Training to yoga respiration selectively increases respiratory sensation in healthy man

https://doi.org/10.1016/j.resp.2004.11.010Get rights and content

Abstract

Because yoga practitioners think they are benefiting from their breath training we hypothesized that yoga respiration training (YRT) could modify the respiratory sensation. Yoga respiration (YR) (“ujjai”) consisted of very slow, deep breaths (2–3 min−1) with sustained breath-hold after each inspiration and expiration. At inclusion in the study and after a 2-month YRT program, we determined in healthy subjects their eupneic ventilatory pattern and their capacity to discriminate external inspiratory resistive loads (respiratory sensation), digital tactile mechanical pressures (somesthetic sensation) and sound-pressure stimulations (auditory sensation). Data were compared to a gender-, age-, and weight-matched control group of healthy subjects who did not undergo the YRT program but were explored at the same epochs. After the 2-month YRT program, the respiratory sensation increased. Thus, both the exponent of the Steven's power law (Ψ = n) and the slope of the linear–linear plot between Ψ and mouth pressure (Pm) were significantly higher, and the intercept with ordinate axis of the Ψ versus Pm relationship was lower. After YRT, the peak Pm developed against inspiratory loads was significantly lower, reducing the load-induced activation of respiratory afferents. YRT induced long-lasting modifications of the ventilatory pattern with a significant lengthening of expiratory duration and a modest tidal volume increase. No significant changes in somesthetic and auditory sensations were noted. In the control group, the respiratory sensation was not modified during a 15-min period of yoga respiration, despite the peak Pm changes in response to added loads were then significantly reduced. These data suggest that training to yoga respiration selectively increases the respiratory sensation, perhaps through its persistent conditioning of the breathing pattern.

Introduction

In the philosophical doctrine of yoga, hatha-yoga is the physical training part combining postural exercise (“asana”), relaxation, and voluntary control of breathing (“pranayama”). Most of previous studies only considered the global influence of yoga and very few focused on the benefits of yoga respiration (YR), i.e., “pranayama”. Some authors denote an improvement of the pulmonary function (Birdel and Edgren, 2000, Makwana et al., 1988, Raju et al., 1973, Telles et al., 1993, Yadav and Das, 2001) but it was difficult to assess whether these effects resulted from yoga postures (“asana”), relaxation, and/or “pranayama” (YR) alone. “Pranayama” involves a diversity of breathing maneuvers among them “ujjai” one, which consists of slow and ample, near-vital capacity (VC) maneuvers, accompanied by apnea at the end of inspiration and expiration.

Until now, there was surprisingly no study on the influence of YR training (YRT) on the respiratory sensation documented by psychometric measurement during added respiratory loads. Indeed, “ujjai” induces long-lasting modifications of the breathing pattern, characterized by increased tidal volume and lowered breathing frequency (Joshi et al., 1992, Stanescu et al., 1981). This could modify the activation of vagal and/or respiratory muscle afferents by mechanical loading of the respiratory apparatus as well as the gain of the central integration of respiratory sensory pathways. Moreover, most of the yoga practitioners think they are benefiting from their breath training (Birdel and Edgren, 2000, Yadav and Das, 2001).

Our several years of clinical experience of “sophrologie” in patients suffering from dypnea at exertion, including training to reproduce an ample and slow breathing pattern, clearly indicates that their “difficulty in breathing” was markedly attenuated after several months of this training program. It is obvious that dyspnea of patients does not equate to the load-induced respiratory sensation in healthy volunteers, despite high values of added resistors also elicit in control subjects the same sensation of “air hunger” than that reported by patients. Our clinical observations encouraged us to test the consequences of training to “ujjai” yoga respiration (YRT) on the respiratory sensation in healthy subjects. We hypothesized that YRT could selectively influence the perceived magnitude of added respiratory loads without affecting other arms of sensitivity. Thus, we explored the respiratory, auditory and somesthetic (tactile) sensations, before and after a 2-month YRT program. Data were compared to those measured at the same epochs in a control group of healthy subjects who targeted breathing at the “ujjai” yoga pattern for a brief (15 min) period during which the respiratory load detection was tested.

