If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Magnetic Resonance Imaging Zygapophyseal Joint Space Changes (Gapping) in Low Back Pain Patients Following Spinal Manipulation and Side-Posture Positioning: A Randomized Controlled Mechanisms Trial With Blinding
Submit requests for reprint to: Gregory D. Cramer, DC, PhD, National University of Health Sciences, Department of Research, 200 East Roosevelt Road, Lombard, IL 60148.
The purpose of this study was to quantify lumbar zygapophyseal (Z) joint space separation (gapping) in low back pain (LBP) subjects after spinal manipulative therapy (SMT) or side-posture positioning (SPP).
Methods
This was a controlled mechanisms trial with randomization and blinding. Acute LBP subjects (N = 112; four n = 28 magnetic resonance imaging [MRI] protocol groups) had 2 MRI appointments (initial enrollment and after 2 weeks of chiropractic treatment, receiving 2 MRI scans of the L4/L5 and L5/S1 Z joints at each MRI appointment. After the first MRI scan of each appointment, subjects were randomized (initial enrollment appointment) or assigned (after 2 weeks of chiropractic treatment appointment) into SPP (nonmanipulation), SMT (manipulation), or control MRI protocol groups. After SPP or SMT, a second MRI was taken. The central anterior-posterior joint space was measured. Difference between most painful side anterior-posterior measurements taken postintervention and preintervention was the Z joint “gapping difference.” Gapping differences were compared (analysis of variance) among protocol groups. Secondary measures of pain (visual analog scale, verbal numeric pain rating scale) and function (Bournemouth questionnaire) were assessed.
Results
Gapping differences were significant at the first (adjusted, P = .009; SPP, 0.66 ± 0.48 mm; SMT, 0.23 ± 0.86; control, 0.18 ± 0.71) and second (adjusted, P = .0005; SPP, 0.65 ± 0.92 mm; SMT, 0.89 ± 0.71; control, 0.35 ± 0.32) MRI appointments. Verbal numeric pain rating scale differences were significant at first MRI appointment (P = .04) with SMT showing the greatest improvement. Visual analog scale and Bournemouth questionnaire improved after 2 weeks of care in all groups (both P < .0001).
Conclusions
Side-posture positioning showed greatest gapping at baseline. After 2 weeks, SMT resulted in greatest gapping. Side-posture positioning appeared to have additive therapeutic benefit to SMT.
A fundamental hypothesis of a beneficial effect of chiropractic spinal manipulative therapy (SMT) is that adhesions developing in hypomobile zygapophyseal (Z) joints are broken during SMT by gapping of the Z joint articular surfaces
Fig 1Theoretical model of one of the beneficial effects of SMT (spinal adjusting). This project assessed step 3 of the model, separation of the Z joint articular surfaces (therapeutic gapping).
Vertebral segmental hypomobility has been identified clinically, and low back pain (LBP) patients with identified vertebral hypomobility have been found to respond more favorably to SMT than those without hypomobility.
Putative reasons for Z joint hypomobility include inactivity; injury; or repetitive, asymmetric motions (eg, assembly line work). Such repetitive motions would tend to result in normal or increased movement of some of the Z joints while chronically loading others. The joints receiving the long-term loading would likely become relatively hypomobile.
Fibrous adhesions are thought to develop in hypomobile Z joints, further preventing normal joint motions.
have been quantified in hypomobile animal Z joints (Fig 1, step 2). Gapping of the Z joints is thought to break-up intra-articular adhesions that have developed during hypomobility and aid in re-establishing normal range of motion to the Z joints (Fig 1, steps 3-5).
and more recently, SMT and side-posture positioning (SPP) have been shown to gap the lumbar Z joints in healthy human volunteers, with SMT resulting in greater gapping than SPP alone.
Effects of side-posture positioning and side-posture adjusting on the lumbar zygapophysial joints as evaluated by magnetic resonance imaging: a before and after study with randomization.
However, no previous studies assessed Z joint gapping in clinical (LBP) patients.
The study reported here was designed to determine whether Z joints gap during lumbar side-posture SMT and SPP in acute LBP patients (Fig 1, step 3). Zygapophyseal joint gapping was assessed from magnetic resonance imaging (MRI) scans taken at initial presentation (M1) and after 2 weeks of chiropractic care (M2). Secondary outcomes assessing pain and functional impairment were also included.
