Journal of Bodywork & Movement Therapies (2015) 19, 434e441
Available online at www.sciencedirect.com
ScienceDirect journal homepage: www.elsevier.com/jbmt
YOGA METHODOLOGY
Sirsasana (headstand) technique alters head/ neck loading: Considerations for safety Rachel Hector, M.S. , Jody L. Jensen, PhD* Department of Kinesiology & Health Education, University of Texas at Austin, 2109 San Jacinto Blvd D3700, UT Mail Code: D3700, Austin, TX 78712-1415, USA Received 19 May 2014; received in revised form 2 October 2014; accepted 17 October 2014
KEYWORDS Headstand; Yoga; Cervical loading; Sirsasana; Headstand forces; Inversion
Abstract Background: This study examined the weight-bearing responsibility of the head and neck at moments of peak force during three headstand techniques. Methods: Three matched groups of 15 each (18e60 years old) were formed based upon lower limb entry/exit technique: symmetrical extended, symmetrical flexed, and asymmetrical flexed. All 45 practitioners performed 3 headstands. Kinematics and kinetics were analyzed to locate peak forces acting on the head, loading rate, center of pressure (COP) and cervical alignment. Findings: During entry, symmetrical extended leg position trended towards the lowest loads as compared to asymmetrical or symmetrical flexed legs (Cohen’s d Z 0.53 and 0.39 respectively). Also, symmetrical extended condition produced slower loading rates and more neutral cervical conditions during loading. Interpretation: Subjects loaded the head with maximums of 40e48% of total body weight. The data support the conclusion that entering the posture with straight legs together may reduce the load and the rate of change of that load. ª 2014 Elsevier Ltd. All rights reserved.
Introduction Since its introduction into American culture over a century ago, yoga has gained popularity as a modern method of
* Corresponding author. Tel.: þ1 512 232 2685; fax: þ1 8 512 471 8914. E-mail addresses: [email protected] (R. Hector), [email protected] (J.L. Jensen). http://dx.doi.org/10.1016/j.jbmt.2014.10.002 1360-8592/ª 2014 Elsevier Ltd. All rights reserved.
obtaining states of meditation, wellness and physical fitness. Now a 6 billion dollar industry, this practice has more than 15.8 million Americans regularly coming to the mat to reap its studied stress-relieving benefits (YIAS, 2008). Practitioners perform complex postures and conscious breathing exercises shown collectively to reduce stress, improve mood, bolster immunity, increase flexibility, improve sleep and aid in recovery processes (Bower et al., 2005; Cohen et al., 2004; Curtis et al., 2011; Gururaja et al., 2011; Hedge et al., 2011). However, as
Sirsasana (headstand) technique alters head/neck loading the study of physiological aspects of yoga expands, the biomechanical aspects of what actually takes place on the mat are being ignored. The body often moves into uncommon positions within the context of yoga. Inquiry into the structural impact of potentially risky positions is needed due to the repetitive nature of yoga practice, the lack of biomechanical research in this arena, and yoga’s growing popularity across all age groups. Various pairings of spinal action (eg: flexion, extension) and body position (eg: kneeling, supine) make up the asana, or posture, portion of a yoga practice. Inversions, often encouraged due to the level of challenge and benefits, present a particularly atypical deviation from daily activity. Improved focus, reduced heart rate, calming effects, encouragement of venous return, movement of lymph, improvement of immune function, flush of toxins and strength gains in the trunk, upper body and respiratory diaphragm are all touted as benefits accrued in yoga inversion practices. Headstand, known as the “king of all postures” is held up in yogic literature as a cure-all for issues of circulation, the common cold, back pain and depression (Iyengar, 1966). Scientists agree with Iyengar when it comes to healing depression with yoga (Woolery et al., 2004) and have observed marked shifts in circulation during and shortly after the pose (Rao, 1968). In 1924, Kuvalayananda, found increased blood pressure readings in 11 healthy headstanding adults slowly dropped during the static phase of the posture (Kuvalayananda, 1924). Fortyfour years later, Shankar Rao validated Kuvalayananda’s work, and noted that heart rate was lower in headstanding than standing, inspiratory capacity was greatest upside down and that oxygen utilization in headstand was greater than in standing or supine positions (Rao, 1962a,b, 1968). In addition to lowering heart rate, Manjunath and Telles (2003) also found that two minutes of headstand increased sympathetic tone in male practitioners. Despite decades of inquiry on the physiological changes during inversions, the literature is markedly devoid of research on the structural aspects of loaded inversions such as headstands. Research findings in other fields have delineated potential at-risk circumstances for the discs and vertebrae of the spine. Cadaver cervical spine failure loads have been reported within a range of 300 Ne17 kN (Cusick and Yoganandan, 2001) with men consistently having 600 N greater loading capacity than women (Pintar et al., 1998). Bearing larger loads without incurring damage does not appear to be a product of practice. Examination of the cervical conditions of African wood-bearers, individuals practiced at loading the head, found increased degeneration and pain among male and female wood bearers bearing increased loads (Jager et al., 1997; Josaab et al., 1994). Although the discs and vertebrae are intended to aid in redistribution of force, the ability to do so depends on size, shape, and condition (Einhorn, 1992; Adams et al., 2000). Intervertebral mobility, especially in the cervical spine, generally decreases with age (Prescher, 1998). Disc degeneration increases both as vertebral mobility decreases and as discs are subjected to large, repeated or asymmetrical loads (Walter et al., 2011; Lotz and Chin, 2000; Matsumoto et al., 2010). Nearly 60% of the current United States yoga population currently falls within at-risk categories in terms of age (35e70) for disc degeneration,
435 thus investigation into magnitude and direction of forces acting on the head is warranted (YIAS, 2008; Matsumoto et al., 2010). This study is an initiation of the biomechanical examination of headstand (sirsasana) with a description and comparison of the loading conditions and factors that contribute to increased, asymmetrical, fixed, or flexed loading during all phases of the posture. In the most commonly practiced version headstand, forearms are on the floor, hands are clasped around the back of the skull, and the crown of the head is also in contact with the floor (Fig. 1). Typical entry (often modified by preferred yogic lineage) occurs in one of three ways: legs asymmetrical, legs symmetrical and bent, and legs symmetrical and straight (Fig. 2). Entry and exit techniques are generally paired and stability, or the static performance of the pose with legs vertical, is without noticeable difference across all techniques. In this study we quantified baseline estimates of average and maximum forces acting on the head and neck during the all three phases of the headstand: entry, stability and exit. Forces were partitioned by technique across the three phases to determine if technique played a role in the magnitude of cervical loading. Loading rate, neck angle, and center of pressure (COP) were determined to estimate the stiffness of the tissues in loadbearing, lateral movement of the head, and potential flexion during loading in headstand.
Methods Participants Forty-five subjects were recruited from the local yoga community in Austin, Texas. Participants were screened prior to arrival and subsequently recruited based upon age (18 years old and up), self-report of freedom from chronic neck injury, and ability to perform a supported headstand for five breath cycles. Individuals performing headstands outside of the context of yoga practice were not
Fig. 1 Visual representation of upper body in sirsasana (headstand).
