Physiology & Biochemistry Thieme

Influence of -induced on Catecholamine Release and Cardiovascular Dynamics

Authors 53127 Bonn 1 1 1 Lars Eichhorn , Felix Erdfelder , Florian Kessler , Germany 2 2 3 Ramona C. Dolscheid-Pommerich , Berndt Zur , Uwe Hoffmann , Tel.: + 49/228/287 14114, Fax: + 49/228/287 14125 1 4 Richard K. Ellerkmann , Rainer Meyer [email protected]

Affiliations 1 Klinik und Poliklinik für Anästhesiologie und Operative Abstract Intensivmedizin, Universitätsklinikum Bonn, Bonn, Germany Prolonged breath-hold causes complex compensatory mechanisms 2 Institut für Klinische Chemie und Klinische Pharmakologie, such as increase in blood , redistribution of blood flow, and Universitätsklinikum Bonn, Bonn, Germany bradycardia. We tested whether apnea induces an elevation of catecho- 3 Institut für Physiologie und Anatomie, Deutsche Sporthochschule lamine- in well-trained apneic divers. Köln, Köln, Germany 11 apneic divers performed maximal dry apnea in a horizontal position. 4 Physiologisches Institut 2, Rheinische Friedrich-Wilhelms-Univer- Parameters measured during apnea included , ECG, and sität Bonn, Bonn, Germany central, in addition to peripheral hemoglobin oxygenation. Peripheral arterial hemoglobin oxygenation was detected by pulse oximetry, Key words whereas peripheral (abdominal) and central (cerebral) tissue oxygena- apnea, blood pressure, catecholamine, hypoxia, NIRS, SpO2 tion was measured by Near Infrared Spectroscopy (NIRS). Exhaled O2 accepted after revision 18.04.2016 and CO2, plasma norepinephrine and epinephrine concentrations were measured before and after apnea. Bibliography Averaged apnea time was 247 ± 76 s. Systolic blood pressure increased

DOI http://dx.doi.org/10.1055/s-0042-107351 from 135 ± 13 to 185 ± 25 mmHg. End-expiratory CO2 increased from Published online: July 25, 2016 | Int J Sports Med 2017; 38: 85–91 29 ± 4 mmHg to 49 ± 6 mmHg. Norepinephrine increased from 623 ± 307 © Georg Thieme Verlag KG Stuttgart · New York to 1 826 ± 984 pg ml − 1 and epinephrine from 78 ± 22 to 143 ± 65 pg ml − 1 ISSN 0172-4622 during apnea. reduction was inversely correlated with in- creased norepinephrine (correlation coefficient − 0.844, p = 0.001). Correspondence Central (cerebral) O2 desaturation was time-delayed compared to pe- Dr. Lars Eichhorn, MD ripheral O2 desaturation as measured by NIRSabdominal and SpO2. Universitatsklinikum Bonn Klinik und Poliklinik für Increased norepinephrine caused by apnea may contribute to blood Anästhesiologie und Operative Intensivmedizin shift from peripheral tissues to the CNS and thus help to preserve cer- Sigmund-Freud-Straße 25 ebral tissue O2 saturation longer than that of peripheral tissue.

Introduction Prolonged apneic diving has been performed for centuries, e. g., sympathetic nerve activity (MSNA). MSNA leads to peripheral va- for harvesting sponges, pearl shells and seafood [16]. In recent dec- soconstriction and thereby to hypertension [7, 20, 22, 28] and brad- ades, apneic diving has become more popular as a recreational ycardia [11, 15, 38] during breath holding. On the contrary, isolat- sport. Interestingly, physiological changes during apneic diving are ed hypoxia is accompanied by vasodilation and an increase in heart not yet completely understood. rate [1, 6, 7]. Bradycardia is described as an -saving mecha- To protect hypoxia sensible organs in case of extended breath- nism [2, 4, 11] and may be part of the diving response, but barore- hold, complex compensatory mechanisms exist to withstand apnea flex-activation due to increased blood pressure following acute hy- caused hypoxia [17]. The so-called diving response includes brad- percapnia [24] and stimulated chemoreceptors also seem to play ycardia [36], vasoconstriction [28], elevated mean arterial blood an important role in influencing bradycardia [29, 30]. pressure (MAP) [18], and increased cerebral blood flow (CBF) The aim of the present study was to characterize the physiolog-