Methods
Project Overview
This controlled mechanisms trial with randomization and blinding used 4 MRI protocol groups (SPP, SMT, and 2 control groups) to assess a component of one of the proposed mechanisms of SMT. Figure 2 shows the general overview of the study. The study was not designed to assess the effectiveness of spinal manipulation as a treatment; other investigators are conducting such effectiveness studies
A practice-based study of patients with acute and chronic low back pain attending primary care and chiropractic physicians: two-week to 48-month follow-up.
; this study was designed to assess gapping of the Z joints with SPP and SMT in LBP subjects. All subjects in this study received the same modalities of care during the treatment phase of the project.
Fig 2Flowchart showing the overview of the project. BQ, Bournemouth Questionnaire of functional impairment,
This study was approved by the Institutional Review Board of the National University of Health Sciences (IRB no. H-0107) and was registered with the US NIH Clinical Trial Registry (NCT00284063).
Screening Examination
Table 1 shows the inclusion and exclusion criteria used at the screening examination. The acute LBP subjects
included in this study closely matched the patients described as “Category 1” (more specifically, categories 1a and 1b) of the Quebec Task Force classification.
Quebec Task Force on Spinal Disorders. Scientific approach to the assessment and management of activity-related spinal disorders: a monograph for clinicians.
Each subject's most painful side (primary treatment side [PTS]) was determined at the examination. The treating clinician (DG) asked the patient to describe her/his pain and to identify the most painful side. The subject's reported most painful side became the PTS and did not change throughout the study. The PTS was the up-side during all SMT or SPP during M1 and M2 appointments.
Table 1Inclusion and exclusion criteria during the screening examination
Inclusion
Exclusion
21 to 69 years old (21 years to ensure fully developed Z joints and <70 years to tolerate side-posture MRI scans) Females, ≤160 lb or BMI of ≤28; males ≤200 lb or BMI of ≤30 (to ensure optimum MRI quality) Pain related to the low back (lower lumbar region, L4/L5, and/or L5/S1 region)—this criterion was determined by the examining physician through subjective complaint and description as well as objectively using inspection, palpation, motion assessment, and standard orthopedic and neurologic tests such as Kemp's, Milgram's, Yeoman's, straight-leg raise, and Valsalva maneuver. A history of LBP lasting for a period of ≤6 wk;
also defined as having ≥1 mo pain free between current and previous episodes of LBP; must have had more pain free days than days with LBP in the past year.
<21 or ≥70 years old (see inclusion criteria) Weighs >160 lb or BMI >28 (if female) or >200 lb or BMI >30 (if male) (subject weighed at baseline examination) Presence of lumbar scoliosis of >5° (Cobb's angle) (due to difficulty in imaging the Z joints) Presence of radiculopathy (This criterion was evaluated by the examining physician by using patient history, standard screening tests, and the results of a detailed orthopedic/neurologic evaluation.) Cauda equina symptoms such as perianal numbness, loss of bowel, and/or bladder control (This criterion was evaluated by the examining physician.) Spine deformity such as current spinal fractures, spinal infections, or tumors of the spine Current history of severe osteoporosis Prior lumbar spine surgery No pain related to L4/L5 and/or L5/S1 region (This criterion was determined by the examining physician through subjective complaint and description as well as objectively using inspection, palpation, motion assessment, and standard orthopedic tests.) Pregnancy or currently breastfeeding (for MRI, although no known risk, and in the event an x-ray is needed to screen for contraindications to manipulation) Intolerance to MRI procedures (including claustrophobia and inability to lie on one's side for 15 min). Claustrophobia will be evaluated before and during the first and second MRI scans. Other significant pathology discovered on MRI scans, as observed by reading radiologist. (This criterion was evaluated during the first MRI visit, immediately after the first MRI scan was taken. Such pathologies may constitute contraindications to chiropractic SMT.) Absence of acute LBP (See “Inclusion criteria,” for definitions of acute LBP.) Current or future litigation for LBP (work injury or motor vehicle accident) Psychiatric illness or lack of cognitive ability (ie, dementia or Alzheimer) Current and known substance abuse Not fluent or literate in English
BMI, body mass index; LBP, low back pain; SMT, spinal manipulative therapy.
