−2.7 to 42.1) 19.9 Annual PM10, O3 Mortality CVD (I00–I99) All ages Cwb C-oceanic UM
30 Yi and Chan 2015 Hong Kong, China 2002–2011 TS Mean 23.4 (8.2 to 31.8) 23.4 Annual PM10, NO2, SO2 Mortality CVD (I00–I99) All ages Cfa C-subtropical H
31 Xing et al. 2020 Beijing, China 2006–2011 TS Mean 12.56 (−14.1 to 33) 12.56 Annual PM2.5, O3, NO2, SO2 Mortality CVD (I00–I99) All ages Dwa D-continental UM
32 Sharovsky et al. 2004 São Paulo, Brazil 1996–1998 TS Mean 19.3 (8.8 to 28.3) 19.3 Annual PM10, NO2, CO Mortality ACS (I21–I22) All ages Cfa C-subtropical UM
33 Achebak et al. 2018 47 cities, Spain 1980–2015 TS Mean NA (NA to NA) NA Summer months NA Mortality CVD (I00–I99) All ages Multi Multi H
34 Lin et al. 2020 Taiwan, China 2000–2014 TS Mean 23.3 (9.5 to 31.1) 23.3 Annual PM2.5, PM10, O3, NO2, CO Both CVD (I00–I99) All ages Cfa C-subtropical H
35 Analitis et al. 2007 15 European cities 1990–2000 CC Mean NA (NA to NA) NA Annual NA Mortality CVD (I00–I99) All ages Multi Multi H
36 Denpetkul and Phosri 2021 65 provinces, Thailand 2010–2017 TS Mean 27.48 (6.95 to 36.6) 27.48 Annual NA Mortality CVD (I00–I99) All ages Aw A-tropical UM
37 Guo et al. 2013 Five cities, China 2004–2008 TS Mean 17.12 (−10.5 to 34.2) 17.12 Annual PM10, NO2 Mortality CVD (I00–I99) All ages NA NA UM
38 Chen et al. 2017 Texas, United States 1992–2011 CC Cold spell 20.4 (−6.4 to 34.4) 20.4 Annual NA Mortality CVD (I00–I99) All ages Dfa D-continental H
39 Zeka et al. 2014 Ireland 1984–2007 TS Tapp 5.8 (3.1 to 8.6) 5.8 Cold (Dec–Feb) NA Mortality CVD (I00–I99) 18+ Cfb C-oceanic H
40 Medina-Ramon and Schwartz 2007 50 cities, United States 1989–2000 CC Min 19.8 (18.5 to 32.1) 19.8 Warm (May–Sep) NA Mortality CHD (I20–I25) All ages NA NA H
41 Ha and Kim 2013 Seoul, South Korea 1993–2009 TS Mean 24.4 (NA to NA) 24.4 Warm (Jun–Aug) NA Mortality CVD (I00–I99) All ages Dwa D-continental H
42 Tian et al. 2012 Beijing, China 2000–2011 CC Mean 13.3 (−7.6 to 30.5) 13.3 Annual NA Mortality CHD (I20–I25) All ages Dwa D-continental UM
43 Sharafkhani et al. 2017 Urmia, Iran 2005–2010 TS Mean NA (NA to NA) NA Annual PM10, NO2, SO2 Mortality CVD (I00–I99) All ages Cfa C-subtropical UM
44 Pan et al. 1995 Taiwan, China 1981–1991 TS Mean NA (9 to 32) NA Annual NA Mortality Stroke (I60–I69), CHD (I20–I25) 45+ Cfa C-subtropical H
45 Yang et al. 2015 Shanghai, China 1981–2012 TS Mean 16.9 (−4.8 to 34.6) 16.9 Annual NA Mortality CVD (I00–I99) All ages Cfa C-subtropical UM
46 Breitner et al. 2014 Bavaria, Germany 1990–2006 TS Mean 9.5 (−14.2 to 29.2) 9.5 Annual PM10, O3 Mortality CVD (I00–I99) All ages Dfb D-continental H
47 Rodrigues et al. 2019 Lisbon, Portugal 2000–2013 TS Mean 17.38 (4.1 to 33.3) 17.38 Annual PM10 Mortality Stroke (I60–I69) All ages Csa C-mediterranean H
48 Chen et al. 2014 Six cities, China 2009–2011 TS Mean 4.9–23 (−24.2 to 33) 15 Annual PM10, NO2, SO2 Mortality CHD (I20–I25) All ages NA NA UM
49 Chan et al. 2012 Hong Kong, China 1998–2006 TS Mean 27.6 (19.7 to 31.8) 27.6 Warm (May–Oct) PM10, O3, NO2, SO2 Mortality CVD (I00–I99) All ages Cfa C-subtropical UM
50 Rocklöv et al. 2011 Stockholm, Sweden 1990–2002 TS Tappmin 21.6 (5.8 to 33.5) 21.6 Cold (Oct–Mar) O3, NO2 Mortality CVD (I00–I99) All ages   E-subarctic H
51 Fu et al. 2018 India 2001–2013 CC Mean NA (NA to NA) NA Annual NA Mortality CVD (I00–I99) All ages   C-subtropical UM
52 Goodman et al. 2004 Dublin, Ireland 1980–1996 TS Min 6.5 (−7.9 to 18.4) 6.5 Annual PM10 Mortality CVD (I00–I99) All ages Cfb C-oceanic H
53 Bai et al. 2014 Three cities, Tibet 2008–2012 TS Mean 5.8–9.7 (−12.2 to 22.6) 7.5 Annual NA Mortality CVD (I00–I99) All ages Dw(x) D-continental UM
54 Liu et al. 2011 Beijing, China 2003–2005 TS Mean 22.6 (6.9 to 32.1) 22.6 Cold (Oct–Mar) PM2.5 Mortality CVD (I00–I99) All ages Dwa D-continental UM
55 Alahmad et al. 2020 Kuwait 2010–2016 TS Mean 27.9 (6.86 to 44.65) 27.9 Annual PM10, O3 Mortality CVD (I00–I99) All ages BWh B-dry H
56 Iranpour et al. 2020 Ahvaz, Iran 2014–2018 TS Mean 26.95 (5.8 to 42.4) 26.95 Annual PM2.5, PM10, O3, NO2, CO, SO2 Mortality CVD (I00–I99) All ages BSh B-dry UM
57 Yin et al. 2019 Beijing, China 2010–2016 TS Mean NA (−14 to 35) NA Annual PM10 Mortality CVD (I00–I99) All ages Dwa D-continental UM
58 Chen et al. 2018 272 cities, China 2013–2015 TS Mean 15 (−0.5 to 25) 15 Annual PM10, O3 Mortality CVD (I00–I99) All ages NA NA UM
59 Hu et al. 2019 89 Zhejiang counties, China 2009–2015 TS Mean 16.9 (−2 to 35.3) 16.9 Annual PM10, O3 Mortality CVD (I00–I99) All ages Multi Multi UM
60 Yang et al. 2015 15 cities, China 2007–2013 TS Mean 5.3–21.6 (−28 to 36.7) 12 Annual NA Mortality CVD (I00–I99) All ages Dwa D-continental UM
61 Chen et al. 2013 Eight cities, China 1996–2008 TS Mean 16 (−22 to 34) 16 Annual PM10, NO2, SO2 Mortality Stroke (I60–I69) All ages Multi Multi UM
62 Zhang et al. 2021 Ganzhou, China 2015–2019 TS Mean 20.4 (−3 to 39) 20.4 Annual PM2.5, PM10, O3, NO2, CO, SO2 Mortality CVD (I00–I99) All ages Cfa C-subtropical UM
63 Rocklov and Forsberg 2008 Stockholm, Sweden 1998–2003 TS Mean NA (NA to NA) NA Annual NA Mortality CVD (I00–I99) All ages Cfb E-subarctic H
64 Yatim et al. 2021 Klang Valley, Malaysia 2006–2015 TS Mean 27.7 (23.5 to 30.9) 27.7 Annual PM10, O3 Mortality CVD (I00–I99) All ages Af A-tropical UM
65 Xu et al. 2022 Jiangsu, China 2015–2019 CC Mean NA (3.2–27.8) NA Annual PM2.5, PM10, O3, NO2, CO, SO2 Mortality CVD (I00–I99) All ages Cfa C-subtropical UM
66 Schlte et al. 2021 seven geographic regions in Switzerland 1998–2016 TS Mean 22 (3 to 40) 22 Annual NA Both CVD (I00–I99) All ages   D-continental H
67 Lu et al. 2021 Queensland, Australia 1997–2013 CC Mean 23.4 (−8.9 to 48.3) 23.4 Annual NA Mortality CVD (I00–I99) All ages Multi Multi H
68 Wang et al. 2015 two cities, China 2007–2009 TS Mean 15.65 (−9.4 to 34.6) 15.65 Annual PM10, NO2, SO2 Mortality CVD (I00–I99) All ages Dwa D-continental UM
69 Polcaro-Pichet 2019 Quebec, Canada 1981–2015 CC Mean NA (NA to NA) NA Cold (Nov–Apr) NA Mortality Stroke (I60–I69) All ages Dfb D-continental H
70 Klot et al. 2012 48 cities in the United States 1992–2000 CC Mean NA (NA to NA) NA Cold (winter) NA Mortality CVD (I00–I99) All ages Multi Multi H
71 Moghadamnia et al. 2018 Rasht, Iran 2005–2014 TS Tapp 17.38 (−2.6 to 38.6) 17.38 Annual NA Mortality CVD (I00–I99) All ages Cfa C-subtropical UM
72 Breitner et al. 2014 Three regions, Germany 1990–2006 TS Mean 9.5 (−15.3 to 28.7) 9.5 Cold (Dec–Feb) PM10, O3 Mortality CVD (I00–I99) All ages Cfb C-oceanic H