436
R. Hector, J.L. Jensen
Fig. 2 Visual representation of three lower limb techniques for entry and exit into Sirsasana: asymmetrical or single leg, symmetrical/double bent, and symmetrical/double straight.
considered. The criteria insured that each individual was a low-risk semi-regular practitioner of the activity. Upon arrival for one 60-min study visit, subjects signed an IRB-approved consent form and filled out a survey indicating age, gender, yoga experience, headstanding experience, and frequency of practice. Forty-five individuals were placed into three groups based upon entry technique. The symmetrical extended group (SE) entered and exited the pose with legs together and straight (SE, n Z 15 (13 female, 2 male), mean weight 60.0 (SD 6.8) kg, mean height 157.3 (SD 5.2) cm mean 36.9 (SD 8.8) years of age, mean 8.1 (SD 4.5) years of yoga experience, mean 6.2 (SD 3.7) years of headstand experience, mean 4.5 (SD 1.4) weekly yoga frequency). The symmetrical flexed group (SF) entered the pose with symmetrical thighs and flexed knees (SF, n Z 15 (13 female, 2 male), mean weight 61.8 (SD 12.5) kg, mean height 159.1 (SD 8.2) cm, mean 37.9 (SD 11.1) years of age, mean 8.8 (SD 5.7) years of yoga experience, mean 3.5 (SD 3.5) years of headstand experience, mean 4.4 (SD 1.7) weekly yoga frequency). The asymmetrical flexed group (AF) entered the pose one leg at a time with the legs in varying degrees of flexion (AF, n Z 15 (13 female, 2 male), mean weight 67.1 kg (SD 17.6) kg, mean height 158.2 (SD 8.4) cm, mean 32.3 (SD 9.2) years of age, mean 6.8 (SD 3.6) years of yoga experience, mean 5.7 (SD 3.9) years of headstand experience, mean 4.2 (SD 1.0) weekly yoga frequency). Individuals performing headstand tend to employ the same technique each time and often cannot achieve the pose otherwise. In order to compare across groups, means were closely matched for age, gender, and yoga experience level. No significant differences were found across groups on age, weight, height, years of yoga or headstand experience, or weekly yoga frequency. All groups contained 13 females and 2 males.
the force plate. Markers were secured to the following locations: chin, center of the forehead, spinous processes of 4 vertebrae (C3, C7, T9, and L5) and the right and left earlobes, lateral epicondyles of the humerus, acromion processes, greater trochanters, lateral condyles of the femur, and bases of the second toe. None of these markers interfered with headstand performance. With the markers affixed, subjects were captured standing on the force plate (Bertec; K70501) to record baseline ground reaction force, natural standing head position, and overall body weight and height. A 4 6 foot platform covered with a thick yoga mat was then placed over the 40 60 cm force plate (Fig. 3). The subject was asked to place the crown of the head inside the red semicircle on the mat. The section of the mat where the
Protocol, data collection and equipment Subjects completed a 10 min warm up. They were instructed to engage in self-guided yoga practice in silence with the knowledge that they would be performing headstand after the allotted time period. After the warm-up, 18 reflective markers for the 10camera motion capture system (F-series; Vicon Nexus 1.6.1) were attached to the subject and one marker was placed on
Fig. 3
Headstand platform.
Sirsasana (headstand) technique alters head/neck loading
437
practitioner placed his or her head was in full contact with the force plate; however, the section under the arms was not. The practitioner was not aware of this fact and could not feel the difference between the free-floating area touching the plate and the area not touching the plate. With this platform method, only the ground reaction force acting on the head was recorded. The subject was instructed to begin when ready and to maintain the pose for five self-paced breaths prior to exiting it. Spotting was provided to reduce risk; however, individuals were not aided into the pose or allowed to lean on the spotter. Three headstand trials were captured. A requisite 2 min rest period was enforced between trials to minimize the effects of fatigue. During each trial, spotting was provided regardless of need and no instruction was given. After the third trial, markers were removed and the subject performed a required 5 min yoga-related cooldown. Sampling rate of 120 Hz was used for x, y, z coordinate data and 1200 Hz for force place data. The force data were smoothed and then down-sampled from 1200 Hz to 120 Hz.
Dependent variables The kinematic data were analyzed to determine the three phases of each headstand. The entry phase was marked when the average vertical velocity of the toe marker was consistently at or greater than 10 mm/s and ended when the velocity dropped below 10 mm/s. The stability phase began at the end of the entry phase ended when the absolute value of the toe velocity was consistently greater than 10 mm/s. The exit phase began at the end of the stability phase and ended when absolute toe velocity was less than 10 mm/s. For single leg practitioners, the entry phase began when the second foot left the ground and the exit phase ended when the first foot landed. Force was the primary dependent measure. In an effort to discern if one of the headstand techniques generated greater loading patterns on the head, maximum load and average load for each phase were calculated for each of the three methods. All force outcomes were calculated and compared as raw values and as percentages, calculated by dividing individual headstand forces by the subject’s total ground reaction force during quiescent stance. Loading rate was defined as the instantaneous rate of change of the resultant force acting through the head. The absolute difference between consecutive forces was divided by the time between samples collected at a sampling rate of 120 hz. Average and maximum loading rates were reported for each phase and compared between the three methods. Neck angle, a non-invasive measure of the head in relationship to the trunk that described the angle about the third cervical vertebrae, was calculated to examine flexion and extension of the neck during the task. Using the values of the left ear (LE), C3 vertebrae, and C7 vertebrae from the lateral perspective, the angle between the vector from C3 to C7 and the vector from C3 to LE was calculated (Fig. 4). Static stance of all subjects was processed as an initial or comparative neck measurement against which inversion neck angles could be compared. Neck angle was
Fig. 4 Depiction of neck angle measurement using kinematic markers for the left ear, C3 Vertebra, and C7 Vertebra.
examined at times of peak force and peak loading rate within each phase and compared across the three methods to determine relative cervical alignment behaviors at points of interest. Center of pressure (COP) was estimated. The trajectory of the ground reaction force vector was calculated in terms of distance traveled over time. COP was divided into anterior-posterior COPx and medio-lateral COPy components and compared across phase and between techniques to determine balancing accommodations. This value was an initial determination of both the stillness of the head as well as the apex of the force vector.
Design and statistical analysis This was an experimental project comparing three headstand techniques. In this study, individuals performed only one variation of entry/exit into supported headstand. Each subject performed three trials. The first trial was considered a warm-up trial and the second two were analyzed on all measures previously described and then averaged together to boost power and reduce variability. The statistical assessment was carried out with 3 3 repeated-measures ANOVAs for each dependent variable with phase as the within-subjects component and method as the between-subjects component. Following the ANOVA, post hoc procedures were calculated to discern notable pairwise comparisons. Levene’s test was used to test for homogeneity of variance. Mauchly’s test was used to test for violations of the sphericity assumption. Significance was reported for p-values less than 0.05. Upon determination of force, loading rate, neck angle and COP during each of the three entry methods; dependent measures were correlated to participant characteristics including issues of experience, frequency, weight, age, and gender.
438
R. Hector, J.L. Jensen
Results Force
Fig. 5 Maximum forces as a percentage of individual weight acting on the head during each time period as modified by method: symmetrical extended (SE), symmetrical flexed (SF), and asymmetrical flexed (AF).
Gaps in marker coordinate data, due to momentarily hidden or untracked markers, were filled via linear interpolation for gaps of up to 0.2 s. Larger gaps were eliminated from the analysis and subject neck angle, the only measure affected by this, was taken from the subject’s other headstand. In the case of COP, a non-linear measure, gaps could not be filled via linear interpolation. Instead, COP was considered stationary during periods of zero data as the practitioner’s head was actually lifted off of the mat/plate during those times as confirmed by a simultaneous absence of force. In this way, when the head returned to the plate, a deviation from center was not recorded. Analysis of the force data did not require interpolation or fill.