[23, 35, 44]. The latter is mainly caused by increased pCO2 [26, 37]. ical effects of prolonged apnea by recording heart frequency, blood Redistribution of blood aims to maintain adequate oxygen supply pressure, central and peripheral oxygen saturation, and endogen to hypoxia sensible organs such as the brain [14, 23, 29]. Addition- catecholamine concentrations. We hypothesized that dry apnea ally, the mentioned compensatory mechanisms are enhanced by might be linked to increased catecholamine levels helping to ex- water immersion of the face [3, 15, 16, 38]. plain inter-individual differences in the amount of heart rate de- The commonly seen bradycardia in apneic divers is more diffi- crease following apnea. cult to explain. Hypoxia and are known to increase

Eichhorn L et al. Influence of Apnea-induced Hypoxia … Int J Sports Med 2017; 38: 85–91 85 Physiology & Biochemistry Thieme

Material and Methods 250 150 start of apnea end of apnea

The design of this study was approved by the local ethics commit- abdominal S

IR 200 tee of the University of Bonn. Our study conforms to the ethical /N 2 standards in sport and exercise science research [21]. 11 static ap- 100 150 2 neic divers were included in the study measuring various physio- HR/SpO O logical parameters during the complete time course of apnea. To rS 100 clearly distinguish apneic-induced effects (i. e., hypoxia and hyper- 50 capnia) from bradycardia-inducing effects such as face immersion 50 [25] or immersion, we chose a static dry apnea setup in a horizon- tal position. Apnea by the individual subject was performed as long 0 blood pressure [mmHg/ as possible. 0 200 400 time [sec] 15 s before initiation of apnea, one breath was analyzed for O2- systoic blood pressure cerebral rSO2 SpO2 [%] and CO2-levels and taken as a baseline. For this, participants ex- NIRSabdominal heart rate [bpm] haled air in a mask connected to the Dräger Scio 4 Oxi plus (Dräger Medical AG&Co.KG, Lübeck, Germany). Immediately ▶Fig. 1 Raw data of one subject. Total apnea time was 411 s. after apnea, the first 5 breaths were analyzed for pO2 and pCO2 Subject exhibited an earlier decrease in NIRSabdominal and SpO2 than breath by breath, and the mean was taken as post-apnea measure- in cerebral rSO2. This subject desaturated steadily (defined as ment. fall > 2 % of baseline-level) in SpO2 after 181 s and in rSO2 after 240 s. Desaturation of NIRS started after 21 s. Re-saturation was Heart rate was continuously measured by a 5-lead electrocardi- abdominal observed in NIRS after 19 s, in rSO after 5 s and in SpO after ® abdominal 2 2 ogram (Dräger monitor system: Infinity M540 Monitor and Infin- 29 s. ity M500 Docking Station). Peripheral oxygen saturation (SpO2) was measured Masimo (SHP ACC MCABLE-Masimo Set, OEM part- ners of Dräger Medical AG&Co.KG, Lübeck, Germany). Central (cer- creasing cerebral rSO2 vs. SpO2 and NIRSabdominal values following ebral) oxygen saturation (rSO2) and peripheral tissue oxygen satu- apnea onset and the corresponding increases following restart of ration (NIRSabdominal) were measured by Near Infrared Spectrosco- were investigated using Repeated Measures ANOVA py (NONIN Medical’s EQUANOX™, Model 7 600 Regional Oximeter with Greenhouse-Geisser correction followed by Bonferroni’s Mul- System, Plymouth, USA). NIRS diodes (EQUANOX Advance™ Sen- tiple Comparison Test. sor, Model 8004CA, suited for measuring cerebral and somatic ox- Total increase of end-expiratory CO2 measurements and total ygen-saturation) were placed on the right forehead and above the decrease of end-expiratory O2 were tested for significance. This dif- right musculus obliquus externus abdominis (2 cm above iliac ference was investigated using 2-tailed paired t-tests after check- crest). ing for violation of the distributional assumptions for increase of

