Cardiac autonomic regulation following a 246-km mountain ultra-marathon: An observational study

Abstract

Physical exercise requires integrated autonomic and cardiovascular adjustments to maintain homeostasis. We aimed to observe acute posture-related changes in blood pressure, and apply a portable noninvasive monitor to measure the heart index for detecting arrhythmia among elite participants of a 246-km mountain ultra-marathon. Nine experienced ultra-marathoners (8 males and 1 female) participating in the Run Across Taiwan Ultra-marathon in 2018 were enrolled. The runners’ Heart Spectrum Blood Pressure Monitor measurements were obtained in the standing and supine positions before and immediately after the race. Their high-sensitivity troponin T and N-terminal proB-type natriuretic peptide levels were analyzed 1 week before and immediately after the event. Heart rate was differed significantly in the immediate postrace assessment compared to the prerace assessment, in both the standing (P = .011; d = 1.19) and supine positions (P = .008; d = 1.35). Postural hypotension occurred in 4 (44.4%) individuals immediately postrace. In 3 out of 9 (33.3%) recruited finishers, the occurrence of premature ventricular complex signals in the standing position was detected; premature ventricular complex signal effect was observed in the supine position postrace in only 1 participant (11.1%). Premature ventricular complex signal was positively correlated with running speed (P = .037). Of the 6 individuals who completed the biochemical tests postrace, 2 (33.3%) had high-sensitivity troponin T and 6 (100%) had N-terminal proB-type natriuretic peptide values above the reference interval. A statistically significant increase was observed in both the high-sensitivity troponin T (P = .028; d = 1.97), and N-terminal proB-type natriuretic peptide (P = .028; d = 2.91) levels postrace compared to prerace. In conclusion, significant alterations in blood pressure and heart rate were observed in the standing position, and postexercise (postural) hypotension occurred among ultra-marathoners. The incidence of premature ventricular complexes was higher after the race than before.

1. Introduction

The autonomic nervous system regulates key aspects of cardiovascular function, including contraction, blood pressure, and heart rate. Alterations in autonomic function following prolonged exercise have been observed in athletes competing in events ranging from 4 to 100 hours.^[1] Exercise-induced hypotension can be a significant symptom/event for autonomic failure.^[2] Profound changes in the mechanisms that result in postexercise hypotension can last approximately 2 hours in healthy individuals.^[3] With the increasing number of individuals participating in mountain ultra-marathon (MUM) running, races longer than the classical marathon distance (42.195 km), the number of individuals affected by postexercise hypotension has the potential to increase in the near future.

Although autonomic imbalance, either increased sympathetic activity or vagal tone, can trigger arrhythmias,^[4] there is currently no consensus on the nature of sports in which long-distance running triggers arrhythmias. Studies on elite athletes who had engaged in marathon running, cycling, or cross-country skiing showed that they were at a 4-fold to 15-fold increased risk of atrial fibrillation (AF) compared with the general, sedentary population.^[5–7] Conversely, utilized Holter recordings, Grab et al^[8] found that the prevalence of arrhythmias in male marathoners during and after a marathon event decreased. In a study by D’Ascenzi et al,^[9] among 301 50-km ultra-marathoners, postrace electrocardiography (ECG) signs of right heart overload were reported, but no arrhythmias.

The Heart Spectrum Blood Pressure Monitor (HSBPM) is a portable, noninvasive device that can detect changes in the pressure of the blood vessels of the heart. The heartbeat frequency can be analyzed to determine whether premature ventricular complex (PVC) signals or AF have occurred. Kao et al^[10] compared the diagnostic efficacy of the HSBPM to a 12-lead ECG in patients with AF. The sensitivity, specificity, positive predictive value, as well as negative predictive value, were 97%, 97%, 97%, and 97%, respectively. Thus, battery-operated HSBPM may be a practical device for evaluating the exercise-induced arrhythmias of athletes in outdoor environments.^[11]