This analysis reviewed the loads applied through the head during three phases of headstand: entry, stability, and exit. Maximum forces on the head expressed, as a percent of individual body weight, were 40e48% of total practitioner body weight. These forces acting on the crown of the head differed across phases, F(2, 84) Z 24.762, P < 0.001 (Fig. 5) and as an interaction with method, F(4, 84) Z 5.137, P < 0.001. Pairwise comparisons further showed significant differences between all three phases (P Z 0.006) with the stability phase exhibiting the greatest forces and exit the least. Between group effects were not significant; however, effect sizes (Cohen’s d) comparing the difference between one method and another displayed moderate effects. During entry, symmetrical extended leg position trended towards the lowest loads with little difference between the other techniques. The effect between SE and AF or SF was moderate (Cohen’s d Z 0.53 and 0.39 respectively) while the difference between SF and AF was low (Cohen’s d Z 0.13). The symmetrical extended variation showed similarly low force values during stability. In relationship to SE, the values were 0.55 for AF and 0.47 for SF with the effect between AF and SF again being nearly absent at 0.04. Technique did not appear to markedly affect exit forces. The effects between SE and the other two methods, AF and SF, were 0.30 and 0.09 respectively. Interestingly, exit effects differentiated
Fig. 6 Maximum forces acting on the head during each time period as modified by method: Symmetrical extended (SE), symmetrical flexed (SF), and asymmetrical flexed (AF).
Sirsasana (headstand) technique alters head/neck loading between SF and AF techniques; AF practitioners had lower exit loads (Cohen’s d Z 0.35). Raw maximum force values were significant across phases, F(2, 84) Z 26.302, P < 0.001, as well as the interaction between phase and method, F(4, 84) Z 5.866, P < 0.01 (Fig. 6). As with percent force, differences were found between all phases (P < 0.01) with the largest loads occurring during stability and the smallest during exit. Maximum loads during entry and stability were above 300N on average for the AF group and below 250N for the SE group (a difference of w15 lbs). The SF group exhibited midrange means in terms of maximum forces; however, it was the most variable of the techniques. The trends between techniques during entry and stability did not hold for exit as AF and SF values dropped while SE remained stable and therefore was no longer the method with the lower loads.
Loading rate Maximum loading rate was significant across phases, F(1.441, 84) Z 18.757, P < 0.01 (Fig. 7). Entry exhibited the fastest loading rate, followed by exit and then stability. Bonferroni procedures revealed a significant difference across all time phases between SE and AF (P Z 0.047). AF practitioners, who often kick, had faster loading rates than those who come up in a controlled double legs straight variation (SE). No differences were found between SF and the other techniques.
439 Table 1 Neck angles in degrees at baseline, time of maximum force, and maximum loading rate. Baseline Entry Max force SE 140.21 SF 140.18 AF 141.94
Stability Max LR Max force
Exit
Max LR Max force
Max LR
143.95 144.22 137.87 137.96 140.46 138.20 137.66 143.38 141.28 140.77 140.56 140.71 140.05 138.41 139.34 137.83 136.32 135.69
(P Z 0.007), which showed more neutral or flexed neck position. No significant differences were found between groups.
Center of pressure Both measures showed a main effect for phase: F(1.54, 84) Z 12.456, P < 0.001 for COPx and F(1.615, 84) Z 9.102, P Z 0.001 for COPy (Fig. 8). Pairwise comparisons of both values were significant between entry and stability (P < 0.01) and stability and exit (P < 0.01). Further analysis revealed a significant difference between lateral and frontal movement during entry F(1, 88) Z 4.649, P Z 0.034, and exit F(1, 88) Z 5.210, P Z 0.025. In other words, practitioners produced more side-to-side movement at the crown of the head while navigating the balance between the arms and head during entry and exit phases regardless of technique used.
Neck angle
Discussion Neck angle, calculated in a standing position, revealed an average between 140 and 142 across all groups. Angles greater than this were in the direction of neck extension and angles smaller than baseline were in the direction of flexion. Points of interest included neck angle at the time of maximum force and at the time of maximum loading rate (Table 1). Neck angle at time of maximum force was not significant across phase, between methods or as an interaction. Neck angle at time of maximum loading rate was significant across phase with maximum neck extension occurring during entry F(1.312, 84) Z 6.80, P Z 0.007. Entry differed from both stability (P Z 0.013) and exit
On average, headstand practitioners loaded the head with maximums of 40e48% of total body weight during headstand. Maximum forces were largest while the practitioner held the full inverted posture for five breaths indicating an overall tendency to transfer weight from the arms during entry to the head and neck during stability. In terms of lower body technique, moderate effects indicate a clear trend: individuals entering with straight legs that lift up together exhibited noticeably lower raw and percent loads, slowed loading rates and less cervical flexion than the other
Fig. 7 Maximum force rate of change during each time period as modified by method: Symmetrical extended (SE), symmetrical flexed (SF), and asymmetrical flexed (AF).
Fig. 8 Lateral (y) and frontal (x) movement of center of pressure during each phase of headstand collapsed across all methods.
440 techniques during entry and during the five breaths of stability. This is an important finding as entry/exit technique is one of the few variables a headstander may be able to control. That said, it should be noted that all variations exhibited raw force values of potential concern according to previous literature. Fifty-one percent of the individuals in the total sample experienced maximums above 300 N, the lower range of previously determined cadaver cervical spine failure loads (Cusick and Yoganandan, 2001). By gender, 38% of the women assessed exceeded this value and, less surprisingly, five out of the six males exceeded this value as the vertebral bodies in a male body can withstand greater forces. Technique modified force values. During entry, more than three-quarters of the subjects entering with straight, symmetrical legs stayed below 300 N of vertical loading while more than one-half of the subjects who entered one leg at a time, experienced loads above that value. The force values for the symmetrical flexed technique proved to be the least predictable and most difficult to perform with consistency and controlled force. Differences between techniques were negligible upon exit because individuals using the straight leg technique continued to show control during exit whereas the asymmetrical technique, which resembles a one-legged kick, had a dramatic decrease in percent force and raw force upon exit. Many of those with the largest loads also exhibited the fastest loading rates. Loading rate modulates the type or incidence of injury due to the altered mechanics of loading. At faster rates, tissues temporarily stiffen thereby increasing failure loads (Pintar et al., 1998). However, the loading of the head during headstand is a more complex scenario due to cervical alignment at the time of load and the period of self-supported static loading that follows the initial load. Sudden stiffening of the tissues and musculature does not support reduced risk in headstanding due to increased intervertebral strain (Wang et al., 2000). Additionally, headstand is not a single event in a practitioner’s life and must be done in a sustainable way. Individuals approaching it rapidly may find the increased load they are initially bearing to be too much to maintain during stability. The relative stability of the forces found in the symmetrical extended technique was partially explained by the markedly slower rate of loading. That technique had lower entrance and stability forces and longer times to entry showing that moving more slowly while loading the head resulted in a decrease of force. Alternatively, subjects performing the single leg entry loaded the head more rapidly than the other two methods. Although entering the pose with one leg at a time can be done slowly, the practitioner often had to use momentum to approach the pose due to physical proportions or strength/flexibility challenges. Cervical alignment shifted as the practitioner moved between phases of headstand regardless of technique. States of direct compression (natural lordotic cervical curve compression) and compression-extension offer a cervical failure rate that is four times greater than that of cervical-flexion (Carter et al., 2002). Neck angle at maximum load velocity during entry was significantly more extended than during stability or exit. Although increased load at any cervical alignment engenders a certain amount
R. Hector, J.L. Jensen of risk, loading during flexion or flattened alignment is more likely to result in neurological damage than flexion or neutral/natural lordotic loading (Yoganandan et al., 1990). Practitioners who entered the posture one leg at a time trended towards the least able to maintain a lordotic cervical curve during points of maximum velocity. During stability and exit, the angle of the neck at maximum loads correlated to the individual’s baseline neck conditions. Center of pressure also shifted as the practitioner moved between phases of headstand regardless of technique. The location and behavior of this force vector acting on the fixed portion of the spine may impact the magnitude of the load and the type of potential injury. Restraint of the head has led to greater forces and increased probability of fractures (Yoganandan et al., 1986) Allowing the base of support to make small movements to adjust to loading is a boon for headstanders; however, most of the movement observed was lateral, or side-to-side, which introduces the potential for lateral fractures and nerve root avulsion (McElhaney and Myers, 1993).