Changes in blood pressure were measured beat-to-beat using CO2 (Shapiro-Wilk W = 0.920, p = 0.318), as well as for decrease of ® a finger-cuff blood pressure sensor on the right hand (Infinity O2 (Shapiro-Wilk W = 0.947, p = 0.611). CNAP™ SmartPod® system, Dräger Medical AG&Co.KG, Lübeck, The statistical association between changes in heart rate (i. e., Germany). heart rate prior to apnea – heart rate at the end of apnea) and NONIN Medical’s EQUANOX™ calculates the normalized hemo- serum norepinephrine/epinephrine (i. e., catecho- globin index (HBI) from the NIRS recordings. HBI may be taken as lamine concentration prior to apnea – catecholamine concentra- a hint for increased tissue [41]. tion at the end of apnea) was assessed using Spearman’s rank cor- For catecholamine analysis, venous blood samples were drawn relation analysis. Changes in heart rate and systolic blood pressure from every participant 3 min prior to and directly after apnea. Anal- were calculated in the same way. yses were performed in a specialized laboratory, which was reac- The continuous 5-lead ECG was stored automatically with a tem- credited by the German accreditation body DAkkS (accreditation poral resolution of 1 s, therefore a beat-to-beat variability in R-R in- numbers D-ML-13403-01-00 and D-PL-13403-01-00). terval was not possible. We decided to compare the heart rate de- viation in each participant during the first minute of apnea with Data analysis methods that of the last, because both are not influenced by the respiratory Changes in MAP during apnea were investigated using Repeated sinus arrhythmia. This comparison was performed by a 2 tailed, Measures ANOVA with Greenhouse-Geisser correction followed by paired t-test. Dunnett’s Multiple Comparison Test comparing the MAP at 20, 40, All results are given with standard deviation ( ± ) and statistical 60 and 100 % of the total apnea time with the MAP before onset of analyses were performed using SPSS 22 software (IBM SPSS, Chi- apnea. cago/IL, USA).

The start point of cerebral rSO2, NIRSabdominal and SpO2-decrease during apnea was defined as a drop of > 2 % compared to stable pre-apneic measurements followed by a monotonic decreasing Results time course. After restart of breathing, the point of increase (cer- An original recording displaying heart rate (HR), systoloic arterial ebral rSO2, NIRSabdominal and SpO2) was defined as steadily increas- pressure (SAP), cerebral rSO2, NIRSabdominal, and SpO2-saturation of ing values in the same way. Mean differences of time delays for de- the subject with the longest apnea time is shown in ▶Fig. 1.

86 Eichhorn L et al. Influence of Apnea-induced Hypoxia … Int J Sports Med 2017; 38: 85–91 ▶Table 1 Subjects physiologic details.

Subject Age Height BMI Heart rate at Heart rate at DAP at end SAP at end Total Gender [kg] [cm] start of end of apnea of apnea of apnea apnea apnea [bpm] [bpm] [mmHg] [mmHg] time [s] 1 30 85 186 24.6 68 74 95 161 411 male 2 30 70 174 23.1 80 83 112 158 248 male 3 50 82 183 24.5 70 57 135 223 226 male 4 38 68 176 22.0 71 72 119 212 335 female 5 37 97 185 28.3 90 35 122 204 248 male 6 37 70 189 19.6 56 46 107 180 319 male 7 32 79 180 24.4 88 89 101 161 189 male 8 41 78 179 24.3 70 76 115 176 193 male 9 53 111 193 29.8 67 57 134 218 200 male 10 63 65 173 21.7 70 73 110 158 124 male 11 47 88 183 26.3 79 56 100 183 226 male Ø 42 81 182 24.4 73.5 65.3 114 185 247 SD 10 13 6 2.8 9.4 15.6 13 25 76

As expected, SAP started to rise after about 150 s reaching a max- imum of more than 200 mmHg after the end of apnea. HR howev- 250 er stayed constant throughout the time course of apnea, but in- 200 creased clearly after restart of breathing. SpO2 usually taken as a measure for peripheral O2-saturation of hemoglobin started to de- 150 crease continuously after 181 s, whereas NIRSabdominal started to decrease shortly after starting breath-hold. Cerebral rSO2 initially increased by 10 % (compared to baseline levels) and remained at mmH g 100 100 % of baseline until 240 s after onset of apnea (▶Fig. 1). All data reported in the following are based on a group of 11 ex- 50 perienced static apneic divers of both genders (▶Table 1). The av- erage apnea time was 247 ± 76 s. A significant elevation of mean 0 arterial pressure (MAP) was seen at 40 % of total apnea time with 020406080 100 % of maximal apnea time continuously rising values until onset of breathing (ANOVA with systolic blood mean arterial diastolic blood Greenhouse-Geisser correction (2.496) F = 42.037, p < 0.001; Dun- pressure pressure pressure nett’s Multiple Comparison Test (20 %: not significant, 40 %: p < 0.05, 60–100 %: p < 0.001) (▶Fig. 2). By the end of breath hold ▶Fig. 2 An average blood pressure course of all participants the diastolic blood pressure (DAP) had increased from 80 ± 11 to (n = 11) plotted vs. normalized apnea time. For orientation, baselines are incorporated in diastolic arterial pressure (DAP) and systolic 114 ± 13 mmHg and the systolic blood pressure had increased from arterial pressure (SAP). * p < 0.05 vs. baseline. 135 ± 13 to 185 ± 25 mmHg. HR did not decrease significantly for all participants during dry apnea performed in this study − 1 − 1 (74 ± 9 min at start; 65 ± 16 min at the end). values increased with a significant delay [SpO2: 32 s (SD = 4 s) and