73 Chen et al. 2019 Augsburg, Germany 1987–2014 CC Mean 9.6 (−5.5 to 23.5) 9.6 Annual PM10, O3, NO2 Mortality CHD (I20–I25) All ages Cfb C-oceanic H
74 Nafstad et al. 2001 Oslo, Norway 1990–1995 TS Mean 12.5 (NA to NA) 12.5 Warm (Apr–Sep) NO2 Mortality CVD (I00–I99) All ages Dfb E-subarctic H
75 Zhang et al. 2018 Yinchuan, China 2010–2015 TS Mean 10.5 (−15 to 30.6) 10.5 Annual NA Mortality CVD (I00–I99) All ages BSk B-dry UM
76 Gholampour et al. 2019 Isfahan, Iran 2008–2016 TS Mean 17.52 (−7.4 to 35.4) 17.52 Annual NA Mortality CVD (I00–I99) All ages BWk B-dry UM
77 Tsoutsoubi et al. 2021 Greece 1999–2012 TS Mean NA(−3 to 42) NA Annual NA Mortality CVD (I00–I99) 70+ Multi Multi H
78 Kim et al. 2015 Seoul, South Korea 1995–2011 TS Mean 12.8 (−15.7 to 30.4) 12.8 Annual PM10 Mortality CVD (I00–I99) All ages Dwa D-continental H
79 Anderson and Bell 2009 107 communities, United States 1987–2000 TS Mean NA (NA to NA) NA Annual PM10, O3 Mortality CVD (I00–I99) All ages Multi Multi H
80 Saucy et al. 2021 Zurich, Switzerland 2000–2015 TS Mean 9 (−14 to 28) 9 Annual PM2.5, NO2 Mortality CVD (I00–I99) All ages Cfb C-oceanic H
81 Ma et al. 2021 47 prefectures, Japan 1972–2015 CC Cold spell NA (−0.5 to 18.6) NA Cold (Nov–Mar) NA Mortality CVD (I00–I99) All ages Multi Multi UM
82 Ebi et al. 2004 Three regions, United States 1983–1998 TS Min NA (NA to NA) NA Annual NA Morbidity CVD (I00–I99) 55+ Csa C-mediterranean H
83 Rocklov et al. 2014 Stockholm, Sweden 1990–2002 TS Cold spell NA (NA to NA) NA Annual NO2 Morbidity CVD (I00–I99) All ages   E-subarctic H
84 Wang et al. 2013 Jinan, China 1990–2009 TS Mean 15 (−10.5 to 35.8) 15 Annual NA Morbidity Stroke (I60–I69) All ages Cwa C-subtropical UM
85 Hajat et al. 2002 London, United Kingdom 1992–1995 CC Mean 8.1 (NA to NA) 8.1 Cold (Oct–Mar) SO2, O3, PM10 Morbidity CVD (I00–I99) 65+ Cfb C-oceanic H
86 Shaposhnikov et al. 2014 Moscow, Russia 1992–2005 TS Mean 5.5 (−17 to 25) 5.5 Annual NA Morbidity Stroke (I60–I69) All ages Dfb D-continental UM
87 Kovats et al. 2004 London, United Kingdom 1994–2000 TS Mean 11.6 (3.1 to 26.7) 11.6 Annual PM10, O3 Morbidity CVD (I00–I99) All ages Cfb C-oceanic H
88 Wu et al. 2011 Taiwan, China 1994–2003 CC Cold spell NA (NA to NA) NA Cold (Nov–Jan) NA Mortality CVD (I00–I99) All ages Cfa C-subtropical UM
89 Kysely et al. 2009 Czech Republic 1994–2006 CC Cold spell NA (−1.9 to 21.9) NA Cold (Dec–Feb) NA Mortality CVD (I00–I99) 25+   C-oceanic H
90 Madrigano et al. 2013 Worcester, United States 1995–2003 CC Cold spell 7.9 (−6.9 to 25.5) 7.9 Annual PM2.5, O3 Mortality ACS (I21–I22) All ages Dfa D-continental H
91 Lu et al. 2020 Queensland, Australia 1995–2016 CC Mean 25.9 (−8.9 to 48.8) 25.9 Annual NA Morbidity CVD (I00–I99) All ages Multi Multi H
92 Bai et al. 2018 Ontario, Canada 1996–2013 TS Mean NA (−33.1 to 32.2) NA Annual PM10, NO2, PM2.5 Morbidity CVD (I00–I99) All ages Dfb D-continental H
93 Chen et al. 2010 Taiwan, China 1997–2003 TS Cold spell NA (NA to NA) NA Annual NA Mortality CVD (I00–I99) All ages Cfa C-subtropical UM
94 Martinez-Solanas and Basagana 2017 Spain 1997–2013 TS Max 20.9 (6.88 to 34.69) 20.9 Annual NA Morbidity CVD (I00–I99) All ages Multi Multi H
95 Mohammad et al. 2020 Hong Kong, China 1998–2011 TS Mean 23.52 (8.2 to 31.8) 23.52 Annual PM10, NO2, O3 Morbidity CVD (I00–I99) All ages Cfa C-subtropical H
96 Ryti et al. 2017 Oulu, Finland 1998–2011 CC Cold spell 1.4 (−41.3 to 33) 1.4 Annual NA Mortality CVD (I00–I99) All ages   E-subarctic H
97 Ryti et al. 2018 Oulu, Finland 1998–2011 CC Cold spell 1.4 (−41.3 to 33) 1.4 Annual NA Mortality Atherosclerotic heart disease (I25.1) 35+   E-subarctic H
98 Sartini et al. 2016 London, United Kingdom 1998–2012 TS Cold spell NA (NA to NA) NA Annual NA Mortality CVD (I00–I99) 60+ Cfb C-oceanic H
99 Wichmann et al. 2012 Copenhagen, Denmark 1999–2006 CC TappMax 16 (0 to 30) 16 Annual PM10, NO2, CO Morbidity ACS (I21–I22) 18+ Cfb C-oceanic H
100 Wang and Lin 2014 Taipei, China 2000 –2009 TS Mean 23.4 (8.3 to 33) 23.4 Annual PM10, NO2, O3 Morbidity CVD (I00–I99) All ages Cfa C-subtropical UM
101 Liang et al. 2008 Taichung, China 2000–2003 TS Mean 27–29 (NA to NA) 28 Annual PM10, NO2, CO, SO2, O3 Morbidity ACS (I21–I22) All ages Cfa C-subtropical H
102 Revich and Shaposhnikov 2008 Moscow, Russia 2000–2006 CC Cold spell NA (NA to NA) NA Annual NA Mortality CHD (I20–I25), stroke (I60–I69) All ages Dfb D-continental H
103 Dahlquist et al. 2016 Stockholm, Sweden 2000–2010 CC Mean 7.1 (−18.2 to 25.2) 7.1 Annual PM10, O3 Morbidity OHCA (I46) All ages Cfb C-oceanic H
104 Lin et al. 2021 Five cities, Taiwan, China 2000–2014 TS Mean 23.1–25.4 (NA to NA) 14 Annual PM10, O3, NO2, SO2 Morbidity CVD (I00–I99) 40+ Cfa C-subtropical H
105 Vaičiulis et al. 2021 Kaunas, Lithuanian 2000–2015 TS Cold spell NA (NA to NA) NA Cold (Nov–Jan) NA Mortality MI (I21–I23) 25+   D-continental H
106 Ma et al. 2013 Shanghai, China 2001–2009 TS Cold spell 17.5 (−3.4 to 39) 17.5 Cold (Jan–Mar) NA Mortality CVD (I00–I99) All ages Dwa D-continental UM
107 Wichmann et al. 2011 Greater Copenhagen, Denmark 2002–2006 CC Tapp 10 (−8 to 30) 10 Cold (Oct–Mar) PM10, NO2, CO Morbidity CVD (I00–I99) All ages Cfb C-oceanic H
108 Goggins et al. 2017 Hong Kong, China 2002–2011 TS Mean 23.4 (8.2 to 31.8) 23.4 Annual PM10, O3 Morbidity CVD (I00–I99) 0–59 ages Cfa C-subtropical H
109 Kim et al. 2021 Seven metropolitan provinces, South Korea 2002–2017 TS Mean NA (NA to NA) NA Annual PM2.5, PM10, O3, NO2, CO, SO2 Morbidity ACS (I21–I22) All ages Multi Multi H
110 Misailidou et al. 2006 Five rural regions, Greece 2003–2004 TS Mean NA (NA to NA) NA Annual NA Morbidity ACS (I21–I22) All ages Multi Multi H
111 Vasconcelos et al. 2013 Lisbon and Oporto, Portugal 2003–2007 TS Mean NA (NA to NA) NA Cold (winter) PM10 Morbidity MI (I21–I23) All ages Csa C-mediterranean H
112 Son et al. 2014 Eight cities, South Korea 2003–2008 TS Mean 14.1 (12.6 to 16.2) 14.1 Annual NA Morbidity CVD (I00–I99) All ages Multi Multi H
113 Ma et al. 2011 Shanghai, China 2005–2008 TS Cold spell 17.7 (−3.1 to 34.1) 17.7 Cold (Jan–Feb) NA Morbidity CVD (I00–I99) All ages Cfa C-subtropical UM
114 Cho et al. 2018 Seoul, South Korea 2005–2009 TS Mean 12.9 (−11.5 to 30.1) 12.9 Annual O3, PM2.5 Morbidity Stroke (I60–I69) All ages Dwa D-continental H
115 Yamazaki and Michikawa 2017 Three prefectures, Japan 2005–2012 CC Mean 16.83 (NA to NA) 16.83 Annual NA Morbidity OHCA (I46) All ages Cfa C-subtropical H
116 Bai et al. 2014 Lhasa, Tibet, China 2005–2012 TS Max 9.6 (−16.1 to 30.4) 9.6 Annual NA Morbidity CVD (I00–I99) All ages Dwb D-continental UM