Study limitations The uniqueness of individual style, independent of performance technique leads to large between-subject variability. Although differences were found, moderate to large effect sizes in measures with low power, indicate that sample size may have been too small to detect some potential differences. Entry/exit technique used was determined by level of ability, instruction, and structural limitations. An individual who regularly performs a headstand task typically does so in the same way every time, both on entry and exit, without varying the technique. Most practitioners simply cannot, and do not, perform all three entry techniques, rather they are more easily able to do one over another or were trained to use a specific technique. Subsequently, subjects were not randomly assigned to groups and doing otherwise would have devalued the practical application of the results. Exit technique was not always identical across trials. Additionally, experience levels of headstand and yoga were self-reported and for those who have many years of experience, values may have been unintentionally inflated or deflated. Lastly, headstanders very new to performing this asana, who may be at the greatest risk, were not represented as the study criteria requiring them to be stable in the pose for five breaths could not be met.
Conclusion Supported Sirasasana (headstand) is a reoccurring practice in a yoga class and the numbers of yoga classes and practitioners are on the rise. By modifying technique during headstand, practitioners may be able to reduce the risks of rapid loading of the vertebral bodies of the cervical spine. Entering the posture with straight legs together may reduce both the load on the head and the rate of change of that load. Conversely, exiting the pose quickly with a push of the arms and a controlled one-legged fall appears to reduce cervical spine involvement upon exit.
Sirsasana (headstand) technique alters head/neck loading Independent of technique, the maximum load carried by the head during yogic headstand approaches half of the practitioner’s total weight and surpasses known failure thresholds in more than 50% of the subjects. The growing population of yoga practitioners should be informed of these values when considering regular performance of this pose. Educated individuals can make an informed decision about weight-bearing through the head as cervical condition is not able to be assessed outwardly by a yoga instructor. Future research should address potential contributions to increased force values such as weight, gender, height, and anthropomorphic measurements. Currently, the average western practitioner will be told to refrain from headstand if they are experiencing hypertension, glaucoma, detached-retina, pregnancy, cervical dysfunction, menstruation, heart conditions or other serious medical concerns. Additionally, instructors should teach or encourage students to approach the posture with a more controlled technique and an awareness of the magnitude of the load incurred.
Acknowledgments This study was conducted without sponsors or any outside sources of funding.
References Adams, M.A., Freeman, B.J., Morrison, H.P., Nelson, I.W., Dolan, P., 2000. Mechanical initiation of intervertebral disc degeneration. Spine 25, 1625e1636. Bower, J.E., Woolery, A., Sternlieb, B., Garet, D., 2005. Yoga for cancer patients and survivors. Cancer Control 12, 165e171. Carter, J.W., Ku, G.S., Nuckley, D.J., Ching, R.P., 2002. Tolerance of the cervical spine to eccentric axial compression. Stapp Car Crash J. 46, 441e459. Cohen, L., Warneke, C., Fouladi, R.T., Rodriguez, M.A., ChaoulReich, A., 2004. Psychological adjustment and sleep quality in a randomized trial of the effects of a Tibetan yoga intervention in patients with lymphoma. Cancer 100, 2253e2260. Curtis, K., Osadchuk, A., Katz, J., 2011. An eight-week yoga intervention is associated with improvements in pain, psychological functioning and mindfulness, and changes in cortisol levels in women with fibromyalgia. J. Pain Res. 4, 189e201. Cusick, J.F., Yoganandan, N., 2001. Biomechanics of the cervical spine 4: major injuries. Clin. Biomech. 17, 1e20. Einhorn, T.A., 1992. Bone strength: the bottom line. Calcif. Tissue Int. 51, 333e339. Gururaja, D., Harano, K., Toyotake, I., Kobayashi, H., 2011. Effect of yoga on mental health: comparative study between young and senior subjects in Japan. Int. J. Yoga 4, 7e12.
441 Hegde, S.V., Adhikari, P., Kotian, S., Pinto, V.J., D’Souza, S., D’Souza, V., 2011. Effect of 3-month yoga on oxidative stress in type 2 diabetes with or without complications: a controlled clinical trial. Diabetes Care 34, 2208e2210. Iyengar, B.K.S., 1966. Light on Yoga. Schocken, New York. Jager, H.J., Gordon-Harris, L., Mehring, U.M., Goetz, G.F., Mathias, K.D., 1997. Degenerative change in the cervical spine and load-carrying on the head. Skelet. Radiol. 26, 475e481. Josaab, M., Torode, M., Rao, P.V., 1994. Preliminary findings on the effect of load-carrying to the structural integrity of the cervical spine. Surg. Radiol. Anat. 16, 393e398. ṅsa (Lonavla, Bombay). Kuvalayananda, 1924. Yoga-mıma Lotz, J.C., Chin, J.R., 2000. Intervertebral disc cell death is dependent on the magnitude and duration of spinal loading. Spine 25, 1477e1483. Manjunath, N.K., Telles, S., 2003. Effects of sirsasana (headstand) practice on autonomic and respiratory variables. Indian J. Physiol. Pharmacol. 47 (1), 34e42. Matsumoto, M., Okada, E., Ichihara, D., Watanabe, K., Chiba, K., Toyama, Y., Fujiwara, H., Momoshima, S., Nishiwaki, Y., Hashimoto, T., Takahata, T., 2010. Age-related changes of thoracic and cervical intervertebral discs in asymptomatic subjects. Spine 35, 1359e1364. McElhaney, J.H., Myers, B.S., 1993. Biomechanical aspects of cervical spine trauma. In: Nahum, A.M., Melvin, J.W. (Eds.), Accidental Injury: Biomechanics and Prevention. Springer-Verlag New York Inc., New York. Pintar, F.A., Yoganandan, N., Voo, L., 1998. Effect of age and loading rate on human cervical spine injury. Spine 23, 1957e1962. Prescher, A., 1998. Anatomy and pathology of the aging spine. Eur. J. Radiol. 27 (3), 181e195. Rao, S., 1962a. Metabolic cost of head-stand posture. J. Appl. Physiol. 17, 117e118. Rao, S., 1962b. Cardiovascular responses to head-stand posture. J. Appl. Physiol. 18, 987e990. Rao, S., 1968. Respiratory responses to head stand posture. J. Appl. Physiol. 24, 697e699. Walter, B.A., Korecki, C.L., Purmessur, D., Roughley, P.J., Michalek, A.J., Iatridis, J.C., 2011. Complex loading affects intervertebral disc mechanics and biology. Osteoarthr. Cartil. 19, 1011e1018. Wang, J.L., Parnianpou, M., Shirazi-Adl, A., Engin, A.E., 2000. Viscoelastic finite-element analysis of a lumbar motion segment in combined compression and sagittal flexion: effect of loading rate. Spine 25, 310e318. Woolery, A., Myers, H., Sternlieb, B., Zeltzer, L., 2004. A yoga intervention for young adults with elevated symptoms of depression. Altern. Ther. Health Med. 10, 60e63. Yoga in America Study, 2008. Harris Interactive Service Bureau. Yoga Journal. Yoganandan, N., Sances Jr., A., Maiman, D.J., Myklebust, J.B., Pech, P., Larson, S.J., 1986. Experimental spinal injuries with vertical impact. Spine 11, 855e860. Yoganandan, N., Sances Jr., A., Pintar, F., 1990. Injury biomechanics of the human cervical column. Spine 15, 1031e1039.