Following apnea, mean end-expiratory CO2 values increased sig- NIRSabdominal: 19 s (SD = 7 s); ANOVA with Greenhouse-Geisser cor- nificantly from baseline level of 29 ± 4 mmHg to 49 ± 6 mmHg at the rection (1.495; F = 80.229, p < 0.001; Bonferroni’s Multiple Com- end of apnea (t(10) = 16.088, p < 0.001, r² = 0.963). These findings parison Test p < 0.001 for all pairs) (▶Fig. 3b). were expected and characterize the quality of apnea performed in The normalized HBI recorded from the forehead did not de- our cohort of well-trained breath-hold divers. Following apnea, crease before 80 % of maximal apnea time, whereas the normalized mean end-expiratory O2 values decreased significantly from a base- HBIabdominal fell steadily from the beginning of breath hold (▶Fig. line level of 113 ± 4 mmHg to 50 ± 17 mmHg at the end of apnea 4). This may be regarded as a sign of a preferential cerebral perfu- (t(10) = 12.011, p < 0.001, r² = 0.935). sion compared to abdominal perfusion. To further characterize apnea, we recorded de- and re-satura- Averaged catecholamine-levels determined from blood sam- tion of peripheral arterial blood and abdominal vs. cerebral tissue ples of all subjects showed a significant increase at the end of apnea − 1 oxygenation. NIRSabdominal started to decrease 72 ± 48 s after onset (epinephrine from 78 ± 22 to 143 ± 65 pg ml ; norepinephrine − 1 of breath-hold, SpO2 began to fall significantly later, i. e., after from 623 ± 307 to 1826 ± 984 pg ml ) (▶Fig. 5a, b). Changes in

158 ± 22 s and cerebral rSO2 after 185 ± 26 s (▶Fig. 3a). After re- heart rate were significantly correlated to the increase in norepi- start of respiration, cerebral rSO2 values increased with a mean nephrine concentration during apnea (Spearman’s rank correlation delay of only 9 s (SD = 3 s), whereas SpO2 values and abdominal NIRS coefficient − 0.844, p = 0.001) while the correlation with epineph-

Eichhorn L et al. Influence of Apnea-induced Hypoxia … Int J Sports Med 2017; 38: 85–91 87 Physiology & Biochemistry Thieme

ab 250 50

200 40 ] 150 ] 30 time [s

100 time [s 20

50 10

0 0

NIRSabdominal SpO2 cerebral rSO2 NIRSabdominal SpO2 cerebral rSO2

▶Fig. 3 a Mean time delay between beginning of apnea and start of decrease of cerebral rSO2, SpO2 and NIRSabdominal (n = 11). * p < 0.001 for all pairs. b Mean time delay between restart of breathing and start of increase of cerebral rSO2, SpO2 and NIRSabdominal (n = 11). * p < 0.001 for all pairs.

cerebral O2 desaturation in comparison to peripheral O2 desatura- 110 tion as measured by NIRSabdominal and SpO2.

Blood pressure, apnea duration and CO2 According to the literature, apnea causes an increase in blood pres- 100 sure [17]. The increase in MAP detected in this study was 41 mmHg. Compared to other studies our measured MAP increase was higher line-value

se (31 mmHg at Cross et al. [12], 32 mmHg at Heusser et al. [22], and

90 17 mmHg at Perini et al. [36]). The mean apnea time lasted % of ba 247 ± 76 s, which is longer than those reported by the mentioned studies (199 ± 11 s [12] and 210 ± 70 s [36]) for trained divers and in the same range as the 234 ± 66 s published by Heusser et al. [22] 80 for elite divers. Thus, the divers in this study may be classified as 0420 06080 100 % of maximal apnea time well-trained in the field of static apnea and the detected apnea de- pendent compensatory mechanisms can be classified as relatively rSO2 NIRSabdominal distinct. A significant increase in blood pressure was seen after 40 % of individual maximal apnea time. This is in parallel with other stud- ▶Fig. 4 Time course of HBI. HBI levels and total apnea time are normalized to 100 % for better comparability. ies, showing a bi- or even tri-phasic change in blood pressure and bradycardia [11, 27, 36]. Although hyperventilation was not allowed for participants, rine was non-significant (Spearman’s rank correlation coeffi- pre-apneic levels of CO2 were lowered [13]. This may be caused by cient − 0.122, p = 0.721). Changes in heart rate were correlated inspiration and stretching exercises that all divers performed be- with increase in systolic pressure (Spearman’s rank correlation co- fore onset of apnea to adapt their pulmonary stress sensors to max- efficient − 0.5701, p = 0.0705). imal inspiration. Deeper in- and expiration as performed in these

