Sleep posture - be wary of anecdotal opinion

We have all had that client rock into the clinic with a 'spasmed' neck, that came on soon after waking. It was an amazing fact that 9 years ago when I started looking into relationships between sleep posture and waking spinal pain, I couldn't find any research to guide me. Ah I thought, now that is interesting.....plenty of anecdotal stuff, but nothing on which to base clinical advice.

"Herniated lumbar disc. The preferred sleep position depends in part on the position of the disc. For a paracentral disc herniation (most common), people tend to do better lying on the stomach." Complete online article written by Dr Richard A. Staehler, MD can be found here.

Really! Clinically would you tell anyone to sleep prone? I even avoid palpating in prone when clients have lumbar spinal pain. Here are three reasons I don't think it is such a good idea to recommend sleeping in prone.

Two mechanisms have been associated with IVD injury and degeneration. Firstly, excessive mechanical loads and secondly, impaired nutritional supply.

1. Excessive Load

In essence, sustained postures load unsupported tissues and under load these tissues have demonstrated the expression of pro inflammatory chemicals, associated muscle spams (protective) and complaints of pain. Subject to the conditions of sustained load, extension could result in viscoelastic creep in spinal tissues (Panjabi, Krag, White, & Southwick, 1977) and two possible outcomes have been proposed. Firstly, that the motion segment rotates further and increases the load on the anulus fibrosus with possible damage (Farfan & Gracovetsky, 1984), or secondly, the motion segment does not rotate further and additional load is borne by the compressed ZPJ (White & Panjabi, 1978). It is likely that degrees of both mechanisms occur, with the dominant tissue resisting the applied torque varying according to age (increasing age, increasing IVD load) and degree of pre-compressive load (less compression, increased IVD load) (Krismer et al., 1996). 

The biomechanical concepts of creep, hysteresis and fatigue failure following sustained or repeated cycles of loading and unloading, have important clinical consequences with respect to sleep postures. While creep is a reversible and an entirely normal physiological response to load, repeated and/or sustained loads do result in tissue changes like hysteresis and fatigue failure, and have been associated with abnormal muscle activity. From a young age, humans develop and consolidate a regular routine of repeated sleeping postures throughout the night, involving both sustained and repeated postures. Authors have proposed that the repeated application of low loads over time to the spine, can have a cumulative effect and overload collagenous tissues (Kumar, 2001; Marras, 2000). It has been noted, that participants over 18 years of age, on average change their posture between 2.1 and 3.6 times per hour (De Koninck et al., 1992). This indicates that sleep postures are commonly sustained for periods of greater than 15 minutes, a time frame in which in vitro (Yahia et al., 1991) and in vivo (Solomonow, Baratta, et al., 2003) viscoelastic collagenous tissue creep has been demonstrated.

Some sleep postures will be supportive of collagenous spinal tissues, while others will increase the load on spinal structures. Relationships between biomechanical creep and spinal symptoms in relation to sustained and repeated low loading, have been extensively explored using in vivo designs with healthy human and feline groups (Solomonow, 2012; Solomonow, Baratta, et al., 2003; Solomonow, Zhou, et al., 2003; Solomonow, Zhou, Lu, & King, 2012). Under sustained rotation loading for 10 minutes in sitting, viscoelastic creep was demonstrated in vivo in asymptomatic males (N = 16) (Shan et al., 2013). In addition to observing a significant increase in trunk axial rotation (M = 10.50, SD = 5.2, p < .001), the authors also reported a significant increase in lumbar pain measured using a visual analogue scale (VAS) (M = 45.1, SD 10.6, p < .001). In another experiment, asymptomatic human participants remained in long sitting for 10 minutes and more than 50% of participants consequently developed muscle spasms (Solomonow, Baratta, et al., 2003). The authors suggested this was indicative of micro-damage to viscoelastic tissues and explored this relationship in feline spines. After calibration, feline spines were exposed to 10 minutes of sustained flexion (20 and 60 Nm), followed by 10 minutes rest which was repeated six times after which the spines were rested for 7 hours (Solomonow, 2012). The author reported significant increases in a range of pro-inflammatory chemicals expressed during the recovery period, indicating acute inflammation and tissue damage in ligaments subjected to the cyclic loading, compared with the control ligaments from the same spine. There were also significant increases in muscle activity during the recovery period, which increased with increases in load, increases in the number of repetitions, and decreased duration of rest period between loading periods.

So young or old (worse old), sleeping in a soft tissue supported/resisted posture is going to load your IVDs more.

2. Impaired Nutritional Supply. 

Lacking a direct blood supply, the IVD is largely avascular and relies on nutrition diffusing from the superficial anular fibres and the vascular tissues of the vertebral end plate (Adams et al., 2006). The supply of nutrition to the IVD is considered barely adequate at the best of times (Urban, Holm, Maroudas, & Nachemson, 1982) and impaired nutritional supply has been associated with cell death and degenerative changes in IVDs (Bibby, Fairbank, Urban, & Urban, 2002; Horner & Urban, 2001). During the day, gravity, muscle contraction and movements associated with upright postures, produce compressive loads that push fluids out of the IVD at a rate of approximately 20% over 6 hours (McMillan et al., 1996). Fluid movement occurs via the vertebral end plate and the peripheral anular fibres. When lying down, the diurnal pressure gradient is reversed and fluid along with nutrients diffuse back into the IVD. The rate of diffusion of nutrients into an area of the IVD is influenced by the distance to the nearest blood vessel. This distance is called the diffusion path length. Flexion postures increase anterior IVD compression and thickening (0.1 mm/degree of movement), while at the same time stretching and thinning occurs in the posterior anulus (Stokes, 1988). However, sleeping postures that involve lordosis (e.g., prone) will potentially have a negative effect on the pressure diffusion gradient of the posterior IVD by increasing the diffusion path length, due to a thickened posterior IVD, resulting in a reduced nutrient inflow.

3. Concomitant pathologies

Extension is known to reduce central and lateral canal diameter and therefore potentially compromise neural structures. Nocturnal exacerbation of neuropathic symptoms that disrupt sleep can occur in people with spinal stenosis, with or without peripheral neuropathy (Goldman, 2005). In this study, participants were advised to use a walking frame during the day to encourage lumbar flexion and to sleep in supine (pillow under knees) or side lying (pillow between knees) or prone (pillow under stomach), in an attempt to limit lumbar extension. Improvement of symptoms in eight of 11 participants was reported, commonly with significant symptom reduction within 24 hours (Goldman, 2005). How compliance with postural recommendations was monitored, was not reported. However, the speed with which symptom control was improved in a chronic condition is noteworthy. It implicates a biomechanical component in the symptom provocation, that when postural loads are altered is reversible, even in a chronic condition.

Prone has been noted to aggravate 49% of participants with existing lumbar pain (Boissonnault & Di Fabio, 1996) and avoiding prone is commonly recommended (Bland, 1987; Grieve, 1988; Kraemer, 2011). With increasing age, prone is adopted less frequently (De Koninck et al., 1992; Gordon et al., 2007a) which may represent a learned protective response. As clinicians we are supposed to "do no harm", and I would suggest the advice given in the above mentioned online article runs contrary to good clinical practice.

What do you think?

Just completed and being presented for the first time in July in the Maldives, you can express your interest in attending our Sleep Right Sleep Tight course for health professionals here.