117 Tian et al. 2016 Hong Kong, China 2005–2012 CC Mean 23.4 (8.7 to 31.8) 23.4 Annual NO2, PM10, O3 Morbidity CVD (I00–I99) All ages Cfa C-subtropical UM
118 Xu et al. 2021 Brisbane, Australia 2005–2013 CC Mean NA (NA to NA) NA Annual PM10, NO2 Morbidity Stroke (I60–I69) All ages Cfa C-subtropical H
119 Moghadamnia et al. 2018 Rasht, Iran 2005–2014 TS Tapp 17.4 (−2.6 to 38.6) 17.4 Annual NA Morbidity ACS (I21–I22) All ages Cfa C-subtropical UM
120 Onozuka et al. 2017 Japan 2005–2014 TS Mean 9.4–23.2 (−10.7 to 33.7) 16 Annual NA Morbidity OHCA (I46) All ages Multi Multi H
121 Shin et al. 2021 Seoul, South Korea 2005–2014 CC Min 12.7 (NA to NA) 12.7 Annual NO2, CO, SO2 Morbidity MI (I21–I23) All ages Dwa D-continental H
122 Cheng et al. 2020 Brisbane, Australia 2005–2015 TS Mean 16 (10 to 25) 16 Annual PM10, NO2, Morbidity MI (I21–I23) All ages Cfa C-subtropical H
123 Zhou et al. 2014 15 provinces in China 2006–2010 CC Cold spell NA (−1.9 to 21.9) NA Cold (Dec–Feb) NA Mortality CVD (I00–I99) All ages Cfa C-subtropical UM
124 Ponjoan et al. 2017 Two regions, Spain 2006–2013 CC Cold spell 28.1 (27.3 to 29) 28.1 Cold (Nov–Jan) PM10, O3, NO2, SO2, CO Morbidity CVD (I00–I99) All ages Multi Multi H
125 Lee et al. 2014 16 cities, South Korea 2006–2014 TS Mean 13.3 (−19.5 to 37.7) 13.3 Annual PM10, NO2, CO, SO2, O3,PM2.5 Morbidity ACS (I21–I22) All ages Multi Multi H
126 Wang et al. 2020 Taiwan, China 2006–2014 TS Mean 23.4 (10.6 to 31) 23.4 Annual PM10,NO2, PM2.5 Morbidity OHCA (I46) All ages Cfa C-subtropical H
127 Moraes et al. 2022 São Paulo, Brazil 2006–2015 CC Cold spell NA (NA to NA) NA Annual PM10 Mortality CVD (I00–I99) All ages Cfa C-subtropical UM
128 Chen et al. 2019 31 cities, China 2007–2013 CC Cold spell NA (NA to NA) NA Annual NA Mortality CVD (I00–I99) All ages Cfa C-subtropical UM
129 Doan et al. 2021 Brisbane, Australia 2007–2019 TS Mean 20.9 (10.4 to 30.1) 20.9 Annual NA Morbidity OHCA (I46) All ages Cfa C-subtropical H
130 Giang et al. 2014 Thai Nguyen, Vietnam 2008–2012 TS Mean 23.6 (21 to 27.5) 23.6 Annual NA Morbidity CVD (I00–I99) 60+ Cfa C-subtropical LM
131 Niu et al. 2016 Guangzhou, China 2008–2013 TS Mean 22.3 (5.1 to 33.5) 22.3 Annual PM10, NO2, SO2 Morbidity OHCA (I46) All ages Cfa C-subtropical UM
132 Thu Dang et al. 2019 Two Central Coast regions, Vietnam 2008–2015 TS Mean 26.1 (15.0 to 36.9) 26.1 Annual NA Morbidity ACS (I21–I22) All ages NA NA LM
133 Bijelović et al. 2017 Novi Sad, Serbia 2010–2011 TS Mean NA (NA to NA) NA Annual NA Morbidity ACS (I21–I22) 19+ Cfa C-subtropical UM
134 Sangkharat et al. 2020 London, United Kingdom 2010–2014 TS Mean 11.8 (−2.2 to 25.4) 11.8 Annual PM10, NO2, CO, SO2, O3, PM2.5 Morbidity CVD (I00–I99) All ages Cfb C-oceanic H
135 Hensel et al. 2017 Hamburg, Germany 2010–2014 TS Mean 10 (NA to NA) 10 Annual NA Morbidity CVD (I00–I99) All ages Cfb C-oceanic H
136 Pourshaikhian et al. 2019 Rasht, Iran 2010–2015 TS Tapp 30.1 (NA to NA) 30.1 Warm (May–Sep) NA Morbidity CVD (I00–I99) All ages Cfa C-subtropical UM
137 Zhan et al. 2022 Fujian province, China 2010–2016 TS TappMean 20 (−2 to 33.8) 20 Annual PM10, NO2, CO, SO2 Morbidity CVD (I00–I99) All ages Cfa C-subtropical UM
138 Han et al. 2017 Jinan, China 2011–2014 TS Cold spell 14.7 (−9.4 to 34) 14.7 Annual NA Mortality CVD (I00–I99) All ages Cwa C-subtropical UM
139 Mohammadi et al. 2021 Sabzevar, Iran 2011–2017 TS Tapp 12.9 (−11.2 to 45.4) 12.9 Annual NA Morbidity CVD (I00–I99) All ages DSk D-continental UM
140 Zhao et al. 2018 Ningxia Hui Autonomous Region, China 2012–2015 TS Mean 8.5 (−18.6 to 29.7) 8.5 Annual NO2, CO, SO2, PM2.5 Morbidity CVD (I00–I99) All ages BWk B-dry UM
141 Luo et al. 2017 Beijing, China 2013–2014 TS Mean 11.6 (−12.9 to 30.1) 11.6 Annual PM2.5 Morbidity Stroke (I60–I69) All ages Cfa C-subtropical UM
142 Guo et al. 2017 Guangzhou, China 2013–2015 TS Mean NA (NA to NA) NA Annual NO2, SO2, O3, PM2.5 Morbidity Stroke (I60–I69) All ages Cfa C-subtropical UM
143 Gao et al. 2019 Hefei, China 2013–2015 TS Cold spell NA (NA to NA) NA Annual NO2, PM10, O3 Morbidity CVD (I00–I99) All ages Cfa C-subtropical UM
144 Lei et al. 2022 272 cities, China 2013–2015 CC Mean NA Annual PM2.5, O3 Mortality CVD (I00–I99) All ages Cfa C-subtropical UM
145 Liu et al. 2018 Beijing, China 2013–2016 TS Mean 12.8 (−16 to 32) 12.8 Annual NA Morbidity ACS (I21–I22) All ages Dwa D-continental UM
146 Aklilu et al. 2020 Beijing, China 2013–2017 CC Mean 13.9 (−14.1 to 32.6) 13.9 Annual PM10, NO2, CO, SO2, O3, PM2.5 Morbidity CVD (I00–I99) All ages Dwa D-continental UM
147 Garcı´a-Lledó et al. 2020 Madrid, Spain 2013–2017 TS Max NA (NA to NA) NA Annual NA Morbidity ACS (I21–I22) All ages BSk B-dry H
148 Guo et al. 2020 Yancheng, China 2013–2018 TS Mean 15.2 (−4.7 to 32.9) 15.2 Annual PM2.5, O3, NO2, CO, SO2 Morbidity ACS (I21–I22) All ages Cfa C-subtropical UM
149 Wang et al. 2021 Qingdao, China 2014–2017 TS Mean 14.9 (NA to NA) 14.9 Annual PM10, PM2.5, SO2, NO2, CO, O3 Morbidity CVD (I00–I99) All ages Cfa C-subtropical UM
150 Wang et al. 2021 Shenzhen, China 2015–2016 TS Mean 23.5 (NA to NA) 23.5 Warm (May–Oct) SO2, O3, PM2.5 Morbidity CVD (I00–I99) All ages Cfa C-subtropical UM
151 Cui et al. 2019 Hefei, China 2015–2017 TS Mean 18.1 (−5.9 to 35.6) 18.1 Annual PM10, NO2, SO2 Morbidity CVD (I00–I99) All ages Cfa C-subtropical UM
152 Mohammad et al. 2018 Sweden 2017–2018 TS Min NA (−11.1 to 11.5) NA Annual NO2, CO, O3, PM2.5 Morbidity CHD (I20–I25) All ages Dfb E-subarctic H
153 Li et al. 2021 Beijing, China 2017–2019 TS Tapp 10 to 12 (−6 to 33) 11 Annual PM10, NO2, CO, SO2, O3, PM2.5 Morbidity ACS (I21–I22) All ages Dwa D-continental UM
154 Borghei et al. 2020 Rasht, Iran 3-year period TS TappMean 17.2 (NA to NA) 17.2 Annual NA Morbidity OHCA (I46) All ages Cfa C-subtropical UM
155 Li et al. 2017 Shenyang, China 2006–2015 TS Mean 8.2 (−24.0 to 29) 8.2 Annual PM10, NO2, SO2 Morbidity DVT (I82) All ages Dwa D-continental UM
156 Chiara et al. 2021 8,084 municipalities of Italy 2006–2015 CC Mean 13.7 (−25.8 to 38.5) 13.7 Annual PM2.5, PM10 Morbidity DVT (I82) All ages Cfb C-oceanic H
157 Chen et al. 2022 11 cities, China 2009–2019 CC Mean NA (NA to NA) NA Annual PM2.5, O3 Morbidity AAD (I71) All ages Cfa C-subtropical UM
158 Yu et al. 2021 Wuhan, China 2011–2018 TS Mean NA (NA to NA) NA Annual SO2, NO2 Morbidity AAD (I71) All ages Cfa C-subtropical UM
159 Zhang et al. 2022 131 cities, China 2015–2020 CC Mean NA (NA to NA) NA Annual PM2.5, O3, NO2, CO, SO2 Morbidity AAD (I71) All ages Cfa C-subtropical UM

