Association between the Consumption of Carbonated Beverages and Out-of-Hospital Cardiac Arrests of Cardiac Origin in Japan

Background: The consumption of carbonated beverages has been shown to increase the risk of developing metabolic syndrome. The associations between the consumption of carbonated beverages and left arterial dimension or left ventricular mass are believed to be likely related to the greater body weight of carbonated beverage drinkers relative to non-drinkers. Nonetheless, little is known about the association between the consumption of carbonated beverages and out-ofhospital cardiac arrests (OHCAs) in Japan. Methods: We compared the age-adjusted incidence of OHCAs to the expenditures on various beverages per person between 2005 and 2011 in the 47 prefectures of Japan. Patients who suffered from OHCAs of cardiac and non-cardiac origin were enrolled in All-Japan Utstein Registry of the Fire and Disaster Management Agency. The expenditures on various beverages per person in the 47 prefectures in Japan were obtained from data published by the Ministry of Health, Labour and Welfare of Japan. Results: There were 797,422 cases of OHCA in the All-Japan Utstein registry between 2005 and 2011, including 11,831 cases who did not receive resuscitation. Among these 785,591 cases of OHCA, 435,064 (55.4%) were classified as cardiac origin and 350,527 (44.6%) were non-cardiac origin. Non-cardiac origin inThe first two authors contributed equally to this work (M.T. and Y.S.). Corresponding author.


Introduction
Several reports have shown that the consumption of soft drinks has been increasing worldwide [1] [2].The consumption of large quantities of soft drinks has been shown to not only lead to hypertension and diabetes, but also to promote atherosclerosis [3]- [8].Furthermore, these drinks contain large amounts of advanced glycation endproducts which accelerate the progression of atherosclerosis [9].Some epidemiologic studies have shown a positive correlation between the consumption of these drinks and the incidence of cardiovascular disease (CVD) and stroke [10]- [12], while other reports have demonstrated that the intake of green tea and coffee reduces the risk and mortality of CVD in Japan [13]- [15].Carbonated beverages, or sodas, have frequently been demonstrated to increase the risk of metabolic syndrome and CVD, such as subclinical cardiac remodeling and stroke [16] [17].
The Fire and Disaster Management Agency of Japan launched a prospective, nationwide, population-based, cohort study in subjects who had an OHCA to evaluate the effect of the nationwide dissemination of public-access Automated External Defibrillators (AEDs) on the rate of survival among patients who had an OHCA, and the Japanese Circulation Society (JCS) Resuscitation Science Study (JCS-ReSS) Group had a suitable database in January 2005 [18].These data showed that in Japan, over 100,000 people experienced OHCA annually, and their survival rate within the first month thereafter was only 5%.Thus, OHCA is still a major public health problem in Japan.As for OHCA of cardiac origin, several studies have demonstrated that some factors (e.g., seafood consumption) reduce the risk of OHCA of cardiac origin [19].However, the association between the consumption of large amounts of carbonated beverages and fatal CVD, or out-of-hospital cardiac arrests (OHCA) of cardiac origin, is not clear.
In the present study, we used an ecological method, which referred to a geographical and temporal study of risk-modifying factors on health outcomes based on populations.We hypothesized that the consumption of large amounts of carbonated beverages might be related to the incidence of OHCA of cardiac origin.Therefore, we considered that the kinds of soft drinks, especially carbonated beverages, consumed in each prefecture, in addition to the total consumption of soft drinks per se in each prefecture, would be associated with OHCA of cardiac origin.We looked for associations between the average consumption of different kinds of soft drinks and the average incidence of OHCA between 2005 and 2011 in the 47 prefectures of Japan.