During the 216-km Badwater Ultra-marathon in 2004, 4 out of 10 runners experienced high-sensitivity troponin T (hs-TNT) concentrations above the upper reference limit (URL), while 9 out of 10 runners showed an increased N-terminal proB-type natriuretic peptide (NT-proBNP) concentration.^[12] After the 118-km Penyagolosa Trails CSP115 race in 2014, 15 out of 30 finishers surpassed the URL for hs-TNT, while 26 out of 30 finishers exceeded the URL for NT-proBNP.^[13] The mechanisms underlying these findings may relate to myocardial processes, ranging from benign leakage from the cardiac myocyte membrane to myocyte necrosis.^[14–16] However, the significance of the elevated levels of these biomarkers in athletes remains uncertain.^[13,17]

We hypothesized that competing in the 246-km MUM may affect autonomic nervous system activity and trigger exercise-induced arrhythmias in runners. Therefore, we designed an observational study to investigate acute posture-related changes in blood pressure, employing HSBPM to assess the occurrence of PVC signals/AFs among the participants. In this study, we also aimed to assess acute changes in cardiac biomarkers after 246-km MUM.

2. Materials and methods

2.1. Study design and population

Thirty-one runners who participated in the event known as the Run Across Taiwan Ultra-marathon in 2018 were enrolled in this study. The study protocol was approved by the TMU-Joint Institutional Review Board (N201802069). Before the competition, all participants were required to complete a questionnaire on their demographic data and medical and training histories. Runners were excluded if they had a history of heart disease, renal dysfunction, seizures, or syncope of an unknown origin. Throughout the race, participants ingested fluids and food ad libitum. Ultimately, the data from 9 finishers were included in the analysis. In addition, cardiac biomarkers were analyzed in 6 of 9 participants. These athletes did not use performance-enhancing drugs (based on history taking and physical examination for the detection of signs indicative of concealed use of performance-enhancing drugs).^[18]

2.2. Race briefing and weather conditions

The competition began at 8 pm on April 13, 2018, and ended at 4 pm on April 15, 2018, and was comprised of a 246-km ultradistance race with a cumulative elevation gain of 4250 m. The race route is illustrated in Figure 1. During the competition, the weather was sultry with a temperature of 26°C in the lowlands during the night of April 13. However, the temperature gradually dropped to a single digit on the foggy and wind summit of Hehuan Mountain on April 14. The weather on April 15 was cool and breezy accompanied by light showers.

Figure 1.

Race route. The run track is from Taiwan’s west coast (7 m above sea level) rising to Wuling, Hehuan Mountain (3275 m above sea level), and descending to the east coast. The participants ran from the lowlands in the humid subtropical region to the high mountain summit.

2.3. Heart Spectrum Blood Pressure Monitor

The HSBPM (size: 6.5 × 13 × 6 cm; weight: 0.42 kg) (P2; OSTAR Meditech Corp., Taiwan) is a device approved by the Taiwan and U.S. Food and Drug Administration that can detect changes in the pressure of the heart’s blood vessels (Fig. 2). Blood pressure and heart rate were measured using an oscillometric method. Each heartbeat causes the heart to emit blood; subsequently, the sensor of the HSBPM on the arm detects the blood pressure and depicts the time-domain pressure wave. The time-domain pressure wave is converted into an energy-domain frequency wave by the device via fast Fourier transformation. Primary frequency peaks were observed when the waves were converted using fast Fourier transformation. When observing an abnormal frequency, related peaks other than the primary frequency peaks are considered heart noises, which is defined as the number of other spikes above 1/20 for each region. A scale factor of 1/20 was determined by removing the background noise from the clinical pretest results. We defined heart index I1 as the sum of noise in the first frequency region, heart index I2 as the sum of noise in the second frequency region, and heart index I3 as the sum of noise in the third frequency region: heart index = I1 + I2 + I3. A heart index of an isolated I1 ≥ 2 was indicative of PVC signal effects, while (I1 + I2 + I3) ≥ 5 and I1 ≥ 2 indicate AF. Further details of these devices are provided elsewhere.^[10]

Figure 2.

The Heart Spectrum Blood Pressure Monitor (A) and heart spectrum recordings (B). I = heart index.

The study protocol was repeated twice, namely before and immediately after the race. During each assessment period, the participants initially rested in the supine position and the HSBPM recording commenced while the individual remained stationary. The participants were then instructed to move to a motionless standing position, after which the HSBPM measurements were repeated.