As increasing catecholamine-levels should also influence HR var- exercises result in lowered CO2-levels and might also contribute to iability, we calculated the standard deviation for the heart rate of longer apnea times. Nevertheless, a CO2-level of 49 ± 6 mmHg, an each participant during the first and last minute of apnea. This may O2-level of 50 ± 17 mmHg, and an increase of 40 % in MAP meas- be taken as a simplified measure for HR variability. The mean SD of ured at the end of apnea demonstrate the expected changes under heart rates decreased significantly from 7.59 ± 3.19 bpm during the apnea. first minute to 3.96 ± 3.02 bpm during the last minute of apnea (t(10) = 2.865, r2 = 0.451, p = 0.017). Differential modulation of blood flow following apnea The diving response caused by apnea is mostly described as periph- Discussion eral vasoconstriction due to sympathetic activity, resulting in hy- In our study we investigated the change of blood pressure, catecho- pertension and vagally induced bradycardia with reduction of car- lamine-concentrations and other physiological parameters during diac output [29]. It is believed that the diving response aims to re- dry apnea (i. e., HR, SpO2, cerebral rSO2, NIRSabdominal, exhaled duce oxygen consumption by inducing peripheral vasoconstriction gases). We could detect an increase in blood pressure and serum [2, 4, 11]. This leads to redistribution of blood flow and ensures catecholamines in all volunteers. The decrease in heart rate follow- ­adequate oxygen supply for hypoxia sensible organs such as CNS ing apnea was inversely correlated with norepinephrine concentra- [23]. Consequently, our data revealed a faster decrease in periph- tions at the end of apnea. Moreover we measured a time-delayed eral O2 saturation (NIRSabdominal and SpO2) compared to cerebral O2

88 Eichhorn L et al. Influence of Apnea-induced Hypoxia … Int J Sports Med 2017; 38: 85–91 ab norepinephrine pg/ml epinephrine pg/ml 2 500 200

2 000 150

1 500 100 pg/ml pg/ml 1 000

50 500

0 0 baseline after end of baseline after end of concentration apnea concentration apnea cd ∆ heart rate vs. ∆ norepinephrine ∆ heart rate vs. ∆ epinephrine 4 000 200 l 3 000 150

2 000 100

1 000 50 ∆ epinephrine pg/ml ∆ norepinephrine pg/m

–60–40 –2020 0 –60–40 –2020 0 ∆ heart rate [bpm] ∆ heart rate [bpm]

▶Fig. 5 Levels of norepinephrine a and epinephrine b in serum before and immediately after the end of apnea. c Changes in heart rate correlated to the increase in norepinephrine concentration during apnea (Spearman’s rank correlation coefficient − 0.844, p = 0.001). d Correlation between heart rate and serum levels of epinephrine while the correlation with epinephrine was not significant (Spearman’s rank correlation coeffi- cient − 0.122, p = 0.721).

saturation (cerebral rSO2), whereas re-saturation occurred faster erogeneous age. Nevertheless, HR reduction is a commonly seen in the CNS than in peripheral tissues. Additionally, we found stable phenomenon during apnea and authors describe it as oxygen con- HBI values in cerebral NIRS recordings compared to steadily de- serving to avoid hypoxic damage [2, 4, 11]. creasing HBI in abdominal NIRS recordings (▶Fig. 4). An early de- However, HR reduction did not reach the level of significance in crease in HBI might be taken as a hint for decreased tissue perfu- our whole cohort. We saw a drop of HR (i. e., decrease of sion [41]. Taken together, we could validate previous results of our HR ≥ 10 bpm) in only 5 of 11 divers. In a previous study Nishiyasu group [14] and underline the preferential oxygen supply of the CNS et al. [34] compared divers developing strong apnea-induced brad- in presence of prolonged apnea. ycardia with those who show only mild bradycardia. Greater brad- ycardia was accompanied by greater vascular responses. Further- Influence on heart rate (HR) and limitations more, they found a linear relationship between HR reduction, an HR reduction (i. e., decrease of HR ≥ 10 bpm) was seen in 5 of 11 di- increase in vasoconstriction, and thus an increase in MAP during vers in our study. The physiological cause yet remains unclear. HR breath-hold [34]. We only found a slight correlation between in- in general is influenced by various mechanisms. Pulmonary stretch crease of systolic blood pressure and heart-rate reduction during receptors and subsets of visceral and somatic receptors with spinal dry apnea. The use of noninvasive blood in- afferents reflexively decrease cardiovagal activity and increase HR stead of invasive measurement might have damped the observed [9]. Voluntary breathing of hypoxic mixtures leads to hyperventi- effects. Interestingly, studies comparing cardiac performance dur- lation and tachycardia via vagal withdrawal and increased sympa- ing dry and immersed apnea show more pronounced effects on thetic nerve activity [11, 33, 42]. In contrast, in case of breath-hold blood pressure, TPR and cardiac function in immersed situations (i. e., no influence from stretch receptors) the enhancement of [31, 32]. Furthermore, water immersion causes profound effects vagal activity results in bradycardia [8, 9]. Vagal afferents are fur- on blood distribution and oxygen consumption [17]; stimulation ther stimulated by arterial baroreceptors, peripheral chemore- of facial cold receptors augments bradycardia and hypertension ceptors, and cardiopulmonary receptors resulting in a HR decrease. [5]. One may speculate that bradycardia may have been more dis- Beside these physiological reasons other personal factors further tinct with immersion and stimulation of facial cold receptors. affect the expression of the baroreflex, such as personal fitness Vasoconstriction and elevated blood pressure during apnea have level, cardiac performance capacity, ability to reach personal own so far been attributed to increased sympathetic activity being de- physical and mental capacities in an experimental setup and het- tected as muscle sympathetic nerve activity (MSNA) [18, 20, 22, 36].

