Rohlmannt et al. (2001) wanted to compare intradiscal pressures and spinal fixator loads in different body positions and exercises in two particular studies by Nachemson (1966) and Wilke et al. (1999).

Below, the full discussion by Rohlmannt et al. (2001) has been copy-pasted, references have been included at the bottom of the post. 

"The results provide a comparison of the relative values of intradiscal pressure and fixator loads determined in two independent in vivo studies for several body positions and dynamic exercises. With the intradiscal pressure measured in one volunteer, some important differences were found compared with the earlier studies by Nachemson (1966, 1981). The limitation of this new study was that it was only done with the one volunteer. Therefore a comparison with loads acting in an internal spinal fixation device which were measured in up to 10 patients is essential, because few other data exist in the literature and many recommendations are based on findings from intradiscal pressure measurements from the 1960s. When lying down, the spine does not have to carry the weight of the trunk. Therefore, the intradiscal pressure and the bending moments in the fixators were found to be low. 

The small differences in the loads for the various lying positions are probably due to slightly differences curvatures of the spine. Slightly lower disc pressure and implant loads were found for sitting than for standing. Nachemson (1966, 1981) reported 40% higher intradiscal pressure values for sitting. The present results contradict this but are in agreement with those of another indirect method for load measurement using audiometry. Althoffm et al. (1992) found an increase in body height when the subjects were sitting after standing for a while. This indicates that the spinal load is lower for sitting than for standing. The differences between the results of Wilke et al. (1999) and Nachemson (1966) may be explained by the use of different pressure transducers. Nachemson’s transducer was integrated in a stiff needle which might have measured artefacts if it was bent due to muscle contraction. The transducer used by Wilke et al. could not be bent because the stiff part was only 7 mm long and was inserted completely into the nucleus of the disc where only a hydrostatic pressure is present in case of a non-degenerated disc. Sitting consciously erect, actively straightening and extending the back, increased the pressure in the disc and the flexion bending moments in the fixators. Higher muscle forces are needed for sitting straight than for sitting relaxed. Higher muscle forces in turn lead to higher spinal loads. 

The load differences were, however, small and the slightly higher pressure is no argument against sitting erect. Intradiscal pressure and fixator loads were in the medium range during sitting in the different positions. Other unpublished measurements taken by the authors showed that these loads could be reduced greatly when leaning against a backrest since the backrest takes over part of the load. It seems that for low back pain the amount of the global load during sitting is not the crucial factor since the measured loads were lower for sitting than for standing and during walking. However, little information exists about the loads on the facet joints and on the ligaments, which may be high for sitting. For the nutrition of the disc, intersegmental movements are probably very important and people, especially those suffering from low back pain during sitting, are recommended to change their posture frequently. 

Changing from a lordotic shape of the spine to a kyphotic shape and vice-versa has only a minor influence on spinal load but is probably advantageous for disc nutrition and may reduce low back pain. When supporting the body only with the hands, as during body lifting in a sitting position or during balancing the body on parallel bars, the spine has to carry not the body weight above, but the body weight below a certain level. This exercise led to nearly the same low pressure and bending moments in the fixators as lying in a supine position. The pulling force on the spine resulting from gravitational force on the body part below the level of interest was obviously compensated by muscle forces. Dynamic motion, as during walking, jogging, stair climbing, bouncing on a physiotherapy ball, jumping on a trampoline or skipping, led to higher spinal loads than normal standing. However, the loads in these cases were often significantly smaller than for flexion of the upper part of the body or lifting a weight. 

The impulses on the feet during the dynamic exercises are mainly damped in the foot and knee joint and by the curved shape of the femur and obviously do not reach the spine. Measurements with instrumented hip endoprostheses have shown that such impulses normally do even not reach the hip (Bergmann et al. 1995). The dynamic load component at the spine is mainly caused by the accelerations of the upper part of the body. The mass of the upper part of the body moves up and down during these exercises and is therefore responsible for the dynamic component of the spinal load. Walking with crutches led to higher disc pressure than walking without crutches. However, the volunteer loaded the crutches very `dynamically’. Implant loads were only briefly reduced when the patients were asked to load the crutches strongly. The use of one crutch often led to higher implant loads than walking without a crutch, since the patients bent the upper part of the body laterally, which increased the load on the ipsilateral fixator (Rohlmann et al. 1997). From these in vivo measurements it is concluded that it does not seem necessary for patients with back problems, for example after implantation of spinal fixators, to use crutches when they can walk safely with respect to losing balance.