2.4. Laboratory assessment

Blood samples (10 mL) were drawn from the antecubital vein using sterile techniques 1 week before and immediately after the race. The hs-TnT levels were measured using a hs-TnT assay (Roche Diagnostics, Indianapolis, IN) on a Cobas 8000 analyzer (lot no. 305646). The 99th percentile among the healthy participants was 14 ng/L. The NT-proBNP levels were measured using an electrochemiluminescence sandwich immunoassay (Elecsys ProBNP; Roche Diagnostics, Basel, Switzerland) with the Roche e601 system. The URL was 125 pg/mL for participants over the age of 75.

2.5. Statistical analysis

To test the normality of our data, the Shapiro–Wilk test (P > .05) and an inspection of the participants’ histograms, normal Q-Q plots, and box plots were used. Descriptive results are reported as median (interquartile range). McNemar Chi-squared test was applied for categorical data, and the dependent-samples t test (or Wilcoxon signed-ranks test, when appropriate) was applied for numerical data to evaluate the physiological and biochemical associations between the prerace and postrace periods. Fisher exact test was used to examine the significance of the association between arrhythmia and postural hypotension. The point-biserial correlation coefficient was used to evaluate the correlation between running speed and occurrence of arrhythmia and postural hypotension among runners. The commercially available statistical software SPSS (version 21.0, IBM Corp., Armonk, NY) was used for statistical analysis. Differences were considered statistically significant at 2-tailed P < .05.

3. Results

3.1. Race details and runners’ performance

Ninety-five runners competed in the race, of which 33 (34.7%) crossed the finish line. Twenty-five out of the 33 (75.8%) finishers were male. The average finishing time was 2437.2 minutes (range: 1802–2633 minutes). Thirty-one out of 95 (32.6%) competitors were willing to participate in our study. Ultimately, 9 out of 33 finishers (eight males and 1 female) completed the HSBPM measurements. The average age of the recruited runners was 45.7 years (range, 37–54 years). The running speeds and rankings of the 9 participants are summarized in Table 1. Of the 6 finishers who underwent biochemical assessments, 5 (83.3%) were men and 1 (16.7%) was a woman.

Table 1.

Personal data of the subjects.

Athlete	Sex	Age, yr	Running speed, km/h	Rank	Postural hypotension	Exercise-induced PVC signal
A	M	46	7.2	2/33	+	+
B	F	37	6.5	5/33	−	+
C	M	54	6.4	8/33	−	−
D*	M	41	6.0	17/33	+	+
E*	M	44	5.8	23/33	−	−
F	M	42	5.7	27/33	−	−
G	M	52	5.6	30/33	−	−
H	M	45	5.6	32/33	+	−
I*	M	50	5.6	33/33	+	−

Running performance of the subjects.

PVC = premature ventricular complex.

^*Runners didn’t assess the cardiac biomarkers.

3.2. Analysis of HSBPM parameters

The HSBPM tests revealed an increase in HR values that were significantly different in the immediate postrace assessment compared to the prerace assessment in both the standing (P = .011; d = 1.19) and supine positions (P = .008; d = 1.35). The exact measurement values are shown in Figure 3. None of the 9 participants met the AF criteria. Specifically, of these 9 runners, 3 (33.3%) exhibited isolated I1 ≥ 2 and PVC signal effects in the standing position in the postrun period (Table 1), and one of these 3 also exhibited isolated I1 ≥ 2 in the supine position. No statistically significant difference was observed between the prerace and postrace occurrence of exercise-induced PVC signals (standing, P = .083; supine, P = .317).

Figure 3.

Heart rate alterations before and after the 246-km mountain ultramarathon. Heart rate responses in standing (A) and supine (B) positions. The time profiles in the figures are presented by boxplots. The lower and upper boundaries of the box represent Q1 (25 percentile) and Q3 (75 percentile) of the data. Each dot represents the heart rate value of a subject, and the heart rate value of the same subject at different time points are linked by dotted lines. p = P-value; d = Cohen d effect size.