Eichhorn L et al. Influence of Apnea-induced Hypoxia … Int J Sports Med 2017; 38: 85–91 89 Physiology & Biochemistry Thieme

Interestingly, MSNA caused by isolated hypoxia did not lead to va- References soconstriction and elevated blood pressure; moreover, MSNA in this case was accompanied by vasodilation and an increase in car- [1] Ainslie PN, Poulin MJ. Ventilatory, cerebrovascular, and cardiovascular diac output [1, 6, 7]. In a direct comparison during apnea, MSNA interactions in acute hypoxia: regulation by carbon dioxide. J Appl Physiol 2004; 97: 149–159 was elevated to a higher degree by hypoxia than by hypercapnia [2] Alboni P, Alboni M, Gianfranchi L. Diving bradycardia: a mechanism of [40]. Taken together, MSNA seems to play a key role in influencing defence against hypoxic damage. J Cardiovasc 2011; 12: 422–427 vasoconstriction, however other influences like increasing catecho- [3] Andersson JP, Evaggelidis L. Arterial oxygen saturation and diving lamines might also be taken into account to explain vasoconstric- response during dynamic in breath-hold divers. Scand J Med tion during apnea. Sci Sports 2009; 19: 87–91 In our study, we could demonstrate a significant increase in nor- [4] Andersson JPA, Linér MH, Rünow E, Schagatay EKA. Diving response epinephrine concentrations following apnea. Interestingly, the and arterial oxygen saturation during apnea and exercise in breath- measured increase showed a large inter-individual variability rang- hold divers. J Appl Physiol 2002; 93: 882–886 ing from 20 % to 870 %. Moreover, the increase of norepinephrine [5] Andersson J, Schagatay E, Gislén A, Holm B. Cardiovascular responses in our study was inversely correlated with the measured heart rate. to cold-water immersions of the forearm and face, and their relationship to apnoea. Eur J Appl Physiol 2000; 83: 566–572 In a recent study Chmura et al. measured a modest increase of nor- epinephrine in 4 free-divers [10]. Kaye et al. have previously shown [6] Breskovic T, Ivancev V, Banic I, Jordan J, Dujic Z. Peripheral chemoreflex sensitivity and sympathetic nerve activity are normal in apnea divers that inhaled high CO elevated norepinephrine concentrations and 2 during training season. Auton Neurosci 2010; 154: 42–47 caused bradycardia [24]. Therefore, the pCO increase during apnea 2 [7] Breskovic T, Valic Z, Lipp A, Heusser K, Ivancev V, Tank J, Dzamonja G, in our study may be responsible for the detected elevated norepi- Jordan J, Shoemaker JK, Eterovic D, Dujic Z. Peripheral chemoreflex nephrine concentrations. Norepinephrine receptors (α1-adreno- regulation of sympathetic vasomotor tone in apnea divers. Clin Auton ceptors) are known to activate smooth muscle cell contraction Res 2010; 20: 57–63 leading to increased total peripheral resistance (TPR). Norepineph- [8] de Burgh Daly M, Kirkman E, Wood LM. Cardiovascular responses to rine seems to play an important role influencing bradycardia via stimulation of cardiac receptors in the cat and their modification by α1-adrenoceptors in the lateral septal area of the limbic structure. changes in respiration. J Physiol 1988; 407: 349–362 This was further substantiated by Scopinho et al., who were able to [9] Chapleau MW, Sabharwal R. Methods of assessing vagus nerve activity and reflexes. Heart