Available online at www.sciencedirect.com
ScienceDirect journal homepage: www.elsevier.com/jbmt
YOGA METHODOLOGY
Sirsasana (headstand) technique alters head/ neck loading: Considerations for safety Rachel Hector, M.S. , Jody L. Jensen, PhD* Department of Kinesiology & Health Education, University of Texas at Austin, 2109 San Jacinto Blvd D3700, UT Mail Code: D3700, Austin, TX 78712-1415, USA Received 19 May 2014; received in revised form 2 October 2014; accepted 17 October 2014
KEYWORDS Headstand; Yoga; Cervical loading; Sirsasana; Headstand forces; Inversion
Abstract Background: This study examined the weight-bearing responsibility of the head and neck at moments of peak force during three headstand techniques. Methods: Three matched groups of 15 each (18e60 years old) were formed based upon lower limb entry/exit technique: symmetrical extended, symmetrical flexed, and asymmetrical flexed. All 45 practitioners performed 3 headstands. Kinematics and kinetics were analyzed to locate peak forces acting on the head, loading rate, center of pressure (COP) and cervical alignment. Findings: During entry, symmetrical extended leg position trended towards the lowest loads as compared to asymmetrical or symmetrical flexed legs (Cohen’s d Z 0.53 and 0.39 respectively). Also, symmetrical extended condition produced slower loading rates and more neutral cervical conditions during loading. Interpretation: Subjects loaded the head with maximums of 40e48% of total body weight. The data support the conclusion that entering the posture with straight legs together may reduce the load and the rate of change of that load. ª 2014 Elsevier Ltd. All rights reserved.
Introduction Since its introduction into American culture over a century ago, yoga has gained popularity as a modern method of
* Corresponding author. Tel.: þ1 512 232 2685; fax: þ1 8 512 471 8914. E-mail addresses: [email protected] (R. Hector), [email protected] (J.L. Jensen). http://dx.doi.org/10.1016/j.jbmt.2014.10.002 1360-8592/ª 2014 Elsevier Ltd. All rights reserved.
obtaining states of meditation, wellness and physical fitness. Now a 6 billion dollar industry, this practice has more than 15.8 million Americans regularly coming to the mat to reap its studied stress-relieving benefits (YIAS, 2008). Practitioners perform complex postures and conscious breathing exercises shown collectively to reduce stress, improve mood, bolster immunity, increase flexibility, improve sleep and aid in recovery processes (Bower et al., 2005; Cohen et al., 2004; Curtis et al., 2011; Gururaja et al., 2011; Hedge et al., 2011). However, as
Sirsasana (headstand) technique alters head/neck loading the study of physiological aspects of yoga expands, the biomechanical aspects of what actually takes place on the mat are being ignored. The body often moves into uncommon positions within the context of yoga. Inquiry into the structural impact of potentially risky positions is needed due to the repetitive nature of yoga practice, the lack of biomechanical research in this arena, and yoga’s growing popularity across all age groups. Various pairings of spinal action (eg: flexion, extension) and body position (eg: kneeling, supine) make up the asana, or posture, portion of a yoga practice. Inversions, often encouraged due to the level of challenge and benefits, present a particularly atypical deviation from daily activity. Improved focus, reduced heart rate, calming effects, encouragement of venous return, movement of lymph, improvement of immune function, flush of toxins and strength gains in the trunk, upper body and respiratory diaphragm are all touted as benefits accrued in yoga inversion practices. Headstand, known as the “king of all postures” is held up in yogic literature as a cure-all for issues of circulation, the common cold, back pain and depression (Iyengar, 1966). Scientists agree with Iyengar when it comes to healing depression with yoga (Woolery et al., 2004) and have observed marked shifts in circulation during and shortly after the pose (Rao, 1968). In 1924, Kuvalayananda, found increased blood pressure readings in 11 healthy headstanding adults slowly dropped during the static phase of the posture (Kuvalayananda, 1924). Fortyfour years later, Shankar Rao validated Kuvalayananda’s work, and noted that heart rate was lower in headstanding than standing, inspiratory capacity was greatest upside down and that oxygen utilization in headstand was greater than in standing or supine positions (Rao, 1962a,b, 1968). In addition to lowering heart rate, Manjunath and Telles (2003) also found that two minutes of headstand increased sympathetic tone in male practitioners. Despite decades of inquiry on the physiological changes during inversions, the literature is markedly devoid of research on the structural aspects of loaded inversions such as headstands. Research findings in other fields have delineated potential at-risk circumstances for the discs and vertebrae of the spine. Cadaver cervical spine failure loads have been reported within a range of 300 Ne17 kN (Cusick and Yoganandan, 2001) with men consistently having 600 N greater loading capacity than women (Pintar et al., 1998). Bearing larger loads without incurring damage does not appear to be a product of practice. Examination of the cervical conditions of African wood-bearers, individuals practiced at loading the head, found increased degeneration and pain among male and female wood bearers bearing increased loads (Jager et al., 1997; Josaab et al., 1994). Although the discs and vertebrae are intended to aid in redistribution of force, the ability to do so depends on size, shape, and condition (Einhorn, 1992; Adams et al., 2000). Intervertebral mobility, especially in the cervical spine, generally decreases with age (Prescher, 1998). Disc degeneration increases both as vertebral mobility decreases and as discs are subjected to large, repeated or asymmetrical loads (Walter et al., 2011; Lotz and Chin, 2000; Matsumoto et al., 2010). Nearly 60% of the current United States yoga population currently falls within at-risk categories in terms of age (35e70) for disc degeneration,
435 thus investigation into magnitude and direction of forces acting on the head is warranted (YIAS, 2008; Matsumoto et al., 2010). This study is an initiation of the biomechanical examination of headstand (sirsasana) with a description and comparison of the loading conditions and factors that contribute to increased, asymmetrical, fixed, or flexed loading during all phases of the posture. In the most commonly practiced version headstand, forearms are on the floor, hands are clasped around the back of the skull, and the crown of the head is also in contact with the floor (Fig. 1). Typical entry (often modified by preferred yogic lineage) occurs in one of three ways: legs asymmetrical, legs symmetrical and bent, and legs symmetrical and straight (Fig. 2). Entry and exit techniques are generally paired and stability, or the static performance of the pose with legs vertical, is without noticeable difference across all techniques. In this study we quantified baseline estimates of average and maximum forces acting on the head and neck during the all three phases of the headstand: entry, stability and exit. Forces were partitioned by technique across the three phases to determine if technique played a role in the magnitude of cervical loading. Loading rate, neck angle, and center of pressure (COP) were determined to estimate the stiffness of the tissues in loadbearing, lateral movement of the head, and potential flexion during loading in headstand.
Methods Participants Forty-five subjects were recruited from the local yoga community in Austin, Texas. Participants were screened prior to arrival and subsequently recruited based upon age (18 years old and up), self-report of freedom from chronic neck injury, and ability to perform a supported headstand for five breath cycles. Individuals performing headstands outside of the context of yoga practice were not
Fig. 1 Visual representation of upper body in sirsasana (headstand).