Comparing BP within the first 3 minutes of standing from the supine position, 4 out of the 9 runners (44.4%) presented postural hypotension with a drop of >20 mm Hg systolic and/or >10 mm Hg diastolic BP (Table 1). In general, a statistically significant difference (P = .045) was observed between the postrace and prerace values. Further analysis of the association between PVC signal and postural hypotension showed a P-value of .524.

3.3. Correlation analyses between running speed and the occurrence of PVC signal and postural hypotension

A strong positive relationship was observed between running speed and the occurrence of PVC signal (r = 0.697 [0.062–0.93], P = .037). However, only a small positive relationship was observed between running speed and the occurrence of postural hypotension (r = 0.114 [−0.595 to 0.723], P = .771).

3.4. Analysis of cardiac biomarkers

The hs-TnT and NT-proBNP levels were measured at 2 different time points: 1 week before the race and immediately after the race. The related values for the 6 participants are shown in Figure 4. A statistically significant increase was observed in both the hs-TnT (P = .028; d = 1.97) and NT-proBNP (P = .028; d = 2.91) levels postrace compared to prerace. Among the 6 runners, none exhibited hs-TnT values above 0.014 ng/mL before the run and 2 (33.3%) exhibited hs-TnT values above the reference interval after the run (Runner G: hs-TnT = 0.023 ng/mL; Runner H: hs-TnT = 0.017 ng/mL). All 6 (100%) runners exhibited NT-proBNP values below 125 pg/mL in the prerace period but above 125 pg/mL at postrace.

Figure 4.

Changes in plasma cardiac biomarkers during the 246-km mountain ultra-marathon. High-sensitivity troponin T (hs-TnT, 4A) and N-terminal proB-type natriuretic peptide (NT-proBNP, 4B) levels were measured at 2 different time points: prerace and immediately postrace. The time profiles in the figures are presented by boxplots. The lower and upper boundaries of the box represent Q1 (25 percentile) and Q3 (75 percentile) of the data. Each dot represents the hs-TnT or NT-proBNP level of a subject, and the hs-TnT or NT-proBNP level of the same subject at different time points are linked by dotted lines. p = P-value; d = Cohen d effect size.

4. Discussion

In this study, both a significant onset of postural hypotension and an increasing trend toward PVC signals were observed immediately after the race. The measurements for blood pressure and heart rate were similar to those previously reported after various ultramarathons. A faster running speed was associated with a higher incidence of postexercise PVC signals. The plasma levels of the cardiac biomarkers hs-TnT and NT-proBNP were found to be significantly higher after the 246-km run than before the race.

Postural hypotension was defined as a decrease in systolic blood pressure by at least 20 mm Hg or diastolic blood pressure by at least 10 mm Hg below the supine values, assuming an upright posture within 3 minutes. In the context of competitive ultra-running, Holtzhausen et al^[19] reported that asymptomatic postural hypotension developed in 68% of competitors in the “1990 80-km Karoo Marathon.” In our 246-km MUM finishers, 44.4% of runners presented postexercise (postural) hypotension. Currently, objective data on this phenomenon in MUMs is scarce. Thus, our data provide additional information regarding the physical effects of endurance sports.

Isolated PVCs, which are commonly diagnosed among athletes, are believed to be caused by imbalances in autonomic tone, or myocardial ischemia after exercise.^[20,21] Cavigli et al^[22] found exercise-induced isolated PVCs in 7% of 68 50-km ultra-marathoners, none of which showed nonsustained ventricular tachycardia before and after the event. In the present study, 33.3% of the runners had PVC signals detected after the MUM event, and none exhibited nonsustained ventricular tachycardia either.^[22] The incidence of exercise-induced PVC signals in our participants was higher than that in Cavigli study. One explanation for this may be that the physical demands of a 246-km MUM are higher than those of a 50-km Ultrarun.

To date, it is not clear whether long-term endurance exercise training is a trigger or primary cause of susceptibility to arrhythmia.^[23,24] Palatini et al^[25] compared the results of 24-hour continuous ECG monitoring of 40 endurance athletes (20 cyclists and 20 runners) with those of 40 healthy sedentary controls and found a higher prevalence of ventricular ectopy in the athletes (70% vs 55%). In the present study, our data showed a strong positive correlation between running speed and the occurrence of PVC signals, with 33.3% higher-ranked participants having PVC signals. These results suggest that there may be an underlying relationship between running speed and exercise-induced PVCs.