Subjects
Patients who suffered from OHCA of cardiac or non-cardiac origin (n = 435,064 and 350,527, respectively) and who were enrolled in the All-Japan Utstein Registry of the Fire and Disaster Management Agency between 2005 and 2011 were included in this analysis.Cardiac arrest was considered the cessation of cardiac mechanical activity, and was confirmed by the absence of a detectable pulse, unresponsiveness, and apnea.It included all patients who had an out-of-hospital cardiac arrest before the arrival of emergency medical services (EMS) personnel, were treated by EMS personnel, and were then transported to medical institutions.This reporting system has been implemented in all of the fire stations in Japan and includes all cases of OHCA.The data were registered by the Fire and Disaster Management Agency (FDMA) of Japan and changed to a suitable database by the Japanese Circulation Society Resuscitation Science Study (JCS-ReSS) Group.Whether the arrest was of cardiac or non-cardiac origin was determined clinically by the physician in charge, in collaboration with the EMS personnel, and was confirmed by a staff member at the FDMA.Patients were followed-up at 1 month, and 1 month survival, Cerebral Performance Category (CPC) score, and Overall Performance Category (OPC) score were checked by the EMS personnel and physicians.CPC scores are used to describe "good (1-2)" and "poor (3-4)" outcomes.The 1 is conscious and alert with normal function or only slight disability, the 2 is conscious and alert with moderate disability, the 3 is conscious with severe disability, the 4 is comatose or persistent vegetative state, and the 5 is brain dead or death from other causes.OPC scores are used to describe "good (1-2)" and "poor (3-4)" outcomes.The 1 is healthy, alert, capable of normal life, the 2 is moderate overall disability, the 3 is severe overall disability, the 4 is coma or vegetative state, and the 5 is death, certified brain dead or dead by traditional criteria.Data regarding the expenditures on beverages in the 47 municipalities were obtained from the Family Income and Expenditure Survey published by the Ministry of Internal Affairs and Communications of Japan The study protocol for analyses was approved by the Ethics Committee of Fukuoka University (FU-#00000403), Japan.

Age-Adjusted Incidence of OHCA
Using the Utstein Registry, we calculated the crude incidence of OHCA by determining the raw number of cases of OHCA by prefecture and then dividing these numbers by the population of the prefecture from 2005 to 2011 [21].The Japanese Model Population in 1985 was used as a standard population, and age-standardization was performed by a direct method.We used a popular method to adjust for age.The Japanese Model Population in 1985 is a standard population (Supplementary Table 1), similar to the World Standard Population reported by Doll and Segi.First, we calculated the age-specific incidence for five-year age groups per person-year by dividing the number of OHCA cases in each age group by the population in each age group in the 47 prefectures.Second, age-standardized incidences were calculated to allow comparison with the Japanese Model Population in 1985, with age-standardization conducted by a direct method.Age-adjusted incidence was calculated as Σ (the incidence of OHCA in the five-year age group x the size of the standard population in the five-year age group)/total standard population.Finally, we determined the average yearly age-adjusted incidence of OHCA by prefecture from 2005 to 2011.

Expenditures on Beverages
Data regarding the expenditures on beverages in the 47 municipalities were obtained from the Family Income and Expenditure Survey published by the Ministry of Internal Affairs and Communications of Japan [20].The survey unit was the household throughout the entire area of Japan.Data were obtained with four different questionnaires: Household Schedule, Family Account Book, Yearly Income Schedule and Savings Schedule.Enumerators filled in the Household Schedule with the number of household members, occupation and industry of earners, type of dwelling, etc.We considered 9 kinds of beverages: carbonated beverages, green tea, tea, coffee, cocoa, juice, fermented milk beverage, milk, and mineral water.Yearly average expenditures per household by prefectural capital city were obtained from this Survey.We calculated the expenditures on these drinks per person by dividing by the number of household members, and determined the averages from 2005 to 2011.

Statistical Analysis
The statistical analysis was performed using SAS software, version 9.4 (SAS Institute, Cary, NC, USA) at Fukuoka University.We used a t-test for continuous variables and chi-squared tests for categorical variables.The Spearman Rank Correlation Coefficient was used to evaluate associations between groups.The values are expressed as the mean ± standard deviation (SD).Statistical significance was defined as a p-value of less than 0.05.

Patient Characteristics in the All-Japan Utstein Registry
There were 797,422 cases of OHCA in the All-Japan Utstein registry between 2005 and 2011, including 11,831 cases who did not receive resuscitation.Table 1 shows the patient characteristics in the All-Japan Utstein Registry between 2005 and 2011, excluding the 11,831 in the no-resuscitation group: 785,591 cases of OHCA: 435,064 (55.4%) of cardiac origin and 350,527 (44.6%) of non-cardiac origin.Non-cardiac origin included cerebrovascular disease, respiratory disease, malignant tumor, and exogenous disease (4.8%, 6.1%, 3.5%, and 18.9%, respectively).Patients with OHCA of cardiac origin were significantly older, and had a lower incidence of male and a higher incidence of 1-month survival, cerebral performance category 1 or 2, and overall performance category 1 or 2 compared to OHCA of non-cardiac origin.The initial rhythms in OHCA of cardiac origin were significantly more likely to be ventricular fibrillation and pulseless ventricular tachycardia, and less likely to be pulseless electrical activity and asystole.We showed the data aggregated by prefecture (Supplementary Table 2).