T-S, time series; C-C, case-crossover; Min, minimum temperature; Max, maximum temperature; Mean, mean temperature; Tapp, apparent temperature; D, daily resolution; W, weekly resolution; M, monthly resolution; ACS, acute coronary syndrome; OHCA, out-of-hospital cardiac arrest; H, high-income; UM, upper-middle income; LM, lower-middle income; NA, not available.

2.4. Study quality assessment

We further appraised the evidence included in the meta analysis by applying the risk of bias (RoB) assessment in each study and assessment of quality and strength of the body of included studies. A detailed description of the criteria for the assessments is provided in Supplementary Tables S3–S5.

2.5. Statistical analysis

A random-effects meta-analysis was used to compute the relative risk (RR) estimates associated with cold exposure. We converted all RRs to RRs associated with a 1°C decrease below the reference temperature points, assuming a log-linear relationship between mortality/morbidity and temperature below the reference temperature points (6, 19). If studies reported multiple lag RRs, we selected the lag associated with the maximum risk to conduct the meta-analysis. Subgroup analysis was carried out to analyze the vulnerabilities stratified by age, sex, national income level, and climate zones (classified by the Köppen–Geiger climate zones) (20). Subgroup interaction was employed to detect the significance of differences among subgroups.

I2 statistics and Cochrane Q were used to examine heterogeneity among effect estimates. The heterogeneity of pooled estimates with p < 0.10 (Cochrane Q test) was deemed significant (21). I2 statistics of 0%–25%, 25%–50%, and >50% indicated low, moderate, and high heterogeneity, respectively (21). Funnel plots and Egger's test were used to evaluate potential publication bias, and the Trim and Fill method was used to examine the sensitivity of the results to publication bias. Sensitivity analyses were carried out, separating studies by temperature metrics, study design (time-series or case-crossover), seasonality, lag effects, and air pollution adjustment for low-temperature exposure, and intensity (classified by cold spell duration) for cold spells. We further examined the influence of individual estimates on the pooled RRs using a leave-one-out analysis. Meta-regressions were used further to explore the heterogeneity of effects and determinants of heterogeneity. Statistical analyses were done using Stata (version 16.0).

3. Results

3.1. Search and study selection results

The initial database searches produced a total of 21,724 articles, of which 1,672 duplicate records were excluded. After screening titles and abstracts, 19,773 articles were eliminated for being irrelevant. We included 166 articles for full-text review and excluded 7: 1 because of an overlapping database (22) and 6 for providing non-standardizable estimates (2328). Ultimately, we identified 159 studies on the basis of the inclusion criteria for the final review (Figure 1).

3.2. Study characteristic

The characteristics of the included studies presented in Table 1. Of these, 135 studies reported low-temperature effects, 21 assessed cold spell effects, and 2 examined the effects of both low temperature and cold spells. A total of 80 studies reported cardiovascular mortality, 64 studies assessed morbidity, and 4 reported both health outcomes. According to the Köppen–Geiger climate zones classification (20), 7 studies were carried out in the tropical zone, 7 in the dry zone, 5 in the Mediterranean zone, 16 in the oceanic climate zone, 63 in the subtropical zone, 29 in the continental area, and 7 in the subarctic zone (Figure 2).

Figure 2.

Figure 2

Geographical distribution of city-specific or region-specific cardiovascular disease mortality (A) and cardiovascular disease morbidity (B) estimates included in the meta-analysis by considering the Köppen–Geiger climate zone. Af, tropical rainforest climate; Am, tropical monsoon climate. Aw, tropical savanna climate. BWh, hot desert climate; BWk, cold desert climate; BSh, hot semiarid climate; BSk, cold semiarid climate; Csa, hot-summer Mediterranean climate; Csb, warm-summer Mediterranean climate; Csc, cold-summer Mediterranean climate; Cwa, monsoon-influenced humid subtropical climate; Cwb, subtropical highland climate or monsoon-influenced temperate oceanic climate; Cwc, cold subtropical highland climate or monsoon-influenced subpolar oceanic climate; Cfa, humid subtropical climate; Cfb, temperate oceanic climate; Cfc, subpolar oceanic climate; Dsa, Mediterranean-influenced hot-summer humid continental climate; Dsb, Mediterranean-influenced warm-summer humid continental climate; Dsc, Mediterranean-influenced subarctic climate; Dsd, Mediterranean-influenced extremely cold subarctic climate; Dwa, monsoon-influenced hot-summer humid continental climate; Dwb, monsoon-influenced warm-summer humid continental climate; Dwc, monsoon-influenced subarctic climate; Dwd, monsoon-influenced extremely cold subarctic climate; Dfa, hot-summer humid continental climate; Dfb, warm-summer humid continental climate; Dfc, subarctic climate; Dfd, extremely cold subarctic climate; ET, tundra climate; EF, ice cap climate.

