J Manipulative Physiol Ther. 2016 (Mar); 39 (3): 192-199 ~ FULL TEXT
Marco Ehrler, B. Med., Cynthia Peterson, DC, M.Med.Ed., Serafin Leemann, DC,
Christof Schmid, DC, Bernard Anklin, DC, B. Kim Humphreys, DC, PhD
Chiropractic Medicine Student,
Chiropractic Medicine Department,
Faculty of Medicine, Orthopaedic University Hospital Balgrist,
University of Zürich, Zürich, Switzerland.
OBJECTIVE: The purpose of this study was to evaluate whether specific MRI features, such as axial location and type of herniation, are associated with outcomes of symptomatic lumbar disc herniation patients treated with spinal manipulation therapy (SMT).
METHODS: MRI and treatment outcome data from 68 patients were included in this prospective outcomes study. Pain numerical rating scale (NRS) and Oswestry physical disability questionnaire (OPDQ) levels were measured at baseline. The Patients Global Impression of Change scale, the NRS and the OPDQ were collected at 2 weeks, 1, 3, 6 months and 1 year. One radiologist and 2 chiropractic medicine master's degree students analyzed the MRI scans blinded to treatment outcomes. ? statistics assessed inter-rater reliability of MRI diagnosis. The proportion of patients reporting relevant improvement at each time point was compared based on MRI findings using the chi-square test. The t test and ANOVA compared the NRS and OPDQ change scores between patients with various MRI abnormalities.
RESULTS: A higher proportion of patients with disc sequestration reported relevant improvement at each time point but this did not quite reach statistical significance. Patients with disc sequestration had significantly higher reduction in leg pain at 1 month compared to those with extrusion (P = .02). Reliability of MRI diagnosis ranged from substantial to perfect (K = .733-1.0).
CONCLUSIONS: Patients with sequestered herniations treated with SMT to the level of herniation reported significantly higher levels of leg pain reduction at 1 month and a higher proportion reported improvement at all data collection time points but this did not reach statistical significance.
KEYWORDS: Intervertebral Disc Displacement; Lumbar Vertebrae; Magnetic Resonance Imaging; Manipulation; Spinal; Spine
From the Full-Text Article:
Lumbar disc herniations (LDH) occur commonly in the population. Not all people with herniations are symptomatic and the diagnosis and therapy must correlate with the patient’s complaints, physical examination findings and clinical history. [1, 2] The symptomatic herniations cause restrictions of functioning, suffering and costs and therefore deserve attention. Van Tulder et al state that LDHs can be the source of the most severe and disabling forms of back pain.  Gore et al showed that patients with chronic low back pain are characterized by greater economic burdens compared with those without.  This also applies to patients suffering with chronic LDH.
There are different causes and risk factors known for LDH. Genetics, family history, lumbar load, hard-working and time urgency are the major risk factors. [5-7] Poor stabilization due to weak muscles and other instabilities are additional possible explanations for a herniated disc. Physical exercises therefore may be protective. 
To confirm that someone’s symptoms are likely due to a herniated disc a magnetic resonance imaging scan (MRI) is necessary with the abnormal imaging findings compatible with the specific complaints of the patient. The indication for such an image is given by the anamnesis (patient history) and the presentation of the patient along with specific physical examination procedures including lumbar range of motion, dermatomal, myotomal and reflex changes and Lasègue sign.  To treat the condition it is essential to know what its effects are and where the herniation is located in terms of spinal level(s) as well as location in the axial plane. On the MRI scan the specific characteristics of the disc herniation are visualized as well as whether a nerve root is compressed.
Throughout the years different nomenclatures for classifying LDH were used. Today there are recommendations of the combined task forces of the North American Spine Society, the American Society of Spine Radiology and the American Society of Neuroradiology.  There are five different types or classifications of disc herniation: bulging, protrusion, extrusion, sequestration and intravertebral herniation (Schmorl’s node). The last one (Schmorl’s node) is of less interest, since these are usually not symptomatic and of relatively little clinical significance as they do not cause nerve root compression.  The other four will be described under the Methods section.
There are many different therapies available for the treatment of patients with symptomatic LDH. They can be grouped in the two categories surgical and conservative care. The outcomes from treatment of patients with symptomatic herniations are quite variable. So far there are no predictors for treatment outcome such as size, shape or location of the hernia. On the contrary, Masui et al showed, that clinical outcomes did not depend on the size of herniation or the grade of degeneration of the intervertebral disc.  However, spinal manipulative therapy (SMT) was not one of the treatments included in their evaluation. The recent study by Leemann et al showed that the vast majority of both acute and chronic MRI confirmed LDH patients treated with high-velocity, low-amplitude SMT reported clinically relevant improvement in their condition at multiple follow-up time periods up to 1 year, with no adverse events reported.  This current study is a follow-up to the Leemann et al study with the purpose to investigate whether the specific MRI features, such as axial location or type of herniation as identified on the MRI scans are associated with outcomes of patients treated with SMT to their spinal level of herniation.
The purpose of this study was to evaluate whether specific MRI features, specifically axial location and type (bulge, protrusion, extrusion, sequestration) of a herniated disc, are associated with the short and long term outcomes of patients treated with high-velocity, low-amplitude SMT specifically to the level of the symptomatic, MRI confirmed, herniation. This is the first study to address this question. Studies searching for predictors of improvement after treatment in previous low back pain patients did not target type and axial location of the herniated discs. [13, 16] Additionally, patients with disc sequestration were not excluded from this study as was done in the randomized controlled clinical trial (RCT) by Santilli et al on SMT in patients with disc herniation.  The only other RCT evaluating SMT in lumbar disc herniation patients does not even mention the issue of disc sequestration at all in their small sample size. 