Time Trends for the Incidence of OHCA and Expenditures on Carbonated Beverages
Figure 1 shows the yearly changes in the expenditures on carbonated beverages and the crude incidence of OHCA (total, cardiac and non-cardiac origin) in the 47 prefectures of Japan from 2005 to 2011.We performed a Spearman rank correlation coefficient analysis on the yearly increase in expenditures on carbonated beverages and the crude incidence of total OHCA, OHCA of cardiac origin, and OHCA of non-cardiac origin.The incidence of OHCA of both total and cardiac origin and the expenditures on carbonated beverages have been increasing yearly since 2005.We also investigated the associations between the yearly increase in the expenditures on carbonated beverages and the yearly increase in the crude incidence of total OHCA (p = 0.02), OHCA of cardiac origin (p = 0.005), and OHCA of non-cardiac origin (p = 0.3).

Incidence of OHCA of Cardiac and Non-Cardiac Origin and the Expenditures on Carbonated Beverages in the 47 Prefectures of Japan
Figure 2 shows the mean age-adjusted incidence of OHCA of cardiac and non-cardiac origin and the mean expenditures on carbonated beverages in the 47 prefectures of Japan between 2005 and 2011.Northern Japan might tend to show a high incidence of OHCA of cardiac origin and high expenditures on carbonated beverages.There was no certification using statistical analysis.

Correlations between the Nine Kinds of Beverage and the Incidence of OHCA of Cardiac and Non-Cardiac Origin in the 47 Prefectures
Table 2 shows the Spearman Rank Correlation Coefficients between the consumption of each type of beverage and the age-adjusted incidence of OHCA of cardiac origin.The expenditure on carbonated beverages was significantly associated with the age-adjusted incidence of OHCA of cardiac origin as also shown in Figure 3 (r = 0.302, p = 0.04), but not non-cardiac origin (r = −0.03,p = 0.8).On the other hand, the expenditures on other  beverages were not significantly associated with the age-adjusted incidence of OHCA of cardiac origin: green tea (r = −0.010,p = 0.94), tea (r = −0.145,p = 0.33), coffee (r = −0.088,p = 0.55), cocoa (r = −0.192,p = 0.20), juice (r = −0.140,p 0.35), milk (r = −0.134,p = 0.37), and mineral water (r = −0.081,p = 0.59).The expenditure on fermented milk beverages tended to be positively associated with OHCAs of cardiac origin (r = 0.25, p = 0.09).The variability of the data was also shown in Supplementary Table 3.