3.3. Meta-analysis of low-temperature effects

An analysis of pooled estimates showed that for every 1°C decrease in temperature, cardiovascular disease–related mortality increased by 1.6% [RR 1.016; 95% confidence interval (CI) 1.015–1.018] (Figure 3 and Supplementary Figure S1), and cardiovascular morbidity increased by 1.2% (RR 1.012; 95% CI 1.010–1.014) (Figure 4 and Supplementary Figure S2). Cause-specific analyses showed positive associations between low temperatures and the mortality of coronary heart disease (CHD) (RR 1.015; 95% CI 1.011–1.019), heart failure (HF) (RR, 1.008; 95% CI 1.003–1.013), and stroke (RR 1.012; 95% CI 1.008–1.016), while cold temperatures showed no significant association with hypertensive diseases and cardiac arrest mortality. For morbidity, low temperatures increased the morbidity of all kinds of cardiovascular diseases, apart from hypertensive diseases. Moreover, higher morbidity risks were attributable to HF (RR 1.030; [95% CI 1.013–1.048]), aortic aneurysm and dissection (AAD) (RR 1.026; 95% CI 1.021–1.031), and out-of-hospital cardiac arrest (RR 1.024; 95% CI 1.012–1.035). We further analyzed the cold effects in a population with different characteristics to explore the population's vulnerability. We found that people aged 65 or older were more vulnerable to cardiovascular disease–related mortality (p = 0.056). Considering the climate zones, a significant greater risk of cardiovascular disease–related mortality was observed in those who lived in tropical (p = 0.004) and subtropical (p < 0.001) climate zones than those in the subarctic climate zone. In addition, cardiovascular morbidity was significantly higher in people living in lower-middle-income countries than in those living in high-income and upper-middle-income countries (p = 0.002).

Figure 3.

Figure 3

The impact of low temperatures on RR and 95% CIs for cardiovascular disease mortality in different groups. RR, relative risk; n, the number of effect estimates; CI, confidence interval.

Figure 4.

Figure 4

The impact of low temperatures on RR and 95% CIs for cardiovascular disease morbidity in different groups. RR, relative risk; n, the number of effect estimates; CI, confidence interval.

3.4. Meta-analysis of cold spell effects

Cold spells had a significant impact on cardiovascular outcomes, which increased cardiovascular disease–related mortality by 32.4% (RR 1.324; 95% CI 1.234–1.421) (Figure 5 and Supplementary Figure S3) and morbidity by 13.8% (RR 1.138; 95% CI 1.015–1.276) (Figure 6 and Supplementary Figure S4). There was no significant difference among cold spells of different intensities. Moreover, in the subarctic climate zone (RR 1.452; 95% CI 1.164–1.811), the effect of the cold spell on cardiovascular mortality was significantly higher than that in the continental area (p = 0.049).

Figure 5.

Figure 5

The impact of cold spells on RR and 95% CIs for cardiovascular disease mortality in different groups. RR, relative risk; n, the number of effect estimates; CI, confidence interval.

Figure 6.

Figure 6

The impact of cold spells on RR and 95% CIs for cardiovascular disease morbidity in different groups. RR, relative risk; n, the number of effect estimates; CI, confidence interval.

3.5. Heterogeneity analysis

We found high heterogeneity in the summary effect estimates of low temperature (heterogeneity p-values < 0.0001, and all I2 > 50%). The stratification by sex and age did not help reduce heterogeneity, while it decreased in hypertensive disease and HF mortality stratum (Figure 3). We further conducted sensitivity analyses, finding no significant differences in the pooled RRs for the associations between cold exposure and cardiovascular disease–associated health outcomes in the leave-one-out analysis (low-temperature mortality RR 1.015–1.018; low-temperature morbidity RR 1.011–1.013). Moreover, for cardiovascular disease–related mortality and morbidity, a series of sensitivity analyses done by separating studies by temperature metrics, study design, seasonality, lag effects, and air pollution adjustment showed consistency in the direction and magnitude of the associations in the reviewed studies (Table 2). We further explored the source of heterogeneity using meta-regression (Supplementary Table S6), which showed that the lower-middle-income level was positively correlated with a 1% decrease in RRs for cold effects on cardiovascular morbidity (RR 1.124; 95% CI 1.035–1.221; p = 0.007; ref = high income level). The heterogeneity in the summary effect estimates of cold spells was large (heterogeneity p-values < 0.0001, and all I2 > 90%). The stratification of cold spell intensity only reduced the heterogeneity of the estimated morbidity RRs. No significant difference in the pooled RRs for the associations between cold spells and cardiovascular mortality was found in the leave-one-out analysis (cold spell morbidity RR 1.279–1.372). However, two (33%) studies could render the pooled effects of cold spells on cardiovascular disease morbidity insignificant when left out from the analysis.

Table 2.

Sensitivity analysis of random-effects meta-analysis showing relative risk (RR) and 95% confidence intervals (CIs), for the association between low temperatures and cardiovascular disease morbidity, with every 1°C decrease in temperature.

k RR lci uci I 2 p Eagger's p-value
Adjusted for air pollution
 PM2.5 mortality 10 1.021 1.012 1.031 95.90% <0.0001 0.013
 PM2.5 morbidity 16 1.010 1.007 1.014 91.80% <0.0001 0.005
 PM10 mortality 40 1.024 1.019 1.028 96.40% <0.0001 <0.0001
 PM10 morbidity 21 1.020 1.013 1.028 91.40% <0.0001 0.001
 O3 mortality 28 1.028 1.021 1.035 96.90% <0.0001 0.003
 O3 morbidity 19 1.028 1.012 1.027 92.10% <0.0001 0.002
 NO2 mortality 31 1.026 1.019 1.032 95.80% <0.0001 0.001
 NO2 morbidity 22 1.011 1.008 1.015 90.70% <0.0001 0.001
 CO mortality 8 1.025 1.010 1.040 96.60% <0.0001 0.658
 CO morbidity 5 1.019 1.008 1.030 87.20% <0.0001 0.865
 SO2 mortality 20 1.026 1.018 1.035 95.60% <0.0001 0.068
 SO2 morbidity 18 1.016 1.009 1.023 79.30% <0.0001 0.001
 Null mortality 28 1.011 1.009 1.014 91.10% <0.0001 0.001
 Null morbidity 21 1.007 1.005 1.009 89.10% <0.0001 <0.0001
Exposure
 Tmean mortality 70 1.015 1.013 1.016 95.30% <0.0001 <0.0001
 Tmean morbidity 48 1.014 1.011 1.017 90.00% <0.0001 <0.0001
 Tmax mortality
 Tmax morbidity 3 1.022 0.998 1.046 83.00% 0.003 0.169
 Tmin mortality 4 1.019 1.009 1.029 65.30% 0.034 0.001
 Tmin morbidity 3 1.012 0.996 1.028 93.70% <0.0001 0.306
 Tapp mortality 5 1.015 1.005 1.025 93.40% <0.0001 0.168
 Tapp morbidity 9 1.003 1.000 1.006 49.80% 0.063 0.028
Season
 Annual mortality 66 1.017 1.015 1.019 95.70% <0.0001 <0.0001
 Annual morbidity 56 1.011 1.009 1.013 88.70% <0.0001 <0.0001
 Cold mortality 9 1.010 1.006 1.015 84.20% <0.0001 0.025
 Cold morbidity 5 1.041 1.011 1.071 92.50% <0.0001 0.235
 Warm mortality 6 1.019 1.015 1.024 0.00% 0.757 0.95
 Warm morbidity 2 1.010 0.989 1.032 79.40% 0.028
Study design
 TS mortality 68 1.019 1.017 1.022 95.50% <0.0001 <0.0001
 TS morbidity 48 1.011 1.090 1.013 89.80% <0.0001 <0.0001
 CC mortality 12 1.007 1.005 1.008 83.20% <0.0001 0.007
 CC morbidity 14 1.016 1.009 1.022 91.50% <0.0001 0.007
Lag days mortality (day)
 Cumulative 0–9 17 1.017 1.010 1.023 96.80% <0.0001 0.142
 Cumulative 10–19 15 1.017 1.012 1.022 88.70% <0.0001 0.0002
 Cumulative >20 10 1.025 1.017 1.034 96.00% <0.0001 0.002
 Single 0–9 13 1.022 1.015 1.028 96.90% <0.0001 0.02
 Single >10 8 1.022 1.013 1.032 92.00% <0.0001 0.083
Lag days morbidity (day)
 Cumulative 0–9 7 1.005 1.001 1.009 74.50% <0.0001 0.054
 Cumulative 10–19 4 1.012 1.003 1.022 91.80% <0.0001 0.628
 Cumulative >20 12 1.043 1.029 1.058 81.30% <0.0001 0.008
 Single 0–9 11 1.015 1.009 1.02 66.70% 0.001 0.032
 Single >10 4 1.003 1 1.005 72.20% 0.013 0.359
Risk of bias (mortality)
 Low 10 1.014 1.009 1.019 83.20% <0.0001 0.007
 Probably low 67 1.016 1.015 1.018 95.50% <0.0001 <0.0001
 Probably high 3 1.021 1.012 1.027 83.00% 0.003 0.169
Risk of bias (morbidity)
 Low 6 1.017 1.009 1.026 96.60% <0.0001 0.658
 Probably low 51 1.012 1.01 1.014 91.50% <0.0001 <0.0001
 Probably high 4 1.023 1.014 1.027 93.40% <0.0001 0.216

RR, relative risk, CI, confidence interval; T-S, time series; C-C, case-crossover; Tmin, minimum temperature; Tmax, maximum temperature; Tmean, mean temperature; Tapp, apparent temperature.