Importantly, in terms of type of the herniated disc, we found that leg pain reduction at 1 month after the first treatment in patients with sequestration is significantly higher compared to patients with extrusion as well as a nearly significant difference between these two groups in the percentage of patients reporting clinically relevant improvement. Over 77% of patients with disc sequestration reported clinically relevant “improvement” compared to 66.7% of patients with extrusion. Although not statistically significant, 100% of patients with sequestration reported clinically relevant improvement at the 3 month data collection time point and at all data collection time points a higher proportion of patients with sequestration reported clinically relevant improvement. There were no significant differences for disc herniation location either by spinal level or in the axial plane for any of the data collection time points. This now calls into question the traditional thinking that disc sequestrations are more dangerous than herniations that remain attached to the parent disc and are more likely to require surgery. [19, 20] However, the studies reporting this did not consider chiropractic spinal manipulative therapy as a treatment option.
Although patients with foraminal location of their disc herniation had significantly less low back pain reduction at the 2 week time point compared to patients whose disc herniations were paracentral plus foraminal in location, it must be noted that only 4 of the 68 patients had a foraminal only disc herniation. Linking a patient’s clinical symptoms with MRI findings is important when determining appropriate treatments rather than basing treatment on MRI findings alone. [1, 2] Little research has been done on this but is starting to emerge. The recent study by Bensler et al found that patients with cervical disc extrusions compared to the other disc morphologies had less pain relief when treated with indirect nerve-root-blocks and were 4 times more likely to subsequently require surgery.  However, only 1 of the 112 patients in that study had a disc sequestration so direct comparisons between this lumbar disc herniation study, which had 22 patients with disc sequestration, and the Bensler et al study on cervical disc herniations cannot be done.
The mechanism as to why patients with sequestered disc herniations had better treatment outcomes compared to patients with disc extrusions is currently unknown and should be investigated. Perhaps these patients with disc sequestration have more normal biomechanics at the affected motion segment compared to patients whose herniations remain attached to the parent disc.  This current study supports SMT as a treatment in patients with all locations and types of disc herniation providing there are no signs of cauda equina compression or deteriorating neurology between treatments.
It is interesting to note that two different SMT procedures were used depending on the location of the disc herniation in the axial plane as determined by the MRI scans. These SMT procedures were developed based on the clinical experience and reflection of the senior treating chiropractor over time and then taught to the younger clinicians. After the 2 week data collection time point there were no significant differences in the proportion of patients reporting clinically relevant improvement based on the location of the herniation in the axial plane. This may indeed support the use of different SMT procedures based on the location of herniation but it does not prove that the improvement was due to the specific manipulative procedure. An RCT would be needed to randomly assign one of the two SMT procedures irrespective of the axial location of the disc herniation.
From the 148 patients at baseline in the study by Leemann et al we looked at the 68 patients whose MRI scans were available on the chiropractic practice computer system. This was a convenience sample and large enough to get some statistically relevant information but it is possible that with a larger sample size those results that were close to being statistically significant would indeed have become significant. Certainly important trends were noted. Thus, a larger sample size may have strengthened the relationship between herniation type, particularly sequestration, and treatment outcomes.
For outcome measures the NRS and OPDQ were used before treatment (baseline) as well as at 2 weeks, 1, 3, 6 and 12 months after treatment along with the primary outcome measure of the PGIC. For type and determination of axial location the recommendations of the combined task forces of the North American Spine Society, the American Society of Spine Radiology and the American Society of Neuroradiology were used, which suggest using the terms bulging, protrusion, extrusion and sequestration and the categories of central, paracentral, foraminal and extraforaminal.  These methods therefore were very adequate. Today however, the Bournemouth Questionnaire would be the preferred outcome measure instead of the OPDQ for these particular types of patients. It gives a broader picture of the limitations of a condition including psychosocial aspects, which are not included in the OPDQ. At the time the study was conducted by Leemann et al the Bournemouth Questionnaire for low back pain was not yet translated and validated into German. 
A possible limitation for the reliability part could be that the 3 persons interpreting the MRI slices were looking at the same screen simultaneously while one rater navigated through the MRI slides. Although none commented on what was seen until all 3 raters wrote down their findings, the way the navigator may have hesitated at certain slices could influence the other two observers. This study of course has some of the same limitations as the one by Leemann et al, which are that this is a cohort outcomes study rather than an RCT and that all patients were treated in a single chiropractic practice.  In this current study, unlike the Leemann et al publication, we did not make a difference between chronic and acute patients because in the original study the proportion of chronic patients reporting improvement was nearly as high as for the acute patients.
The majority of patients in this study had either extruded or sequestered disc herniations. Patients with sequestered herniations treated with SMT to the level of herniation reported significantly higher levels of leg pain reduction at 1 month and a higher proportion reported improvement at all data collection time points compared to patients with extruded disc herniations but this did not reach statistical significance. Further investigation is needed to determine mechanisms for this finding. This also calls into question the seriousness of disc sequestration in determining appropriate treatment.
Most patients had either lumbar disc extrusions or sequestrations.
A higher proportion of patients with sequestration reported clinically
relevant improvement at all data collection time points
Patients with disc sequestrations had significantly higher leg pain
change scores (ie, greater pain reduction) at 1 month
compared to patients with extruded disc herniations.
There were no adverse events from spinal manipulation to patients with
symptomatic, MRI confirmed, lumbar disc herniations.