Effects of side-posture positioning and side-posture adjusting on the lumbar zygapophysial joints as evaluated by magnetic resonance imaging: a before and after study with randomization.
Each of the 112 subjects received 2 MRI scans (Hitachi MRP 5000, 0.2-T MRI unit, Hitachi Medical Systems America, Inc, Twinsburg, OH) on 2 separate occasions, the M1 and M2 (Fig 3, Fig 4).
Fig 3Enrollment and exclusion of study subjects. Exclusions are listed along the right side of the flowchart. Withdrawals during the 2-week treatment period and second MRI scan are summarized in the last row of boxes (protocols 1-4). The following abbreviations are used in the bottom row of boxes: neutral, supine position; side, SPP. All SMT was performed with the most painful side (PTS) as the up-side.
Fig 4Procedures used for each of the 4 study protocol groups. The protocols are described in the “MRI Scanning” subsection of the Methods. Notice that all protocols began with an MRI scan in the neutral position (first row). Subjects were then randomized into 1 of 4 protocol groups (second and third rows) and were then scanned a second time in either the neutral or side-posture position (third row). Although subjects are shown receiving SMT or SPP with the left side as the up-side, the up-side was the subject's most painful side (PTS) at the examination appointment, which was frequently the right side (see Table 3).
Figure 4 shows the protocols used for the 4 protocol groups of the study. The most painful side, the PTS, was always the up-side for SPP or SMT. The 4 protocol groups were as follows: protocol 1 (SPP group): neutral positioning, followed by SPP, remaining in SPP for second MRI scan; protocol 2 (SMT control): neutral positioning, followed by side-posture SMT, followed by neutral positioning for second MRI; protocol 3 (SMT group): neutral positioning, followed by side-posture SMT and remaining in side-posture for second MRI; and protocol 4 (SPP control, primary control): neutral positioning, followed by brief SPP, followed by neutral positioning for second MRI.
Magnetic resonance imaging scans were taken with the subjects in the original neutral position and in the final position. The first scan of each MRI appointment was taken in the neutral (supine) position. This allowed for a baseline Z joint space (gapping) measurement to be obtained for each subject. The initial (neutral position) MRI was followed by an intervention (side-posture SMT or SPP), which was immediately followed by a second scan. The second MRI scan was taken either back in the supine (control protocol groups 2 and 4) or side-posture position (protocol groups 1 and 3). Imaging in the neutral position for the second MRI scans of protocol groups 2 and 4 served as controls for the SPP and SMT protocol groups imaged in the side-posture position for the second MRI scan (protocol groups 1 and 3). Previous studies showed that reloading the spine by placing the subject in the neutral position following the intervention resulted in no Z joint gapping difference between the first and second MRI scans.
Effects of side-posture positioning and side-posture adjusting on the lumbar zygapophysial joints as evaluated by magnetic resonance imaging: a before and after study with randomization.
Consequently, groups 2 and 4 were the control groups in the study (neutral position for the second MRI scan). The 2 MRI scans and interposed intervention were conducted over approximately 30 minutes (12:19 minutes per MRI scan; thus, SPP was held for approximately 12 minutes during the second MRI scan of protocol groups 1 and 3). All interventions were performed directly on the MRI gantry table after the first scan (Fig 3, Fig 4).
First MRI appointment
The M1 appointment was conducted before any treatment began. Fig 3, Fig 4 summarize the design of the appointment.
During the M1 appointment, the subjects were randomized into 1 of 4 MRI protocol groups (see MRI Scanning, above). A technician not involved in patient contact used a random number generator to develop the randomization scheme. If a subject was eligible for study participation after the first scan of the M1 appointment, the scanning radiologist would leave the area to be blinded to SMT or no-SMT intervention, and the clinical research assistant pulled the next male or female randomization envelope from a safe. The research clinician then performed the SMT and/or positioned the subject (SPP) according to randomized protocol. Once the clinician had completed the protocol with the subject remaining on his/her side, the scanning radiologist was called back and was told whether the subject should be positioned in the neutral or side-posture position for the second MRI scan.