In the literature, there are pronounced changes indicative of acute exercise-induced right heart electrical and/or structural adaptation.^[26] Lord et al^[27] utilized right-sided ECG in 30 100-mile ultra-marathoners and reported significant changes in their P wave, ST segment, and T wave amplitude postrace. Christou et al^[28] examined 27 runners who finished a 246-km SPARTATHLON race and found a decrease in the left ventricular end-diastolic dimensions and an increase in the left ventricular wall thickness, as well as minimal dilatation and alteration in the systolic function of the right ventricle. In agreement with the above results, in the present study, increased NT-proBNP levels were observed in 6 athletes immediately after the 246-km MUM. In contrast, only 33.3% of the runners presented slight hs-TnT elevations after the race. Both were subclinical without any adverse physical symptoms. These results imply that the postrace elevation of cardiac biomarkers is a ventricular response to exercise, rather than due to myocardial damage.

In recent decades, the number of individuals participating in ultra-marathons has increased exponentially. Thus, the potential for serious medical issues is likely to increase in the near future. MUMs are usually located in remote environments where first-aid access may be extremely limited. Therefore, there is a need for the use of reliable health-monitoring devices with which to assess and treat athletes requiring medical evaluation. Furthermore, novel diagnostic tests have not been widely used for these events. Kao et al^[10] demonstrated that HSBPM can be used effectively for the detection of AF in hospital settings. Our results imply that HSBPM could be a useful assessment tool in outdoor environments.

Due to the fact that taking measurements during MUMs is challenging, this study has some limitations. First, the most important limitation was the small sample size, which resulted in decreased power to detect significant postrace changes in the parameters evaluated. As MUM is an extreme sport, it is difficult to enroll a large number of individuals. Second, our study did not include a control group, which raises questions about whether our small sample size reflects the occurrence of arrhythmia signals and blood pressure responses in the general population. Third, due to the lack of wireless technology in our HSBPM, we were unable to obtain physiological measurements during MUM. Finally, we did not perform follow-up HSBPM or cardiac biomarkers tests on our participants beyond the immediately postrace period. Further studies with multiple time-point samplings during and after exercise will be needed to determine the long-term effects of MUMs on ultrarunners. Although far from conclusive, our results provide valuable insights into the incidence of arrhythmia episodes in individuals participating in long-distance running sports.

5. Conclusion

Postexercise (postural) hypotension and PVC signals were observed in participants after a 246-km MUM. A faster average running speed was found to have a significant effect on the postexercise occurrence of PVC signals. Postrace elevations in the hs-TnT and NT-proBNP levels are physiological responses to exercise. Adaptation to endurance races may play a pivotal role in the distinct cardiovascular responses. In this context, HSBPM represent a tool with which to measure arrhythmia signals and blood pressure responses in individuals during ultra-marathons.

Acknowledgments

The authors thank all the ultra-marathon runners who participated in this study and all the doctors, nurses, and emergency medical technicians who provided professional care at this ultramarathon race. The authors also express gratitude to the Chinese Taipei Association of Ultra Runners, all of which assisted at the ultramarathon event. The authors further feel an immense gratitude toward Chia-Chun Lin from the Department of Post-Baccalaureate Veterinary Medicine, Asia University, Taiwan for helping in the generation of our graph.

Author contributions

Data curation: I-Hsun Tsai, Wei-Fong Kao, Li-Hua Li, Yu-Hui Chiu.

Writing—original draft: I-Hsun Tsai.

Conceptualization: Wei-Fong Kao, Lu-Chih Kung, Yu-Hui Chiu.

Formal analysis: Chorng-Kuang How, Yen-Kuang Lin, Yu-Hui Chiu.

Writing—review & editing: Chorng-Kuang How, Lu-Chih Kung, Yu-Hui Chiu, Ding-Kuo Chien, Wen-Han Chang.

Funding acquisition: Yu-Hui Chiu.

Methodology: Yu-Hui Chiu.

Project administration: Yu-Hui Chiu.