Section snippets

Subjects

Seventeen healthy subjects (11 females and 6 males) without experience of hatha-yoga, were trained for two months in yoga respiration and constituted the YRT group. We compared them to nine gender-, age-, and weight-matched control subjects (six females/three males). Their morphological characteristics are shown in Table 1. None of them had any experience in respiratory physiology (they were relatives of nurses and technicians). None of the subjects suffered from and had no antecedents of

YRT-induced changes in eupneic breathing pattern

The pulmonary function did not vary in the YRT group after the 2-month training program.

As shown in Fig. 2, YRT induced long-lasting changes in the eupneic breathing pattern which persisted at distance of any “ujjai” maneuvers. There was no significant change in tidal volume (before YRT: 0.68 ± 0.10 l BTPS body temperature pressure saturated; after YRT: 0.92 ± 0.22 l BTPS) but the respiratory frequency was significantly (p < 0.05) reduced (before YRT: 19.6 ± 2 min−1; after YRT: 13.6 ± 3 min−1) due to a

Discussion

The major findings of this study were that training in “ujjai” yoga respiration selectively increased the exponent of the Ψ = n relationship between the subjective estimate of added inspiratory load (Ψ) and the physical stimulus (Φ = peak mouth pressure). The expression of data using the linear Ψ versus Φ relationship also revealed that YRT not only increased the slope of the linear regression but also markedly reduced its intercept with the ordinate axis. The absence of changes in the k

References (35)

  • T. Chonan et al.

    Effects of voluntary constraining of thoracic displacement during hypercapnia

    J. Appl. Physiol.

    (1987)
  • P.W. Davenport et al.

    Cerebral cortical evoked potentials elicited by cat intercostals muscle mechanoreceptors

    J. Appl. Physiol.

    (1993)
  • B.H. Demediuk et al.

    Dissociation between dyspnea and respiratory effort

    Am. Rev. Respir. Dis.

    (1992)
  • A.F. DiMarco et al.

    Sensation of inspired volume in normal subject and in quadriplegic patients

    J. Appl. Physiol.

    (1982)
  • K.S. Gopal et al.

    Effect of yogosanas and pranayamas on blood pressure, pulse rate and some respiratory functions

    Indian J. Physiol. Pharmacol.

    (1973)
  • S.B. Gottfried et al.

    Perception of changes in airflow resistance in obstructive pulmonary disorders

    Am. Rev. Respir. Dis.

    (1981)
  • S.B. Gottfried et al.

    Respiratory sensation in chronic obstructive pulmonary disease

    Am. Rev. Respir. Dis.

    (1985)
  • Cited by (26)

    • A systematic review and meta-analysis on the effects of yoga on weight-related outcomes

      2016, Preventive Medicine
      Citation Excerpt :