Effects of side-posture positioning and side-posture adjusting on the lumbar zygapophysial joints as evaluated by magnetic resonance imaging: a before and after study with randomization.
was also used in this trial (Fig 4, center row protocols 2 and 3). One intent of the procedure is to open (gap) the up-side targeted joints, in this study, the L4/L5 and L5/S1 Z joints.
Two Weeks of Treatment
After the M1 appointment, all subjects received chiropractic care for 2 weeks (1-3 visits per week as recommended by the treating clinician). The care included SMT and other modalities as deemed appropriate, including hot moist packs, ultrasound, and/or interferential nerve stimulation. SMT only was provided during the MRI appointments. Analgesic use between appointments was recorded at every appointment. Study participants were asked to avoid any other form of care for his/her low back the 2 days before their MRI appointments. The subjects were also asked not to engage in heavy lifting (eg, weight training) or prolonged walking or jogging during the same period. These recommendations were made to avoid excessive loading of the Z joints for the 2 days before the MRI appointments. These were the only restrictions placed on subjects regarding outside care. Outside care was tracked at every visit (see Results for a description of the 3 subjects who sought outside care).
Second MRI Appointment
The M2 appointment occurred after 2 weeks of treatment (Fig 2). The M2 appointment was identical to the M1 appointment with the exception that each subject was assigned to the protocol group “opposite” the one to which she/he was randomized at the M1 appointment. That is, M1 protocol 1 was assigned to M2 protocol 2 and vice versa, and M1 protocol 3 was assigned to M2 protocol 4 and vice versa. This way, during the study, all subjects were in both an intervention and control group, and all subjects were in an SPP and SMT group. Completion of the M2 appointment signified the completion of data collection for this study. After the M2 appointment, subjects were provided as needed care for up to 2 additional weeks; however, no information entered into their files during this final 2 weeks of care was used in the study.
Morphometry
The methods used to choose the MRI images to be measured and the procedures for making the measurements of the Z joint space from the images were described previously.
Reliability of zygapophysial joint space measurements made from magnetic resonance imaging scans of acute low back pain subjects: comparison of 2 statistical methods.
A trained radiologist marked the Z joint images to be measured. During MRI, the scans were coded using random numbers so that all investigators, including the radiologist, were blinded to protocol group, MRI appointment (first or second), MRI scan (first or second for each appointment), and all subject identifying information.
Using the procedures described and found to be reliable in a previous study,
Reliability of zygapophysial joint space measurements made from magnetic resonance imaging scans of acute low back pain subjects: comparison of 2 statistical methods.
3 trained observers measured the central anterior-posterior (A-P) Z joint space (Fig 5) on the coded scans. Measurements were made of the left and right L4/L5 and L5/S1 levels using a backlit GTCO Calcomp Drawing Board III digitizer (Source Graphics, Anaheim, CA) and were rounded to the nearest 0.1 mm. The observers were blinded to the other observers' measurements.
Fig 5Illustration (A) and MRI scan (B) showing the central A-P measurement of the Z joints that were made from the left and right L4/L5 and L5/S1 Z joints in this study.
After all measurements were completed, the identification codes of the MRI scans were broken, and the measurements from the first and second scans for each subject were paired. The value for each joint obtained from the first scan (preintervention) was subtracted from that of the second (postintervention) scan to yield the “gapping difference” for each joint. A positive gapping difference indicated an increase in gapping after the SMT or SPP. Spinal manipulative therapy is not as specific as was previously supposed.
Before the project began, the investigators determined that the L4/L5 or L5/S1 Z joint on the PTS with the largest gap was the segment that received the primary force of the manipulation or the primary torque of the SPP. Therefore, the segment with the largest gap was used as the primary outcome for each subject for all of the protocol groups, including the controls (protocol groups 2 and 4); the greatest gapping difference on the PTS (GGDPTS) was the primary outcome of the study and was calculated for all subjects in all 4 MRI protocol groups for both the M1 and M2 appointments.
Visual Analog Scale
A 100-mm visual analog scale (VAS) was used at the beginning of the M1 appointment to assess LBP (anchors, “No Pain” and “Worst Pain Imaginable”).