3.6. Rob and study quality assessment

We assessed the RoB of the included studies and rated the overall RoB according to the key components such as exposure, outcome, and confounding bias. The details of the RoB assessment criteria and individual studies’ assessment are given in Supplementary Table S2 and Supplementary Figure S5. In summary, of the 154 (96%) studies that were rated with low risk or probably with a low risk of bias, 8 (4%) were rated with probably high risk, and no study was rated with a high risk of overall bias (Supplementary Figure S6). The initial quality rating was moderate, as the evidence was derived from observational studies. The evidence quality of studies on the effects of cold exposures (low temperature and cold spells) on cardiovascular disease–related mortality and morbidity was downgraded because of inconsistent results. All the I² >50% and 80% prediction intervals (PIs) included unity and were more than twice the random-effects meta-analysis confidence interval. Further, we upgraded the quality rating of the studies to moderate for the evident exposure-response gradients, except for the study on the effect of the cold spell on cardiovascular morbidity for its inconsistent dose response across studies (Supplementary Table S8).

4. Discussion

The present study aimed to clarify the effects of cold exposure (low temperature and cold spell) on cardiovascular disease–related health outcomes (mortality and morbidity) and explore the population’s susceptibilities to cold-induced cardiovascular diseases. We systematically reviewed 159 articles in the synthesis to strengthen the evidence on the increase in cardiovascular disease risk due to cold ambient exposures and to clarify the magnitude of cold impact. We provided new knowledge that the risk of cold exposure to cardiovascular diseases varies among climate zones. The meta-analysis indicated that following every 1°C decrease, cardiovascular-associated mortality increased by 1.6% and morbidity by 1.2%. A more substantial effect was observed in the morbidity of cardiac arrest and AAD, while the impact of cold exposure on hypertensive disease outcomes was not significant. Notably, cold spells significantly increased cardiovascular-related mortality and morbidity by 32.4% and 13.8%, respectively.

Our results update the findings of the previous studies and clarify the impact of cold exposure on cardiovascular outcomes with regard to both its direction and magnitude (11, 13, 14). Knowledge generated from previous studies was consistent in terms of direction, showing the positive association between cold exposure and cardiovascular disease outcomes, while its magnitude remained disputable. To better understand the extent of cold impact on cardiovascular disease, we conducted a wide-ranging search and analysis of current evidence with available information on daily temperature, location, and International Classification of Diseases-coded cause of death. Specifically, 80 studies exploring the association between low temperatures and cardiovascular disease were included in the meta-analysis. We found that with every 1°C drop in temperature, the RR of cardiovascular disease increased by 1.6%. We further conducted a series of sensitivity analyses by carrying out an elaborate stratification on included literature, considering the confounding factors. An analysis of different stratifications of the study also showed similar results for both direction and magnitude. These results suggested the robustness of our conclusion, which may be more in accord with the actual situation. Furthermore, we analyzed the impact of low temperatures on cardiovascular morbidity using the same method, which showed a 1.2% increase in cardiovascular mortality with every 1°C decrease. Notably, the impact of cold spells on cardiovascular disease was considerable, which increased mortality by 32.4% and morbidity by 13.8%. Our results provided the latest and unbiased evidence of the association between cold exposure and cardiovascular disease, which may help researchers better evaluate the impact of climate change.

The varied magnitude of cold impact suggests the existence of some crucial factors that could influence cold impact on cardiovascular health. Exploring these influential factors and the population’s susceptibility to cold-induced cardiovascular diseases is an important finding in our review. Here, we analyzed the cold effects in different climate conditions by stratifying the included articles using the Köppen–Geiger climate zones classification (20). As the results showed, the increased mortality caused by low-temperature exposure was more pronounced in a location with a higher mean daily temperature, such as the tropical climate zone (24.23°C; RR 1.023), Mediterranean climate zone (20.19°C; RR 1.024), and subtropical climate zone (19.78°C; RR1.032). In comparison, the cold effects were less pronounced in those with a lower mean daily temperature, such as the oceanic climate zone (10.38°C; RR 1.009), continental climate zone (14.60°C; RR 1.010), and subarctic climate zone (9.23°C; RR 1.009). Similar results have been found in clinical research worldwide (7, 29, 30). For example, Ebi and Mills reported that cold-related mortality increased significantly in regions with higher winter temperatures in the United Kingdom (29). Furthermore, Guo et al. found that the cold effects in southern China were more pronounced than in northern cities (7). Locations with higher mean temperatures tend to have higher optimal temperatures and to be intolerant to a fall in temperature, probably through physical adaptation (1). More importantly, social adaptation may play an even more critical role, as it is a known fact that the susceptible population, such as the elderly and patients with cardiovascular disease, should wear protective clothing and remain active in cold weather outdoors (29). However, The Eurowinter Group reported that in relatively warm countries, such a population often does not follow such practices because they do not feel the need (30). These findings suggest that excessive deaths in some instances could be avoided by way of the authorities taking several steps to promote subjective measures and public measures such as wind-proofing bus shelters. In addition, cold-related mortality is significantly higher in countries with lower-middle-income levels. The social capacity to adapt is also probably tied to economic development. People living in such countries may have less capacity to adapt to decreased temperatures, potentially exacerbating health inequalities across countries.

We further examined the cold impact on different kinds of cardiovascular diseases classified by the International Classification of Diseases-coded. Among them, cold exposure showed the most potent impact on the mortality of CHD and the morbidity of out-of-hospital cardiac arrest and AAD. In contrast, its role in hypertensive disease outcomes was not significant. Mechanically, the autonomic nervous system and humoral regulation system consist of a precise network to maintain blood pressure, which may not easily be disturbed by a change in ambient temperature. Moreover, our finding is consistent with that of a previous meta-analysis that explored the association between low temperatures and blood pressure. It was reported that a 1°C decrease in the mean daily outdoor temperature increased the systolic and diastolic blood pressure by 0.26 and 0.13 mmHg, respectively (12). These results suggested a possible correlation between decreased temperature and the incidence of hypertensive diseases, while the precise relationship remained largely unknown, which warrants future research. For example, research with a more detailed classification of the extent of temperature change and patients with underlying diseases is still needed. Recently, a meta-analysis that explored the effect of heat exposure on cardiovascular diseases reported a 2.8% and 17% increase in cardiovascular mortality followed by high temperatures and heat wave exposure, respectively (19). Coincidentally, both heat and cold exposure exercised the most substantial impact on cardiac arrest and minimum effect on hypertensive disease (19). Future exploration of the critical mechanism elicited by non-optimal temperatures may explain the results.

Cold temperatures could impact cardiovascular activity through many mechanisms. For example, cold exposure increases blood viscosity by elevating blood, platelet count, and red blood cell count in a few hours, which may increase the risk of ischemic heart disease and stroke (3133). This could explain our study’s finding of increased risk of CHD and stroke after cold exposure. Furthermore, the present analysis suggested a high correlation between low temperatures and cardiac arrest morbidity, which may be explained by cold-induced autonomic nervous system disruption and inflammation–coagulation cascade activation (3436). In addition, cold exposure was found to be associated with several risk factors for cardiovascular disease. It was reported that exposure to lower temperatures could be associated with a higher risk of metabolic derangement, including higher plasma glucose and more insulin resistance (37). Moreover, patients with diabetes were more prone to cold-related cardiovascular disease (38). Similarly, cold exposure impacted lipid metabolism disorder and influenza epidemics (39) and may induce more fat and alcohol intake.