      Roughly symmetrical funnel plots were regarded to indicate low risk of publication bias; and asymmetrical funnel plots were regarded to indicate high risk of publication bias. The literature search retrieved 435 non-duplicate records of which 158 full-texts were assessed for eligibility, and 31 of them were eligible (Bera and Rajapurkar, 1993; Blumenthal et al., 1989; Chen et al., 2008; Chen et al., 2010; Cohen et al., 2008a; DiPietro et al., 1998; Elavsky and McAuley, 2007a; Harbans et al., 2011; Hegde et al., 2013; Kanaya et al., 2014; Kanojia et al., 2013; Lee et al., 2012; Lu and Wang, 2007; Mahajan et al., 1999; Manjunath and Telles, 2012; McDermott et al., 2014; Mooventhan and Khode, 2014; Ray et al., 2001; Sakuma et al., 2012; Seo et al., 2012; Shukla and Gehlot, 2014; Stachenfeld et al., 1998; Telles et al., 2014; Telles et al., 2013; Thiyagarajan et al., 2015; Tracy and Hart, 2013; Van Puymbroeck et al., 2007; Yang et al., 2011; Khatri et al., 2007; Kim et al., 2013; Manchanda et al., 2013), while 127 of those full-texts were excluded (Table 1) (McIver, 2010; Cheung et al., 2012; Corey et al., 2014; LaCroix et al., 2012; Sharma et al., 2013; Singh et al., 2011; Hoogbruin et al., 2012; Lai et al., 2010; Trivedi and Mishra, 2014; Bock et al., 2010; Bock et al., 2014; Cheema et al., 2011; Cohen et al., 2013; Kandula et al., 2013; Neumark-Sztainer et al., 2011; Flaherty, 2014; Guarracino et al., 2006; Ades et al., 2003; Ankad et al., 2011; Ankad Roopa et al., 2011; Armstrong and Scott Smedley, 2003; Bhutkar et al., 2011; Choudhary and Mishra, 2013; Hunter et al., 2013; Malhotra et al., 2005; Murthy et al., 2011; Nagarathna and Nagendra, 1987; Raj et al., 2011; Ramos-Jimenez et al., 2009; Sarvottam et al., 2013; Satyanarayana et al., 1992; Sivasankaran et al., 2006; Telles et al., 1993; Telles et al., 2010; Thomley et al., 2011; Villien et al., 2005; Yadav et al., 2012; Yadav et al., 2014; Yang and James, 2014; Begum et al., 2012; Cajka et al., 2013; Čajka et al., 2012; Ce et al., 2015; Chatterjee and Mondal, 2014; Chaya et al., 2006; Chaya et al., 2008; Dhananjai et al., 2013; Goncalves et al., 2011; Halder et al., 2015; Hegde et al., 2011; Herur et al., 2011; Kim et al., 2014; Krejci, 2011; Kubo et al., 2011; Madanmohan et al., 2008; Malhotra et al., 2010; Maninder et al., 2013; Miles et al., 2013; Santhi Sri et al., 2014; Singh et al., 2008; Sukhsohale and Phatak, 2012; Tyagi et al., 2014; Narendran et al., 2005a; Narendran et al., 2005b; Stein et al., 2014; Sprod et al., 2015; Emery and Blumenthal, 1990; Bar et al., 2014; Chu et al., 2014a; Cramer et al., 2014b; Hartley et al., 2014; Innes and Vincent, 2007; Jayasinghe, 2004; Liu et al., 2014; Manchanda and Madan, 2014; Patel et al., 2012; Brady, 2007; Ades et al., 2005; Manchanda et al., 2000; Pal et al., 2013; Pal et al., 2011; Raghuram et al., 2014; Cadmus-Bertram et al., 2013; Littman et al., 2012; Carei et al., 2010; McIver et al., 2009; Mitchell et al., 2007; Yurtkuran et al., 2007; Cade et al., 2010; McCaffrey et al., 2005; Murugesan et al., 2000; Rahnama et al., 2011; Cheung et al., 2014; Ebnezar et al., 2012; Sharma et al., 2015; Nidhi et al., 2012; Nidhi et al., 2013a; Nidhi et al., 2013b; Rakhshani et al., 2015; Innes and Selfe, 2012; Ikai et al., 2013; Amita et al., 2009; Gordon et al., 2008; Nishanth et al., 2011; Shantakumari et al., 2013; Skoro-Kondza et al., 2009; Visweswaraiah and Telles, 2004; Dandekar, 2013; Göring et al., 2013; Jorrakate et al., 2015; Rajajeyakumar et al., 2014; Santaella et al., 2011; Schmidt et al., 1997; Sharma et al., 2014; Blumenthal et al., 1991; Elavsky and McAuley, 2007b; Elavsky and McAuley, 2007c; Arciero et al., 2014; Boxer et al., 2010; Bryan and Zipp, 2014; DeBar et al., 2012; Kenny et al., 2010; Kim et al., 2008; Park et al., 2010; Ramen et al., 2013; Sabet Sarvestani et al., 2009; Ziv et al., 2013). Thirty-one full-text articles reporting on 30 RCTs involving a total of 2173 participants were included in the qualitative analysis; and 27 articles on 26 RCTs were meta-analyzed (Fig. 1).

    • Body Sensations: The Conscious Aspects of Interoception

      2021, Body Sensations: The Conscious Aspects of Interoception
    View all citing articles on Scopus
    View full text