The load on the trunk is significantly increased during flexion of the upper part of the body as well as when lifting and carrying a weight. These exercises strongly increased intradiscal pressure but had only a minor effect on the flexion bending moments in the fixators. In the region bridged by a fixator, the spinal load is shared by the spine and the implant. Further bending has only a minor influence on the bending of the fixator rods when there is anterior bony support. Therefore activities during bending forward might show lower loads in the fixator compared with the intradiscal pressure. How this load is shared depends mainly on the stiffness of the bridged region. This stiffness depends, among other reasons, on the surgical procedure, the time after surgery, and the weight carried. Spinal stiffness is normally higher in an upright than in a lying body position. This is due to the higher load, which probably leads to a better interlocking of the facets (Wilke et al. 1995). If a spinal fixator is flexed to a spinal segment the whole system is stiffness than the fixator itself. In an upright body position, the stiffness of the bridged region is therefore higher than that of the fixator alone. The vertebral bodies are mainly loaded in compression while the fixators are loaded predominantly in bending. 

Crouching on hands and knees led to an intradiscal pressure of only 40% of that when standing and a flexion bending moment in the fixator of 72%. This is astonishingly low. Only for lying positions and when the body weight was balanced on the hands were lower disc pressures and lower fixator loads measured. Obviously only low muscle forces are needed for stabilising the back in this position. Arching and hollowing the back while supporting the body on hands and knees leads to a great deformation of the spine. Hollowing the back caused a pressure increase, but the maximum value for this posture was still less than for standing. The bending moment in the fixators was also lower than for standing. 

The increase of disc pressure for arching the back was higher than for hollowing the back but lower than during flexion of the upper part of the body while standing. The fixator load for arching was nearly the same as for standing. The following limitations of these comparisons have to be noted. In principle it would be best to apply both techniques in the same subjects, but ethical reasons did not of course allow this. Intradiscal pressure depends on several factors including disc quality, disc level, body weight and others. 

Although intradiscal pressure measurements were performed in only one subject, the relative results for standing and sitting are confirmed by published reports of indirect measurements using stadiometry (Althoffm et al. 1992). Loads on internal spinal fixators depend mainly on indication for surgery (compression fracture, degenerative instability), surgical procedure (distraction or compression of the bridged region), and level of bridged vertebra (Rohlmann et al. 1999, 2000). These factors varied in the patients, but for most body positions and activities the trends in the bending moments for the different exercises were similar for all patients. Small differences in the relative loads have probably only minor relevance since intradiscal pressure was measured in only one volunteer and fixator loads often in only one patient. For the standard activities measured in 10 patients the standard deviation was relatively large."

Rohlmann et al (2014) has produced a newer study which makes interesting reading! 

REFERENCES:
--------------------
Althoffm et al. (1992): An improved method of stature measurement for quantitative determination of spinal loading, Spine, 17, 683-693. 

Bergmann et al. (2000): Influence of load carrying on loads in internal spinal fixators, Journal of Biomechanics, 33, 1099-1104. 

Nachemson (1966): http://journals.lww.com/…/The_Load_on_Lumbar_Disks_in_Diffe…

Nachemson (1981): http://www.ncbi.nlm.nih.gov/pubmed/7209680

Rohlmannt et al. (1997): Loads on an internal spinal fixation device during walking, Journal of Biomechanics, 30, 41-47.

Rohlmannt et al. (1999): Loads on internal spinal fixators measured in different body positions, European Spine Journal, 8, 354-359. 

Rohlmannt et al. (2000): Influence of load carrying on loads in internal spinal fixators, Journal of Biomechanics, 33, 1099-1104. 

Rohlmannt et al. (2001): https://www.researchgate.net/…/11890544_Comparison_of_intra…

Rohlmann et al (2014): http://journals.plos.org/plosone/article…

WIlke et al. (1995): Stability increase of the lumbar spine with diŒerent muscle groups, Spine, 20, 192-198.