The present synthesis showed a substantial interstudy heterogeneity. Considering the significant number of studies included in this meta-analysis, it is hard to avoid some inherent differences related to factors such as study design, meteorological variables, study population, and statistical mode. To analyze the source of heterogeneity, we carried out sensitivity analysis, subgroup analysis, and meta-regression, considering various covariant aspects such as temperature metrics, study design, study season, lag effects, air pollution adjustment, and cold spell intensity. However, all these analyses failed to reduce heterogeneity, indicating that other unmeasured factors still contribute to the cold effects on cardiovascular diseases as covariants, which still needs future research. In addition, a series of sensitivity analyses done by separating studies by various covariants and let-one-out analyses showed consistency in direction and magnitude, except for the impact of cold spells on cardiovascular morbidity. In this study, two (33%) studies could render the pooled effects of cold spells on cardiovascular disease morbidity insignificant when left out from the analysis, indicating the instability of the result. This inconsistency may be attributed to the small amount of evidence present in the synthesis. More importantly, there needs to be a clear definition and reference periods for cold spells, which may cause significant heterogeneity and various estimated effects (40, 41).

The main advantages of our estimates of risk attributed to cold exposure are as follows. To our knowledge, this review is the first to focus on the impact of cold weather on cardiovascular disease and to analyze the influential factors that cause differences in terms of cold impact. Notably, we found that cold exposure had the most powerful impact on CHD and AAD. Moreover, we identified the climate zone as an essential influential factor in terms of the impact of ambient cold exposure on cardiovascular disease. We also provided strong evidence of the impact of cold exposure on cardiovascular disease with regard to both its direction and magnitude by conducting a wide-ranging search and analysis of the current evidence and carrying out a series of sensitivity analyses that attest to the robustness of our findings. However, our study still has some limitations to be addressed. First, we unbiasedly included relative peer-reviewed literature. However, the available studies are far from conclusive, and the quality of several studies is a matter of concern. These may inevitably affect the quality of the pooled results, which suggests the requirement for rigor and better instruments in future research. Second, we found a high heterogeneity among included studies. Although we employed a series of subgroup analyses and meta-regression, the source of heterogeneity was not identified. Therefore, we used a random-effects model to pool individual estimates in studies quantitatively. However, considering the undetected source of heterogeneity and confounders, caution should be exercised when interpreting these pooled effect estimates. Third, we referred to the methodology of the previous meta-analysis and chose the lag RRs with the maximum risks (18, 19), which could lead to mistakes in the pooled results. For example, such extracted data could inevitably induce higher estimated RRs. Moreover, temperatures in the following lag days could affect the results in the form of an unadjusted confounder. However, which lag RR gives a true picture of cold impact remains largely unknown, and it is unlikely to make the best choice on the basis of the available evidence. This suggests the need for future research on the relationship between the lag days and the impact of temperature. Fourth, despite the great amount of literature included in the pooled estimates, there were still a small number of estimates in some subgroups such as the mortality and morbidity of hypertensive disease and health outcomes in lower-middle-income-level countries.

This systematic review and meta-analysis used the most up-to-date data assessment method and included 159 pieces of literature on cold exposure and cardiovascular disease outcomes. This study provided updated evidence that cold exposure (both low temperatures and cold spells) could elevate the risk of cardiovascular disease–related mortality and morbidity. Findings from this review also highlight that people living in warmer climate zones and lower-middle-income countries are more susceptible to cold-induced cardiovascular diseases. This study helps evaluate the current risk factors for cardiovascular diseases and provides important implications for future healthcare prevention strategies and resource allocation for high-risk populations. Given the increases in the frequency and intensity of consecutive cold climatic extremes, urgent attention is called for to devise more successful strategies to reduce risks.

Acknowledgments

We sincerely acknowledge the support and help from statistical professor Ying-Yi Qin (from the Department of Statistical Analysis, NMU).

Funding Statement

This work was supported by the Funding of the Chinese Logistics Scientific Research Project (BHJ21J008) and the Key Construction Projects for Naval Academy Disciplines and Specialties (21JX017).

Data availability statement

The original contributions presented in the study are included in the article/Supplementary Material, further inquiries can be directed to the corresponding authors.

Author contributions

JF-F, Y-CX, Y-KW, and W-ZW contributed to the conception and design of the study. Y-FF organized the database. J-FF and L-YN performed the statistical analysis. J-FF wrote the first draft of the manuscript. XT, J-SC, Y-QL, and W-YL wrote sections of the manuscript. W-ZW and Y-KW reviewed and edited the manuscript. All authors contributed to the article and approved the submitted version.

Conflict of interest

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Publisher's note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Supplementary material

The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fcvm.2023.1084611/full#supplementary-material.

Datasheet1.pdf (1.9MB, pdf)