Fail Rev 2011; 16: 109–127 decrease baroreflex dependent bradycardia by a selective α1-adrenoceptor antagonist in the situation of high blood pressure [10] Chmura J, Kawczyński A, Medraś M, Jóźków P, Morawiec B. The impact of on psychomotor performance and blood catecholamine [39]. This finding is supported by our result of an inverse correla- concentration. Undersea Hyperb Med 2014; 41: 111–117 tion between HR and norepinephrine concentration. [11] Costalat G, Pichon A, Joulia F, Lemaître F. Modeling the diving Bradycardia seems to be a result of a pronounced baroreflex. A bradycardia: Toward an “oxygen-conserving breaking point”? Eur J norepinephrine concentration correlated inversely with HR may Appl Physiol 2015; 115: 1475–1484 also be taken as an indication that bradycardia is mediated by in- [12] Cross TJ, Kavanagh JJ, Breskovic T, Johnson BD, Dujic Z. Dynamic creased cardiac afterload. It has been reported, that the heart re- cerebral autoregulation is acutely impaired during maximal apnoea in acts to an exacerbated cardiac afterload in a negative chronotrop- trained divers. PloS One 2014; 9: e87598 ic manner [43]. Furthermore, the appearance of cardiac arrhyth- [13] Donald MJ, Paterson B. End tidal carbon dioxide monitoring in mias under apnea has been attributed to increased cardiac afterload prehospital and retrieval medicine: a review. Emerg Med J 2006; 23: 728–730 [19, 29]. Interestingly, the observed lower HR variability at the end [14] Eichhorn L, Erdfelder F, Kessler F, Doerner J, Thudium MO, Meyer R, of apnea may underscore the general role of increased catechola- Ellerkmann RK. Evaluation of near-infrared spectroscopy under mine concentrations. apnea-dependent hypoxia in humans. J Clin Monit Comput 2015; 29: 749–757 [15] Ferretti G. Extreme human breath-hold diving. Eur J Appl Physiol 2001; Conclusion 84: 254–271 This study documents various physiological responses during pro- [16] Ferretti G, Costa M. Diversity in and adaptation to breath-hold diving longed apnea. Norepinephrine concentrations showed a significant in humans. Comp Biochem Physiol A Mol Integr Physiol 2003; 136: increase due to apnea and were inversely correlated with HR. Fur- 205–213 thermore, norepinephrine may contribute to blood flow from pe- [17] Foster GE, Sheel AW. The human diving response, its function, and its control. Scand J Med Sci Sports 2005; 15: 3–12 ripheral tissues to the CNS and thus help to preserve cerebral tis- [18] Guaraldi P, Serra M, Barletta G, Pierangeli G, Terlizzi R, Calandra-Buon- sue O2 saturation longer than that of peripheral tissue. This finding aura G, Cialoni D, Cortelli P. Cardiovascular changes during maximal might extend our view of the underlying mechanism behind the breath-holding in elite divers. Clin Auton Res 2009; 19: 363–366 diving response. [19] Hansel J, Solleder I, Gfroerer W, Muth CM, Paulat K, Simon P, Heitkamp H-C, Niess A, Tetzlaff K. Hypoxia and cardiac arrhythmias in breath- hold divers during voluntary immersed breath-holds. Eur J Appl Physiol Acknowledgements 2009; 105: 673–678 Special thanks to all volunteers of the study. [20] Hardy JC, Gray K, Whisler S, Leuenberger U. Sympathetic and blood pressure responses to voluntary apnea are augmented by hypoxemia. J Appl Physiol 1994; 77: 2360–2365