436
R. Hector, J.L. Jensen
Fig. 2 Visual representation of three lower limb techniques for entry and exit into Sirsasana: asymmetrical or single leg, symmetrical/double bent, and symmetrical/double straight.
considered. The criteria insured that each individual was a low-risk semi-regular practitioner of the activity. Upon arrival for one 60-min study visit, subjects signed an IRB-approved consent form and filled out a survey indicating age, gender, yoga experience, headstanding experience, and frequency of practice. Forty-five individuals were placed into three groups based upon entry technique. The symmetrical extended group (SE) entered and exited the pose with legs together and straight (SE, n Z 15 (13 female, 2 male), mean weight 60.0 (SD 6.8) kg, mean height 157.3 (SD 5.2) cm mean 36.9 (SD 8.8) years of age, mean 8.1 (SD 4.5) years of yoga experience, mean 6.2 (SD 3.7) years of headstand experience, mean 4.5 (SD 1.4) weekly yoga frequency). The symmetrical flexed group (SF) entered the pose with symmetrical thighs and flexed knees (SF, n Z 15 (13 female, 2 male), mean weight 61.8 (SD 12.5) kg, mean height 159.1 (SD 8.2) cm, mean 37.9 (SD 11.1) years of age, mean 8.8 (SD 5.7) years of yoga experience, mean 3.5 (SD 3.5) years of headstand experience, mean 4.4 (SD 1.7) weekly yoga frequency). The asymmetrical flexed group (AF) entered the pose one leg at a time with the legs in varying degrees of flexion (AF, n Z 15 (13 female, 2 male), mean weight 67.1 kg (SD 17.6) kg, mean height 158.2 (SD 8.4) cm, mean 32.3 (SD 9.2) years of age, mean 6.8 (SD 3.6) years of yoga experience, mean 5.7 (SD 3.9) years of headstand experience, mean 4.2 (SD 1.0) weekly yoga frequency). Individuals performing headstand tend to employ the same technique each time and often cannot achieve the pose otherwise. In order to compare across groups, means were closely matched for age, gender, and yoga experience level. No significant differences were found across groups on age, weight, height, years of yoga or headstand experience, or weekly yoga frequency. All groups contained 13 females and 2 males.
the force plate. Markers were secured to the following locations: chin, center of the forehead, spinous processes of 4 vertebrae (C3, C7, T9, and L5) and the right and left earlobes, lateral epicondyles of the humerus, acromion processes, greater trochanters, lateral condyles of the femur, and bases of the second toe. None of these markers interfered with headstand performance. With the markers affixed, subjects were captured standing on the force plate (Bertec; K70501) to record baseline ground reaction force, natural standing head position, and overall body weight and height. A 4 6 foot platform covered with a thick yoga mat was then placed over the 40 60 cm force plate (Fig. 3). The subject was asked to place the crown of the head inside the red semicircle on the mat. The section of the mat where the
Protocol, data collection and equipment Subjects completed a 10 min warm up. They were instructed to engage in self-guided yoga practice in silence with the knowledge that they would be performing headstand after the allotted time period. After the warm-up, 18 reflective markers for the 10camera motion capture system (F-series; Vicon Nexus 1.6.1) were attached to the subject and one marker was placed on
Fig. 3
Headstand platform.
Sirsasana (headstand) technique alters head/neck loading
437
practitioner placed his or her head was in full contact with the force plate; however, the section under the arms was not. The practitioner was not aware of this fact and could not feel the difference between the free-floating area touching the plate and the area not touching the plate. With this platform method, only the ground reaction force acting on the head was recorded. The subject was instructed to begin when ready and to maintain the pose for five self-paced breaths prior to exiting it. Spotting was provided to reduce risk; however, individuals were not aided into the pose or allowed to lean on the spotter. Three headstand trials were captured. A requisite 2 min rest period was enforced between trials to minimize the effects of fatigue. During each trial, spotting was provided regardless of need and no instruction was given. After the third trial, markers were removed and the subject performed a required 5 min yoga-related cooldown. Sampling rate of 120 Hz was used for x, y, z coordinate data and 1200 Hz for force place data. The force data were smoothed and then down-sampled from 1200 Hz to 120 Hz.
Dependent variables The kinematic data were analyzed to determine the three phases of each headstand. The entry phase was marked when the average vertical velocity of the toe marker was consistently at or greater than 10 mm/s and ended when the velocity dropped below 10 mm/s. The stability phase began at the end of the entry phase ended when the absolute value of the toe velocity was consistently greater than 10 mm/s. The exit phase began at the end of the stability phase and ended when absolute toe velocity was less than 10 mm/s. For single leg practitioners, the entry phase began when the second foot left the ground and the exit phase ended when the first foot landed. Force was the primary dependent measure. In an effort to discern if one of the headstand techniques generated greater loading patterns on the head, maximum load and average load for each phase were calculated for each of the three methods. All force outcomes were calculated and compared as raw values and as percentages, calculated by dividing individual headstand forces by the subject’s total ground reaction force during quiescent stance. Loading rate was defined as the instantaneous rate of change of the resultant force acting through the head. The absolute difference between consecutive forces was divided by the time between samples collected at a sampling rate of 120 hz. Average and maximum loading rates were reported for each phase and compared between the three methods. Neck angle, a non-invasive measure of the head in relationship to the trunk that described the angle about the third cervical vertebrae, was calculated to examine flexion and extension of the neck during the task. Using the values of the left ear (LE), C3 vertebrae, and C7 vertebrae from the lateral perspective, the angle between the vector from C3 to C7 and the vector from C3 to LE was calculated (Fig. 4). Static stance of all subjects was processed as an initial or comparative neck measurement against which inversion neck angles could be compared. Neck angle was
Fig. 4 Depiction of neck angle measurement using kinematic markers for the left ear, C3 Vertebra, and C7 Vertebra.
examined at times of peak force and peak loading rate within each phase and compared across the three methods to determine relative cervical alignment behaviors at points of interest. Center of pressure (COP) was estimated. The trajectory of the ground reaction force vector was calculated in terms of distance traveled over time. COP was divided into anterior-posterior COPx and medio-lateral COPy components and compared across phase and between techniques to determine balancing accommodations. This value was an initial determination of both the stillness of the head as well as the apex of the force vector.
Design and statistical analysis This was an experimental project comparing three headstand techniques. In this study, individuals performed only one variation of entry/exit into supported headstand. Each subject performed three trials. The first trial was considered a warm-up trial and the second two were analyzed on all measures previously described and then averaged together to boost power and reduce variability. The statistical assessment was carried out with 3 3 repeated-measures ANOVAs for each dependent variable with phase as the within-subjects component and method as the between-subjects component. Following the ANOVA, post hoc procedures were calculated to discern notable pairwise comparisons. Levene’s test was used to test for homogeneity of variance. Mauchly’s test was used to test for violations of the sphericity assumption. Significance was reported for p-values less than 0.05. Upon determination of force, loading rate, neck angle and COP during each of the three entry methods; dependent measures were correlated to participant characteristics including issues of experience, frequency, weight, age, and gender.
438
R. Hector, J.L. Jensen
Results Force
Fig. 5 Maximum forces as a percentage of individual weight acting on the head during each time period as modified by method: symmetrical extended (SE), symmetrical flexed (SF), and asymmetrical flexed (AF).
Gaps in marker coordinate data, due to momentarily hidden or untracked markers, were filled via linear interpolation for gaps of up to 0.2 s. Larger gaps were eliminated from the analysis and subject neck angle, the only measure affected by this, was taken from the subject’s other headstand. In the case of COP, a non-linear measure, gaps could not be filled via linear interpolation. Instead, COP was considered stationary during periods of zero data as the practitioner’s head was actually lifted off of the mat/plate during those times as confirmed by a simultaneous absence of force. In this way, when the head returned to the plate, a deviation from center was not recorded. Analysis of the force data did not require interpolation or fill.