References

  • 1.Collaborators GBDRF. Global burden of 87 risk factors in 204 countries and territories, 1990–2019: a systematic analysis for the Global Burden of Disease Study 2019. Lancet. (2020) 396:1223–49. 10.1016/S0140-6736(20)30752-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Gasparrini A, Guo Y, Hashizume M, Lavigne E, Zanobetti A, Schwartz J, et al. Mortality risk attributable to high and low ambient temperature: a multicountry observational study. Lancet. (2015) 386:369–75. 10.1016/S0140-6736(14)62114-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Chen R, Yin P, Wang L, Liu C, Niu Y, Wang W, et al. Association between ambient temperature and mortality risk and burden: time series study in 272 main Chinese cities. Br Med J. (2018) 363:k4306. 10.1136/bmj.k4306 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Ma C, Yang J, Nakayama SF, Iwai-Shimada M, Jung CR, Sun XL, et al. Cold spells and cause-specific mortality in 47 Japanese prefectures: a systematic evaluation. Environ Health Perspect. (2021) 129:67001. 10.1289/EHP7109 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Shi J, Cui L, Wen K, Tian Z, Wei P, Zhang B. Trends in the consecutive days of temperature and precipitation extremes in China during 1961-2015. Environ Res. (2018) 161:381–91. 10.1016/j.envres.2017.11.037 [DOI] [PubMed] [Google Scholar]
  • 6.Luo Q, Li S, Guo Y, Han X, Jaakkola JJK. A systematic review and meta-analysis of the association between daily mean temperature and mortality in China. Environ Res. (2019) 173:281–99. 10.1016/j.envres.2019.03.044 [DOI] [PubMed] [Google Scholar]
  • 7.Guo Y, Li S, Zhang Y, Armstrong B, Jaakkola JJ, Tong S, et al. Extremely cold and hot temperatures increase the risk of ischaemic heart disease mortality: epidemiological evidence from China. Heart. (2013) 99:195–203. 10.1136/heartjnl-2012-302518 [DOI] [PubMed] [Google Scholar]
  • 8.Gibelin P. Do weather and seasons influence our heart?. Presse Med. (2015) 44:821–7. 10.1016/j.lpm.2015.07.004 [DOI] [PubMed] [Google Scholar]
  • 9.Manou-Stathopoulou V, Goodwin CD, Patterson T, Redwood SR, Marber MS, Williams RP. The effects of cold and exercise on the cardiovascular system. Heart. (2015) 101:808–20. 10.1136/heartjnl-2014-306276 [DOI] [PubMed] [Google Scholar]
  • 10.Kim HK, Sohn DW, Lee SE, Kim YJ, Oh BH, Park YB. Coronary blood flow after cold exposure and the effect of warm-up exercise. J Am Soc Echocardiogr. (2006) 19:386–90. 10.1016/j.echo.2005.11.006 [DOI] [PubMed] [Google Scholar]
  • 11.Phung D, Thai PK, Guo Y, Morawska L, Rutherford S, Chu C. Ambient temperature and risk of cardiovascular hospitalization: an updated systematic review and meta-analysis. Sci Total Environ. (2016) 550:1084–102. 10.1016/j.scitotenv.2016.01.154 [DOI] [PubMed] [Google Scholar]
  • 12.Wang Q, Li C, Guo Y, Barnett AG, Tong S, Phung D, et al. Environmental ambient temperature and blood pressure in adults: a systematic review and meta-analysis. Sci Total Environ. (2017) 575:276–86. 10.1016/j.scitotenv.2016.10.019 [DOI] [PubMed] [Google Scholar]
  • 13.Moghadamnia MT, Ardalan A, Mesdaghinia A, Keshtkar A, Naddafi K, Yekaninejad MS. Ambient temperature and cardiovascular mortality: a systematic review and meta-analysis. PeerJ. (2017) 5:e3574. 10.7717/peerj.3574 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Song X, Wang S, Hu Y, Yue M, Zhang T, Liu Y, et al. Impact of ambient temperature on morbidity and mortality: an overview of reviews. Sci Total Environ. (2017) 586:241–54. 10.1016/j.scitotenv.2017.01.212 [DOI] [PubMed] [Google Scholar]
  • 15.Ren C, Williams GM, Tong S. Does particulate matter modify the association between temperature and cardiorespiratory diseases? Environ Health Perspect. (2006) 114:1690–6. 10.1289/ehp.9266 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Bai L, Woodward A, Liu Q. Temperature and mortality on the roof of the world: a time-series analysis in three Tibetan counties, China. Sci Total Environ. (2014) 485-486:41–8. 10.1016/j.scitotenv.2014.02.094 [DOI] [PubMed] [Google Scholar]
  • 17.Liberati A, Altman DG, Tetzlaff J, Mulrow C, Gotzsche PC, Ioannidis JP, et al. The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate healthcare interventions: explanation and elaboration. Br Med J. (2009) 339:b2700. 10.1136/bmj.b2700 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Sun Z, Chen C, Xu D, Li T. Effects of ambient temperature on myocardial infarction: a systematic review and meta-analysis. Environ Pollut. (2018) 241:1106–14. 10.1016/j.envpol.2018.06.045 [DOI] [PubMed] [Google Scholar]
  • 19.Liu J, Varghese BM, Hansen A, Zhang Y, Driscoll T, Morgan G, et al. Heat exposure and cardiovascular health outcomes: a systematic review and meta-analysis. Lancet Planet Health. (2022) 6:e484–95. 10.1016/S2542-5196(22)00117-6 [DOI] [PubMed] [Google Scholar]
  • 20.Beck HE, Zimmermann NE, McVicar TR, Vergopolan N, Berg A, Wood EF. Present and future Koppen-Geiger climate classification maps at 1-km resolution. Sci Data. (2018) 5:180214. 10.1038/sdata.2018.214 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Higgins JP, Thompson SG, Deeks JJ, Altman DG. Measuring inconsistency in meta-analyses. Br Med J. (2003) 327:557–60. 10.1136/bmj.327.7414.557 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Yu W, Guo Y, Ye X, Wang X, Huang C, Pan X, et al. The effect of various temperature indicators on different mortality categories in a subtropical city of Brisbane, Australia. Sci Total Environ. (2011) 409:3431–7. 10.1016/j.scitotenv.2011.05.027 [DOI] [PubMed] [Google Scholar]
  • 23.Hurtado-Diaz M, Cruz JC, Texcalac-Sangrador JL, Felix-Arellano EE, Gutierrez-Avila I, Briseno-Perez AA, et al. Short-term effects of ambient temperature on non-external and cardiovascular mortality among older adults of metropolitan areas of Mexico. Int J Biometeorol. (2019) 63:1641–50. 10.1007/s00484-019-01778-y [DOI] [PubMed] [Google Scholar]
  • 24.Xu W, Thach TQ, Chau YK, Lai HK, Lam TH, Chan WM, et al. Thermal stress associated mortality risk and effect modification by sex and obesity in an elderly cohort of Chinese in Hong Kong. Environ Pollut. (2013) 178:288–93. 10.1016/j.envpol.2013.03.020 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Ren C, Williams GM, Morawska L, Mengersen K, Tong S. Ozone modifies associations between temperature and cardiovascular mortality: analysis of the NMMAPS data. Occup Environ Med. (2008) 65:255–60. 10.1136/oem.2007.033878 [DOI] [PubMed] [Google Scholar]
  • 26.Ha J, Kim H. Changes in the association between summer temperature and mortality in Seoul, South Korea. Int J Biometeorol. (2013) 57:535–44. 10.1007/s00484-012-0580-4 [DOI] [PubMed] [Google Scholar]
  • 27.Gao H, Lan L, Yang C, Wang J, Zhao Y. The threshold temperature and lag effects on daily excess mortality in Harbin, China: a time series analysis. Int J Occup Environ Med. (2017) 8:85–95. 10.15171/ijoem.2017.979 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Yang LT, Chang YM, Hsieh TH, Hou WH, Li CY. Associations of ambient temperature with mortality rates of cardiovascular and respiratory diseases in Taiwan: a subtropical country. Acta Cardiol Sin. (2018) 34:166–74. 10.6515/ACS.201803_34(2).20171101A [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Ebi KL, Mills D. Winter mortality in a warming climate: a reassessment. WIRES Climate Change. (2013) 4:203–12. 10.1002/wcc.211 [DOI] [Google Scholar]
  • 30.Keatinge WR. Cold exposure and winter mortality from ischaemic heart disease, cerebrovascular disease, respiratory disease, and all causes in warm and cold regions of Europe. The Eurowinter group. Lancet. (1997) 349:1341–6. 10.1016/S0140-6736(96)12338-2 [DOI] [PubMed] [Google Scholar]
  • 31.Wang X, Cao Y, Hong D, Zheng D, Richtering S, Sandset EC, et al. Ambient temperature and stroke occurrence: a systematic review and meta-analysis. Int J Environ Res Public Health. (2016) 13. 10.3390/ijerph13070698 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Keatinge WR, Coleshaw SR, Cotter F, Mattock M, Murphy M, Chelliah R. Increases in platelet and red cell counts, blood viscosity, and arterial pressure during mild surface cooling: factors in mortality from coronary and cerebral thrombosis in winter. Br Med J (Clin Res Ed). (1984) 289:1405–8. 10.1136/bmj.289.6456.1405 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.He F, Su W, Wu R, Li H, Lou L, Wu A, et al. The effect of cold exposure on serum cholesterol is dependent upon ApoE. J Therm Biol. (2021) 99:102972. 10.1016/j.jtherbio.2021.102972 [DOI] [PubMed] [Google Scholar]
  • 34.Brook RD, Weder AB, Rajagopalan S. “Environmental hypertensionology” the effects of environmental factors on blood pressure in clinical practice and research. J Clin Hypertens (Greenwich). (2011) 13:836–42. 10.1111/j.1751-7176.2011.00543.x [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Huang M, Yoo JK, Stickford ASL, Moore JP, Hendrix JM, Crandall CG, et al. Early sympathetic neural responses during a cold pressor test linked to pain perception. Clin Auton Res. (2021) 31:215–24. 10.1007/s10286-019-00635-7 [DOI] [PubMed] [Google Scholar]
  • 36.Zha YP, Wang YK, Deng Y, Zhang RW, Tan X, Yuan WJ, et al. Exercise training lowers the enhanced tonically active glutamatergic input to the rostral ventrolateral medulla in hypertensive rats. CNS Neurosci Ther. (2013) 19:244–51. 10.1111/cns.12065 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Valdes S, Doulatram-Gamgaram V, Lago A, Garcia Torres F, Badia-Guillen R, Olveira G, et al. Ambient temperature and prevalence of diabetes and insulin resistance in the Spanish population: Di@bet.es study. Eur J Endocrinol. (2019) 180:273–80. 10.1530/EJE-18-0818 [DOI] [PubMed] [Google Scholar]
  • 38.Ikaheimo TM, Jokelainen J, Hassi J, Hiltunen L, Keinanen-Kiukaanniemi S, Laatikainen T, et al. Diabetes and impaired glucose metabolism is associated with more cold-related cardiorespiratory symptoms. Diabetes Res Clin Pract. (2017) 129:116–25. 10.1016/j.diabres.2017.04.022 [DOI] [PubMed] [Google Scholar]
  • 39.Chen Z, Liu P, Xia X, Wang L, Li X. The underlying mechanisms of cold exposure-induced ischemic stroke. Sci Total Environ. (2022) 834:155514. 10.1016/j.scitotenv.2022.155514 [DOI] [PubMed] [Google Scholar]
  • 40.Ryti NR, Guo Y, Jaakkola JJ. Global association of cold spells and adverse health effects: a systematic review and meta-analysis. Environ Health Perspect. (2016) 124:12–22. 10.1289/ehp.1408104 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Wang L, Liu T, Hu M, Zeng W, Zhang Y, Rutherford S, et al. The impact of cold spells on mortality and effect modification by cold spell characteristics. Sci Rep. (2016) 6:38380. 10.1038/srep38380 [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Datasheet1.pdf (1.9MB, pdf)

Data Availability Statement

The original contributions presented in the study are included in the article/Supplementary Material, further inquiries can be directed to the corresponding authors.


Articles from Frontiers in Cardiovascular Medicine are provided here courtesy of Frontiers Media SA

RESOURCES