90 Eichhorn L et al. Influence of Apnea-induced Hypoxia … Int J Sports Med 2017; 38: 85–91 [21] Harriss DJ, Atkinson G. Ethical Standards in Sport and Exercise Science [33] Naeije R. Physiological adaption of the cardiovascular system to high Research: 2016 Update. Int J Sports Med 2015; 36: 1121–1124 altitude. Prog Cardiovasc Dis 2010; 52: 456–466 [22] Heusser K, Dzamonja G, Tank J, Palada I, Valic Z, Bakovic D, Obad A, [34] Nishiyasu T, Tsukamoto R, Kawai K, Hayashi K, Koga S, Ichinose M. Ivancev V, Breskovic T, Diedrich A, Joyner MJ, Luft FC, Jordan J, Dujic Z. Relationships between the extent of apnea-induced bradycardia and Cardiovascular regulation during apnea in elite divers. Hypertension the vascular response in the arm and leg during dynamic two-legged 2009; 53: 719–724 knee extension exercise. Am J Physiol 2012; 302: H864–H871 [23] Joulia F, Lemaitre F, Fontanari P, Mille ML, Barthelemy P. Circulatory [35] Palada I, Obad A, Bakovic D, Valic Z, Ivancev V, Dujic Z. Cerebral and effects of apnoea in elite breath-hold divers. Acta Physiol 2009; 197: peripheral hemodynamics and oxygenation during maximal dry 75–82 breath-holds. Respir Physiol Neurobiol 2007; 157: 374–381 [24] Kaye J, Buchanan F, Kendrick A, Johnson P, Lowry C, Bailey J, Nutt D, [36] Perini R, Tironi A, Gheza A, Butti F, Moia C, Ferretti G. Heart rate and Lightman S. Acute carbon dioxide exposure in healthy adults: blood pressure time courses during prolonged dry apnoea in evaluation of a novel means of investigating the stress response. J breath-hold divers. Eur J Appl Physiol 2008; 104: 1–7 Neuroendocrinol 2004; 16: 256–264 [37] Przybyłowski T, Bangash M-F, Reichmuth K, Morgan BJ, Skatrud JB, [25] Kinoshita T, Nagata S, Baba R, Kohmoto T, Iwagaki S. Cold-water face Dempsey JA. Mechanisms of the cerebrovascular response to apnoea immersion per se elicits cardiac parasympathetic activity. Circ J 2006; in humans. J Physiol 2003; 548: 323–332 70: 773–776 [38] Schagatay E, Andersson J. Diving response and apneic time in humans. [26] Kjeld T, Pott FC, Secher NH. Facial immersion in cold water enhances Undersea Hyperb Med 1998; 25: 13–19 cerebral blood velocity during breath-hold exercise in humans. J Appl [39] Scopinho AA, Alves FHF, Resstel LBM, Correa FMA, Crestani CC. Lateral Physiol 2009; 106: 1243–1248 septal area α1- and α2-adrenoceptors differently modulate baroreflex [27] Laurino M, Menicucci D, Mastorci F, Allegrini P, Piarulli A, Scilingo EP, activity in unanaesthetized rats. Exp Physiol 2012; 97: 1018–1029 Bedini R, Pingitore A, Passera M, L’abbate A, Gemignani A. Mind-body [40] Somers VK, Mark AL, Zavala DC, Abboud FM. Contrasting effects of relationships in elite apnea divers during breath holding: a study of hypoxia and hypercapnia on ventilation and sympathetic activity in autonomic responses to acute hypoxemia. Front Neuroeng 2012; 5: 4 humans. J Appl Physiol 1989; 67: 2101–2106 [28] Leuenberger UA, Hardy JC, Herr MD, Gray KS, Sinoway LI. Hypoxia [41] Takami Y, Tajima K, Masumoto H. Near-infrared spectroscopy for augments apnea-induced peripheral vasoconstriction in humans. J noninvasive evaluation of chest wall ischemia immediately after left Appl Physiol 2001; 90: 1516–1522 internal thoracic artery harvesting. Gen Thorac Cardiovasc Surg 2008; [29] Lindholm P, Lundgren CEG. The physiology and pathophysiology of 56: 281–287 human breath-hold diving. J Appl Physiol 2009; 106: 284–292 [42] Teppema LJ, Dahan A. The ventilatory response to hypoxia in [30] Lin YC, Shida KK, Hong SK. Effects of hypercapnia, hypoxia, and mammals: mechanisms, measurement, and analysis. Physiol Rev 2010; rebreathing on circulatory response to apnea. J Appl Physiol 1983; 54: 90: 675–754 172–177 [43] Vincent J-L. Understanding cardiac output. Crit Care 2008; 12: 174 [31] Marabotti C, Piaggi P, Menicucci D, Passera M, Benassi A, Bedini R, [44] Willie CK, Ainslie PN, Drvis I, MacLeod DB, Bain AR, Madden D, Maslov L’Abbate A. Cardiac function and oxygen saturation during maximal PZ, Dujic Z. Regulation of brain blood flow and oxygen delivery in elite breath-holding in air and during whole-body surface immersion. breath-hold divers. J Cereb Blood Flow Metab 2015; 35: 66–73 Diving Hyperb Med 2013; 43: 131–137 [32] Marabotti C, Scalzini A, Cialoni D, Passera M, L’Abbate A, Bedini R. Cardiac changes induced by immersion and breath-hold diving in humans. J Appl Physiol 106: 293–297

Eichhorn L et al. Influence of Apnea-induced Hypoxia … Int J Sports Med 2017; 38: 85–91 91