This analysis reviewed the loads applied through the head during three phases of headstand: entry, stability, and exit. Maximum forces on the head expressed, as a percent of individual body weight, were 40e48% of total practitioner body weight. These forces acting on the crown of the head differed across phases, F(2, 84) Z 24.762, P < 0.001 (Fig. 5) and as an interaction with method, F(4, 84) Z 5.137, P < 0.001. Pairwise comparisons further showed significant differences between all three phases (P Z 0.006) with the stability phase exhibiting the greatest forces and exit the least. Between group effects were not significant; however, effect sizes (Cohen’s d) comparing the difference between one method and another displayed moderate effects. During entry, symmetrical extended leg position trended towards the lowest loads with little difference between the other techniques. The effect between SE and AF or SF was moderate (Cohen’s d Z 0.53 and 0.39 respectively) while the difference between SF and AF was low (Cohen’s d Z 0.13). The symmetrical extended variation showed similarly low force values during stability. In relationship to SE, the values were 0.55 for AF and 0.47 for SF with the effect between AF and SF again being nearly absent at 0.04. Technique did not appear to markedly affect exit forces. The effects between SE and the other two methods, AF and SF, were 0.30 and 0.09 respectively. Interestingly, exit effects differentiated
Fig. 6 Maximum forces acting on the head during each time period as modified by method: Symmetrical extended (SE), symmetrical flexed (SF), and asymmetrical flexed (AF).
Sirsasana (headstand) technique alters head/neck loading between SF and AF techniques; AF practitioners had lower exit loads (Cohen’s d Z 0.35). Raw maximum force values were significant across phases, F(2, 84) Z 26.302, P < 0.001, as well as the interaction between phase and method, F(4, 84) Z 5.866, P < 0.01 (Fig. 6). As with percent force, differences were found between all phases (P < 0.01) with the largest loads occurring during stability and the smallest during exit. Maximum loads during entry and stability were above 300N on average for the AF group and below 250N for the SE group (a difference of w15 lbs). The SF group exhibited midrange means in terms of maximum forces; however, it was the most variable of the techniques. The trends between techniques during entry and stability did not hold for exit as AF and SF values dropped while SE remained stable and therefore was no longer the method with the lower loads.
Loading rate Maximum loading rate was significant across phases, F(1.441, 84) Z 18.757, P < 0.01 (Fig. 7). Entry exhibited the fastest loading rate, followed by exit and then stability. Bonferroni procedures revealed a significant difference across all time phases between SE and AF (P Z 0.047). AF practitioners, who often kick, had faster loading rates than those who come up in a controlled double legs straight variation (SE). No differences were found between SF and the other techniques.
439 Table 1 Neck angles in degrees at baseline, time of maximum force, and maximum loading rate. Baseline Entry Max force SE 140.21 SF 140.18 AF 141.94
Stability Max LR Max force
Exit
Max LR Max force
Max LR
143.95 144.22 137.87 137.96 140.46 138.20 137.66 143.38 141.28 140.77 140.56 140.71 140.05 138.41 139.34 137.83 136.32 135.69
(P Z 0.007), which showed more neutral or flexed neck position. No significant differences were found between groups.
Center of pressure Both measures showed a main effect for phase: F(1.54, 84) Z 12.456, P < 0.001 for COPx and F(1.615, 84) Z 9.102, P Z 0.001 for COPy (Fig. 8). Pairwise comparisons of both values were significant between entry and stability (P < 0.01) and stability and exit (P < 0.01). Further analysis revealed a significant difference between lateral and frontal movement during entry F(1, 88) Z 4.649, P Z 0.034, and exit F(1, 88) Z 5.210, P Z 0.025. In other words, practitioners produced more side-to-side movement at the crown of the head while navigating the balance between the arms and head during entry and exit phases regardless of technique used.
Neck angle
Discussion Neck angle, calculated in a standing position, revealed an average between 140 and 142 across all groups. Angles greater than this were in the direction of neck extension and angles smaller than baseline were in the direction of flexion. Points of interest included neck angle at the time of maximum force and at the time of maximum loading rate (Table 1). Neck angle at time of maximum force was not significant across phase, between methods or as an interaction. Neck angle at time of maximum loading rate was significant across phase with maximum neck extension occurring during entry F(1.312, 84) Z 6.80, P Z 0.007. Entry differed from both stability (P Z 0.013) and exit
On average, headstand practitioners loaded the head with maximums of 40e48% of total body weight during headstand. Maximum forces were largest while the practitioner held the full inverted posture for five breaths indicating an overall tendency to transfer weight from the arms during entry to the head and neck during stability. In terms of lower body technique, moderate effects indicate a clear trend: individuals entering with straight legs that lift up together exhibited noticeably lower raw and percent loads, slowed loading rates and less cervical flexion than the other
Fig. 7 Maximum force rate of change during each time period as modified by method: Symmetrical extended (SE), symmetrical flexed (SF), and asymmetrical flexed (AF).
Fig. 8 Lateral (y) and frontal (x) movement of center of pressure during each phase of headstand collapsed across all methods.
440 techniques during entry and during the five breaths of stability. This is an important finding as entry/exit technique is one of the few variables a headstander may be able to control. That said, it should be noted that all variations exhibited raw force values of potential concern according to previous literature. Fifty-one percent of the individuals in the total sample experienced maximums above 300 N, the lower range of previously determined cadaver cervical spine failure loads (Cusick and Yoganandan, 2001). By gender, 38% of the women assessed exceeded this value and, less surprisingly, five out of the six males exceeded this value as the vertebral bodies in a male body can withstand greater forces. Technique modified force values. During entry, more than three-quarters of the subjects entering with straight, symmetrical legs stayed below 300 N of vertical loading while more than one-half of the subjects who entered one leg at a time, experienced loads above that value. The force values for the symmetrical flexed technique proved to be the least predictable and most difficult to perform with consistency and controlled force. Differences between techniques were negligible upon exit because individuals using the straight leg technique continued to show control during exit whereas the asymmetrical technique, which resembles a one-legged kick, had a dramatic decrease in percent force and raw force upon exit. Many of those with the largest loads also exhibited the fastest loading rates. Loading rate modulates the type or incidence of injury due to the altered mechanics of loading. At faster rates, tissues temporarily stiffen thereby increasing failure loads (Pintar et al., 1998). However, the loading of the head during headstand is a more complex scenario due to cervical alignment at the time of load and the period of self-supported static loading that follows the initial load. Sudden stiffening of the tissues and musculature does not support reduced risk in headstanding due to increased intervertebral strain (Wang et al., 2000). Additionally, headstand is not a single event in a practitioner’s life and must be done in a sustainable way. Individuals approaching it rapidly may find the increased load they are initially bearing to be too much to maintain during stability. The relative stability of the forces found in the symmetrical extended technique was partially explained by the markedly slower rate of loading. That technique had lower entrance and stability forces and longer times to entry showing that moving more slowly while loading the head resulted in a decrease of force. Alternatively, subjects performing the single leg entry loaded the head more rapidly than the other two methods. Although entering the pose with one leg at a time can be done slowly, the practitioner often had to use momentum to approach the pose due to physical proportions or strength/flexibility challenges. Cervical alignment shifted as the practitioner moved between phases of headstand regardless of technique. States of direct compression (natural lordotic cervical curve compression) and compression-extension offer a cervical failure rate that is four times greater than that of cervical-flexion (Carter et al., 2002). Neck angle at maximum load velocity during entry was significantly more extended than during stability or exit. Although increased load at any cervical alignment engenders a certain amount
R. Hector, J.L. Jensen of risk, loading during flexion or flattened alignment is more likely to result in neurological damage than flexion or neutral/natural lordotic loading (Yoganandan et al., 1990). Practitioners who entered the posture one leg at a time trended towards the least able to maintain a lordotic cervical curve during points of maximum velocity. During stability and exit, the angle of the neck at maximum loads correlated to the individual’s baseline neck conditions. Center of pressure also shifted as the practitioner moved between phases of headstand regardless of technique. The location and behavior of this force vector acting on the fixed portion of the spine may impact the magnitude of the load and the type of potential injury. Restraint of the head has led to greater forces and increased probability of fractures (Yoganandan et al., 1986) Allowing the base of support to make small movements to adjust to loading is a boon for headstanders; however, most of the movement observed was lateral, or side-to-side, which introduces the potential for lateral fractures and nerve root avulsion (McElhaney and Myers, 1993).
