Out-hospital sudden cardiac arrest (OHCA) is a serious health problem. In Spain, the estimated annual incidence of OHCA is between 24,0001 and 50,000 cases,2 and the emergency medical services (EMS) play a crucial role in managing these patients. A cardiac origin of OHCA has been reported in over 80% of all cases,3 though some recent studies lower this percentage to about 60%.4

The measures adopted in recent years with the purpose of improving survival in patients with OHCA have focused mainly on the consolidation of EMS, the rapid activation of these services, witness-administered basic cardiopulmonary resuscitation, the diffusion and application of early defibrillation by both the EMS and non-medical emergency services, and the introduction of public access to defibrillation programs. However, adequate application of all these measures requires knowledge of the true situation of OHCA, determination of the response times of the installed devices, and knowledge of when and where they can be found.

Different studies5,6 have shown that in addition to myocardial infarction, other cardiovascular processes are also characterized by circadian variations in their appearance. The chronobiological analysis of a disease process allows us to determine its time of greatest incidence, and in contrast to acute myocardial infarction, this aspect has been little studied in the case of OHCA.7

Time variations have been described in cardiovascular event-triggering factors such as physical or psychological activity, environmental temperature, heart rate, blood pressure, sympathetic tone, fibrinolytic activity, platelet aggregation and even intrinsic peripheral vascular cell function.6,8–10

Time variations have also been demonstrated in the appearance of cardiac arrest, both during the day and over weekly or yearly periods.11,12 However, few studies have examined the influence of these variations upon patient survival,13–15 and none have used statistical methods specifically applied to chronobiological analysis. On the other hand, it has been shown that variations in EMS response times can influence mortality.16

Considering the above, analysis and knowledge of the chronobiological characteristics of OHCA, and the EMS response times, are of key interest as a basic source of information on the true situation of the process, and are useful in adopting preventive measures and in developing resource plans designed to offer a more adequate response.

The present study analyzes the chronobiological data and response times of the EMS referred to the OHCA cases registered in a Spanish Autonomous Community (Castilla y León), and their influence upon initial survival.

Of the 452,052 cases registered in the EMS database of Castilla y León during the 18months study period, we initially selected a total of 2018 cases corresponding to patients who had suffered cardiac arrest. Subsequently, application of the exclusion criteria led us to discard the following cases: 218 traumatisms, 123 other external causes, 251 non-cardiac medical causes, 24 patients under 18years of age, and 116 cases with insufficient data support. The final study sample thus comprised 1286 cases, the general characteristics of which are shown in Table 1.

The univariate analysis revealed statistically significant differences in the form of older patient age, more frequent presentation in public places, and a greater frequency of passerby or PC personnel alert activation, in the cases occurring in the first and last of the three periods of the day (8–16h and 16–24h). In contrast, the cases manifesting in the nighttime hours (0–8h) involved comparatively younger patients, with a greater incidence in the home, and a greater frequency of alerts by relatives or cohabitating witnesses. There were no statistically significant differences among the three periods in terms of patient gender, first monitored rhythm, outcome or time intervals between alert and patient care at the alert site or team activation and patient care at the site (Table 2).

The distribution of the median time intervals between the alert call (dial 112) and EMS care, and between emergency team activation and care at the alert site is shown in Figs. 1 and 2.

Figure 1.

Distribution of median time intervals between the alert call (dial 112) and emergency team care.

(0.1MB).

Figure 2.

Distribution of median time intervals between activation of the emergency team and care at the alert site.

(0.09MB).

The chronobiological analysis based on the simple cosine test in seek of 24-h rhythms revealed a global acrophase at 11:16h, and a global batyphase at 5:20h (Fig. 3). This was concurrent to the demonstration of a statistically significant 24-h or circadian harmonic rhythm (p=0.000). The grouping of cases over the hours of the day showed clear peaks, with a greater number of events at 10, 14 and 20h.

Figure 3.

Case curve (standardized) distributed over the hours of the day.

(0.19MB).

The chronobiological analysis based on the simple cosine test in seek of 7-day rhythms revealed a global acrophase on the third day of the week (Wednesday), and a batyphase on day 7 (Sunday). This was concurrent to the demonstration of a statistically significant 7-day or circaseptan rhythm (p=0.019). The grouping of cases over the days of the week showed two peaks with a greater number of events, corresponding to Wednesday and Saturday (Fig. 4).

