When the human brain goes diving: using near-infrared spectroscopy to measure cerebral and systemic cardiovascular responses to deep, breath-hold diving in elite freedivers

Abstract

Continuous measurements of haemodynamic and oxygenation changes in free living animals remain elusive. However, developments in biomedical technologies may help to fill this knowledge gap. One such technology is continuous-wave near-infrared spectroscopy (CW-NIRS)-a wearable and non-invasive optical technology. Here, we develop a marinized CW-NIRS system and deploy it on elite competition freedivers to test its capacity to function during deep freediving to 107 m depth. We use the oxyhaemoglobin and deoxyhaemoglobin concentration changes measured with CW-NIRS to monitor cerebral haemodynamic changes and oxygenation, arterial saturation and heart rate. Furthermore, using concentration changes in oxyhaemoglobin engendered by cardiac pulsation, we demonstrate the ability to conduct additional feature exploration of cardiac-dependent haemodynamic changes. Freedivers showed cerebral haemodynamic changes characteristic of apnoeic diving, while some divers also showed considerable elevations in venous blood volumes close to the end of diving. Some freedivers also showed pronounced arterial deoxygenation, the most extreme of which resulted in an arterial saturation of 25%. Freedivers also displayed heart rate changes that were comparable to diving mammals both in magnitude and patterns of change. Finally, changes in cardiac waveform associated with heart rates less than 40 bpm were associated with changes indicative of a reduction in vascular compliance. The success here of CW-NIRS to non-invasively measure a suite of physiological phenomenon in a deep-diving mammal highlights its efficacy as a future physiological monitoring tool for human freedivers as well as free living animals. This article is part of the theme issue 'Measuring physiology in free-living animals (Part II)'.

Keywords: SpO2; breath-hold diving; cererbal oxygenation; diving physiology; freediving; near-infrared spectroscopy.

Figure 1.

Monitoring diving physiology and behaviour in elite freedivers. (a) Location of NIRS sensor head over the pre-frontal cortex (red box). (b) Respective placements of NIRS sensor body (red box, right) and Little Leonardo W1000-PD3GT (yellow box, left). (Online version in colour.)

Figure 4.

Raw [O₂Hb] traces visualizing heart rate and cardiac waveform. (a) Shape and magnitude of a normal cardiac waveform at 41 bpm. (b) Cardiac waveform with increased magnitude and pronounced tidal waves at a heart rate of 27 bpm. (c) Cardiac waveforms across four cardiac cycles with three key features identified—(i) percussive wave (originating from the contraction of the left ventricle and ejection of blood); (ii) tidal wave (caused by the elasticity of aortic wall); and (iii) diastolic minimum (caused by relaxation of the heart)—showing the transient changes in cardiac waveform on the longest interbeat interval (5.4 s) recorded. (Online version in colour.)

Figure 2.

Monitoring physiological variables in freedivers. Two example dives to 67 and 97 m from two freedivers, showing (a) dive and temperature data and (b) cerebral haemodynamic responses. (c) Heart rate, cerebral and arterial blood oxygen responses. High-frequency peaks and troughs in the accelerometry signal are indicative of leg movements associated with swimming in the CWT dive in example 1 and arm movements of pulling along the rope in an FIM dive in example 2. During early descent, the divers swim intensely to overcome positive buoyancy which is seen as high-amplitude peaks on the acceleration trace, while later during the descent, the negatively buoyant divers free fall. At the bottom of the dive, when the divers turn and swim upwards against negative buoyancy, the acceleration trace, again, shows high-amplitude peaks throughout ascent. Oscillations in TSI from minutes 4–5 in example 2 were probably the result of involuntary breathing movements [14]. TSI measurement in example 2 was lost on surfacing as one of three NIRS channels lost contact with the diver's head.

Figure 3.

(a) Arterial oxygen saturation (SpO₂—red line) and depth data from a dive resulting in the lowest arterial blood oxygenation recorded (25%). (b) Heart rate and depth profiles for a human diver (80 kg) perform the deepest dive recorded (107 m) and an 81 kg California sea lion (Zalophus californianus) diving to 240 m (data re-drawn with permission from McDonald & Ponganis [32]). (Online version in colour.)

Figure 5.

Cardiac pulsation of [ΔHbT] compared between a grey seal and human diver. The cardiac pulsation of [ΔHbT] for one example dive (a) is compared to the resting state cardiac pulsation in a grey seal (b). The colour encodes the underlying heart rate, revealing increasing pulse magnitude with bradycardia heart rate. (c) Comparing the human diver from (a) and the grey seal in (b), the average [ΔHbT] (sum of oxyhaemoglobin and deoxyhaemoglobin) magnitude, with errorbars spanning the standard deviation within the 5 bpm interval in heart rate, reveals similar trends. The pulse shape change in the human diver is emphasized by normalizing the pulses in length and height. Pulses are then averaged based on the underlying diving depth (d) and heart rate (e), with shaded error bars spanning the standard deviation of the normalized [ΔHbT] at any given time of the pulse. (Online version in colour.)