Study limitations The uniqueness of individual style, independent of performance technique leads to large between-subject variability. Although differences were found, moderate to large effect sizes in measures with low power, indicate that sample size may have been too small to detect some potential differences. Entry/exit technique used was determined by level of ability, instruction, and structural limitations. An individual who regularly performs a headstand task typically does so in the same way every time, both on entry and exit, without varying the technique. Most practitioners simply cannot, and do not, perform all three entry techniques, rather they are more easily able to do one over another or were trained to use a specific technique. Subsequently, subjects were not randomly assigned to groups and doing otherwise would have devalued the practical application of the results. Exit technique was not always identical across trials. Additionally, experience levels of headstand and yoga were self-reported and for those who have many years of experience, values may have been unintentionally inflated or deflated. Lastly, headstanders very new to performing this asana, who may be at the greatest risk, were not represented as the study criteria requiring them to be stable in the pose for five breaths could not be met.
Conclusion Supported Sirasasana (headstand) is a reoccurring practice in a yoga class and the numbers of yoga classes and practitioners are on the rise. By modifying technique during headstand, practitioners may be able to reduce the risks of rapid loading of the vertebral bodies of the cervical spine. Entering the posture with straight legs together may reduce both the load on the head and the rate of change of that load. Conversely, exiting the pose quickly with a push of the arms and a controlled one-legged fall appears to reduce cervical spine involvement upon exit.
Sirsasana (headstand) technique alters head/neck loading Independent of technique, the maximum load carried by the head during yogic headstand approaches half of the practitioner’s total weight and surpasses known failure thresholds in more than 50% of the subjects. The growing population of yoga practitioners should be informed of these values when considering regular performance of this pose. Educated individuals can make an informed decision about weight-bearing through the head as cervical condition is not able to be assessed outwardly by a yoga instructor. Future research should address potential contributions to increased force values such as weight, gender, height, and anthropomorphic measurements. Currently, the average western practitioner will be told to refrain from headstand if they are experiencing hypertension, glaucoma, detached-retina, pregnancy, cervical dysfunction, menstruation, heart conditions or other serious medical concerns. Additionally, instructors should teach or encourage students to approach the posture with a more controlled technique and an awareness of the magnitude of the load incurred.
Acknowledgments This study was conducted without sponsors or any outside sources of funding.
References Adams, M.A., Freeman, B.J., Morrison, H.P., Nelson, I.W., Dolan, P., 2000. Mechanical initiation of intervertebral disc degeneration. Spine 25, 1625e1636. Bower, J.E., Woolery, A., Sternlieb, B., Garet, D., 2005. Yoga for cancer patients and survivors. Cancer Control 12, 165e171. Carter, J.W., Ku, G.S., Nuckley, D.J., Ching, R.P., 2002. Tolerance of the cervical spine to eccentric axial compression. Stapp Car Crash J. 46, 441e459. Cohen, L., Warneke, C., Fouladi, R.T., Rodriguez, M.A., ChaoulReich, A., 2004. Psychological adjustment and sleep quality in a randomized trial of the effects of a Tibetan yoga intervention in patients with lymphoma. Cancer 100, 2253e2260. Curtis, K., Osadchuk, A., Katz, J., 2011. An eight-week yoga intervention is associated with improvements in pain, psychological functioning and mindfulness, and changes in cortisol levels in women with fibromyalgia. J. Pain Res. 4, 189e201. Cusick, J.F., Yoganandan, N., 2001. Biomechanics of the cervical spine 4: major injuries. Clin. Biomech. 17, 1e20. Einhorn, T.A., 1992. Bone strength: the bottom line. Calcif. Tissue Int. 51, 333e339. Gururaja, D., Harano, K., Toyotake, I., Kobayashi, H., 2011. Effect of yoga on mental health: comparative study between young and senior subjects in Japan. Int. J. Yoga 4, 7e12.
441 Hegde, S.V., Adhikari, P., Kotian, S., Pinto, V.J., D’Souza, S., D’Souza, V., 2011. Effect of 3-month yoga on oxidative stress in type 2 diabetes with or without complications: a controlled clinical trial. Diabetes Care 34, 2208e2210. Iyengar, B.K.S., 1966. Light on Yoga. Schocken, New York. Jager, H.J., Gordon-Harris, L., Mehring, U.M., Goetz, G.F., Mathias, K.D., 1997. Degenerative change in the cervical spine and load-carrying on the head. Skelet. Radiol. 26, 475e481. Josaab, M., Torode, M., Rao, P.V., 1994. Preliminary findings on the effect of load-carrying to the structural integrity of the cervical spine. Surg. Radiol. Anat. 16, 393e398. ṅsa (Lonavla, Bombay). Kuvalayananda, 1924. Yoga-mıma Lotz, J.C., Chin, J.R., 2000. Intervertebral disc cell death is dependent on the magnitude and duration of spinal loading. Spine 25, 1477e1483. Manjunath, N.K., Telles, S., 2003. Effects of sirsasana (headstand) practice on autonomic and respiratory variables. Indian J. Physiol. Pharmacol. 47 (1), 34e42. Matsumoto, M., Okada, E., Ichihara, D., Watanabe, K., Chiba, K., Toyama, Y., Fujiwara, H., Momoshima, S., Nishiwaki, Y., Hashimoto, T., Takahata, T., 2010. Age-related changes of thoracic and cervical intervertebral discs in asymptomatic subjects. Spine 35, 1359e1364. McElhaney, J.H., Myers, B.S., 1993. Biomechanical aspects of cervical spine trauma. In: Nahum, A.M., Melvin, J.W. (Eds.), Accidental Injury: Biomechanics and Prevention. Springer-Verlag New York Inc., New York. Pintar, F.A., Yoganandan, N., Voo, L., 1998. Effect of age and loading rate on human cervical spine injury. Spine 23, 1957e1962. Prescher, A., 1998. Anatomy and pathology of the aging spine. Eur. J. Radiol. 27 (3), 181e195. Rao, S., 1962a. Metabolic cost of head-stand posture. J. Appl. Physiol. 17, 117e118. Rao, S., 1962b. Cardiovascular responses to head-stand posture. J. Appl. Physiol. 18, 987e990. Rao, S., 1968. Respiratory responses to head stand posture. J. Appl. Physiol. 24, 697e699. Walter, B.A., Korecki, C.L., Purmessur, D., Roughley, P.J., Michalek, A.J., Iatridis, J.C., 2011. Complex loading affects intervertebral disc mechanics and biology. Osteoarthr. Cartil. 19, 1011e1018. Wang, J.L., Parnianpou, M., Shirazi-Adl, A., Engin, A.E., 2000. Viscoelastic finite-element analysis of a lumbar motion segment in combined compression and sagittal flexion: effect of loading rate. Spine 25, 310e318. Woolery, A., Myers, H., Sternlieb, B., Zeltzer, L., 2004. A yoga intervention for young adults with elevated symptoms of depression. Altern. Ther. Health Med. 10, 60e63. Yoga in America Study, 2008. Harris Interactive Service Bureau. Yoga Journal. Yoganandan, N., Sances Jr., A., Maiman, D.J., Myklebust, J.B., Pech, P., Larson, S.J., 1986. Experimental spinal injuries with vertical impact. Spine 11, 855e860. Yoganandan, N., Sances Jr., A., Pintar, F., 1990. Injury biomechanics of the human cervical column. Spine 15, 1031e1039.