Discussion
In the present study, the incidence of OHCA of cardiac origin increased, and the total number of OHCAs reached over 780,000 during the observation period in Japan.The consumption of carbonated beverages also increased from 2005 to 2011, and the population in northern Japan tended to consume more carbonated beverages.
Although we found a significant positive relation between the consumption of carbonated beverages and the age-adjusted incidence of OHCA of cardiac origin, similar associations were not seen for the consumption of green tea, tea, milk, fermented milk beverages, coffee, cocoa, juice or mineral water.Epidemiological data have shown unequivocally that an increased intake of soft drinks is associated with metabolic syndrome, CVD and stroke [10]- [12].In our study, however, only the consumption of carbonated beverages was significantly and positively correlated with the incidence of OHCA of cardiac origin.As far as we know, this study is the first to demonstrate an association between the consumption of carbonated beverages and OHCA of cardiac origin.
Previous studies have demonstrated that the consumption of carbonated beverages and other soft drinks was positively associated with increases in blood sugar, insulin, triglyceride and low-density lipoprotein cholesterol, and these metabolic abnormalities promoted CVD [3]- [8] [10]- [12].In our study, however, only carbonated beverages, and not the other beverages, were positively correlated with the age-adjusted incidence of OHCA of cardiac origin.Thus, our results did not completely agree with the results of previous studies.Although it is still unclear why and how only the consumption of carbonated beverage was positively associated with the age-adjusted incidence of OHCA of cardiac origin, one possible explanation is that the acid contained in carbonated beverages might play an important role in this association.Some previous papers have shown that an increased risk of CVD was associated with the intake of not only soft drinks, but also diet drinks [22] [23].In these studies, an increase in sugar intake was shown to be related to an increase in metabolic abnormalities and to promote CVD, while other factors might also play important roles in promoting CVD.In one report, carbonated drinks were associated with a reduced telomere length in leukocytes, which is related to the cell cycle and cell aging [24].In addition, there was no association between non-carbonated soft drinks and telomere length.This suggests that carbonated acid might promote atherosclerosis through its association with cell aging.Therefore, we thought that the acid in carbonated beverages might play an important role in promoting atherosclerosis, and lead to the increase in the age-adjusted incidence of OHCA of cardiac origin observed in our study.
According to previous reports, the consumption of green tea and coffee is significantly and negatively associated with the incidence of CVD [13]- [15].In our study, green tea and coffee did not show a significant negative correlation with the age-adjusted incidence of OHCA of cardiac origin.Thus, our results did not completely correspond with these previous reports.There are several possible explanations for this discrepancy.First, the consumption of these beverages was expressed in terms of money spent and this may not accurately reflect their actual consumption.Second, not all cases of OHCA of cardiac origin are caused by CVD based on atherosclerosis.Since the OHCA database we used included all age groups in Japan, some cases may have been due to some other heart disease, such as arrhythmia or congenital heart disease, which are often observed in younger patients [25].In addition, some papers have reported that fermented milk beverages had a positive impact on blood pressure [26]- [28].Inconsistent with these findings, fermented milk beverages tended to increase the age-adjusted incidence of OHCA of cardiac origin in our study.Further studies will be needed to clarify whether these drinks actually increase the risk and incidence of OHCA of cardiac origin.

Limitations
Since this was an ecological study and the baseline characteristics in these surveys are different, we need to consider an ecological fallacy, and thus the results may not be completely accurate.In addition, an ecological study is an epidemiologic analysis in which the units of analysis are populations or groups of people rather than individuals.The results of this group study are suggestive, and are insufficient to prove any associations.Further individual studies will be needed to check these results.The data on the average consumption of various beverages in the 47 prefectures of Japan only included information for the respective prefectural capitals.However, this information should reflect the characteristics of the prefectures.Furthermore, the results were estimated from data on consumer spending rather than from a dietary questionnaire.Family Income and Expenditure Survey does not provide a detailed classification of carbonated beverages.Other confounding factors, such as the climate or lifestyle in the 47 prefectures, should also be considered.

Conclusion
In Japan, the consumption of carbonated beverages may be associated with the risk of OHCA of cardiac origin, although further individual studies will be needed to confirm this association.
[20].The populations in the 47 prefectures of Japan were obtained from the Annual Report on Current Population Estimates (2005-2011) published by the Ministry of Internal Affairs and Communications of Japan [21].

Figure 1 .
Figure 1.Yearly changes in the expenditures on carbonated beverages and the incidence of OHCA in the 47 prefectures of Japan from 2005 to 2011.The green line (left vertical axis) indicates the annual expenditures on carbonated beverages per person in the 47 prefectures between 2005 and 2011.The orange, red, and blue lines (right vertical axis) indicate the annual crude incidence of OHCA in the 47 prefectures in all patients and in those of cardiac and non-cardiac origin between 2005 and 2011, respectively.

Figure 2 .
Figure 2. Mean age-adjusted incidence of OHCA of cardiac origin (red map in the left panel), non-cardiac origin (blue map in the middle panel), and expenditures on carbonated beverages (green map in the right panel) in the 47 prefectures of Japan between 2005 and 2011.

Figure 3 .
Figure 3. Correlations between the age-adjusted incidence of OHCA and the expenditures on different kinds of beverages in the 47 prefectures of Japan.Red lines and circles indicate the correlation to OHCA of cardiac origin.

Table 2 .
Associations between the cost of kinds of drink and the age-adjusted incidence of out-of-hospital cardiac arrest of cardiac origin.

Table 3 .
The variability of using parameters in 47 prefectures of Japan.