Figure 4.

Distribution of cases during the days of the week.

(0.11MB).

There were no statistically significant differences in the chronobiological analysis of daily rhythm, distinguishing between the cases of ROSC and death (Fig. 5): ROSC, acrophase 11:28h, batyphase 5:03h; death, acrophase 11:12h, batyphase 5:24h (p=0.963). The chronobiological analysis of weekly rhythm likewise revealed no significant differences (p=0.879).

Figure 5.

Case curves (standardized and adjusted data) of ROSC and death, distributed over the hours of the day.

(0.14MB).

In the present study, based on a large population database, circadian variation was observed in the occurrence of OHCA, with no differences in relation to the initial outcome of the cases. A prevalent peak was recorded in the first hours of the morning, with another peak in the afternoon, as well as a clear decrease in the nighttime hours–in coincidence with the observations of other authors.

Regarding the analysis of the time variations in the presentation of OHCA, it should be noted that although a circadian rhythm had been described in 1987,6 very few further studies have been published,7,8,11–14 and none have been conducted in our particular setting to date. In a recently published study in Asia,18 a daily rhythm in appearance of the disorder has been reported–with two peak incidences (at 8:00h and at 19:00h).

Other publications on time variations in cardiovascular events have examined the hypotheses proposed to explain such variations. In this context, certain triggering physiopathological factors have been suggested to intervene at certain times of the day, including sympathetic tone or platelet activity.19,20 Other purported triggering factors comprise different types of physical activity, and changes in environmental temperature.9,10

On the other hand, a recent study21 whose results have not been confirmed in our work reports that OHCA survivors–in contrast to non-survivors–present a circadian rhythm in the appearance of the disorder, with an increased frequency between 18 and the 22h.

Thus, a large morning peak is noted, with a less marked afternoon peak, confirming the need to adopt preventive and therapeutic measures (“chronotherapy”) fundamentally in patients at risk, and suited for adjustment in time to attenuations in the triggering phenomena.

Regarding the time analysis of the differences observed in the course of the days of the week, our data do not coincide with the results of Brooks et al.22 and Gruska et al.23, who reported an increased incidence on Mondays and Saturdays, respectively, while in our series the peak incidence corresponded to Wednesdays. On the other hand, there were no differences in relation to outcome according to the day of the week on which OHCA occurred. This lack of agreement among the different studies referred to the weekly variations of OHCA points to the need for larger and more specific studies capable of clarifying this issue.

The more frequent presentation of OHCA in the home of the victims, witnessed by relatives or cohabitants during the nighttime period (between 0h and 8h), would justify the lesser tendency towards ROSC, probably due to a lesser availability of people trained in life support or of medical resuscitation teams in this particular scenario.

The data referred to the time intervals between the alerting call, activation of the EMS team, and the time of care at the alert site coincide with those published elsewhere.16 Although there were no significant differences among the three time periods analyzed, the intervals were longer in the period between 0h and 8h. This fact, and the observation of a lesser ROSC rate in that same period, suggest that delays in processing the alert call, in activating the EMS team, and in providing care at the alert site could have an influencing effect, along with other factors such as the known lesser incidence of rhythms amenable to defibrillation in OHCA occurring in the home.24 In this context, studies comparing the intervals of different EMS involving similar settings and characteristics would be needed to establish guiding references.

Our study offers information on the EMS activation circuits in cases of OHCA and on the time variations in the appearance of the disorder and in the times to intervention. Such data can be used to analyze and project new strategies for dealing with OHCA; in this context, specific registries are absolutely necessary in our setting,25 in order to confirm the results obtained. Emphasis also should be placed on the need to implement the first two links in the survival chain. In Spain, the response system has been developed in a generalized manner through the EMS, offering demonstrated quality and competence.26 However, much work remains to be done in reference to early activation of the EMS and to the provision of basic life support on the part of witnesses of the event.27 In this context, increased social awareness is needed, with the development of comprehensive training programs targeted to the general population.28

Lastly, regarding the limitations of the study, mention must be made of the lack of information on the previous health condition of the patients, the time interval between the event and the alert call (dial 112), the performance or not of resuscitation maneuvers by the witnesses, patient survival and neurological condition at hospital discharge, and possible data introduction errors, and the observational and retrospective nature of the study.

The authors declare no conflicts of interest.