Chiropractic chronic low back pain sufferers and self-report assessment methods. Part I. A reliability study of the Visual Analogue Scale, the Pain Drawing, and the McGill Pain Questionnaire.
(VNPRS) was administered twice at each MRI appointment. After the initial, neutral-position MRI scan, a research assistant asked subjects to rate their current pain on a scale of 0 to 10 with 0 being no pain and 10 being the worst pain imaginable. The subject then received an intervention (SMT and/or SPP) while remaining on the MRI gantry table. The subject provided a second VNPRS after the intervention, which was immediately before the second MRI scan.
Bournemouth Questionnaire
Function was assessed at the beginning of the M1 appointment by the reliable, 7-question Bournemouth questionnaire (BQ) that scores impairment of function as a result of LBP on a scale of 0 to 70 (0, no impairment; 70, maximum impairment).
Sample size was determined by conducting a power analysis using the gapping differences and data variability of a previous study on healthy subjects and a pilot study on acute and chronic LBP patients.
The effects of side-posture adjusting on the lumbar zygapophysial joints of low back pain patients as evaluated by magnetic resonance imaging: a preliminary study.
The GGDPTS values of each MRI appointment were analyzed to determine if differences existed between the 4 protocol groups (analysis of variance [ANOVA] with Tukey-Kramer post hoc analysis). The Kruskal-Wallis [KW] test (ANOVA for nonparametric data) and Dunn's post hoc analysis were used when the data did not pass the normality test. Data were graphed and assessed for outliers. If outliers were found, secondary analyses were conducted with the outliers removed. Greatest gapping difference of the PTS analyses were conducted for both M1 and M2 data.
Greatest Gapping Differences of Males vs Females
Previous authors have emphasized the importance of assessing the influence of sex on measurements of anatomical structures of the spine
; consequently, GGDPTS of males and females for all protocol groups were assessed together using 2-sided t tests. If a difference was found, then subanalyses were conducted for each of the 4 protocol groups. These analyses were conducted for both M1 and M2 data.
Pain, Function, and Gapping at M2 Appointment (After 2 Weeks of Treatment)
Visual analog scale and BQ data were assessed for all subjects before and after 2 weeks of treatment (ie, at the beginning of the M1 and M2 appointments) to give an indication of the overall change in pain and functional impairment, respectively. This was done by descriptive statistics and 2-sided t tests.
Difference in Pain (VNPRS) Between the First and Second MRI Scans of Each MRI Appointment (M1 and M2)
Verbal numeric pain rating scale differences before and after intervention were calculated for each MRI appointment (M1 and M2). Verbal numeric pain rating scale differences among the 4 MRI protocol groups were then compared using ANOVA. This was done separately for M1 and M2 data to determine changes in pain due to the intervention that took place at each MRI appointment.
Results
Figure 3 shows the flow of subjects through the study and includes numbers of subjects enrolled and numbers of subjects excluded or dropped out at each stage of the study. Table 2 shows the total number of exclusions for the study. Many subjects met multiple exclusion criteria; consequently, the number of exclusions in Table 2 is much higher than the number of subjects excluded, as shown in Figure 3.
Table 2Reasons for subject exclusions
Exclusion reason
Telephone screen
Baseline visit
MRI visit
Total excluded
<21 years old or >69 years old
27
0
0
27
Weight/BMI above allowable criteria
312
36
0
348
No LBP or inability to reproduce pain at L4/L5 and/or L5/S1
50
16
0
66
Chronic LBP or recurrent episode of LBP >6 wk
1273
71
0
1344
Scoliosis >5°
7
3
3
13
Presence of radiculopathy
147
45
0
192
Claustrophobia or other intolerance to MRI procedures
25
0
3
28
Pregnant or nursing
3
0
0
3
Transitional L4/L5 or L5/S1 segment
0
0
17
17
Severe arthritic change or osseous bridging
0
0
14
14
Disc protrusion/extrusion >5 mm
0
0
26
26
Spondylolysis or spondylolisthesis
0
0
12
12
Other significant pathology or contraindication to study participation
Recruitment and enrollment lasted 22 months. Seven withdrawals occurred after the first MRI appointment. The reasons for the withdrawals included: not showing up for scheduled appointments or no longer interested in continuing (5), claustrophobia during the second M2 appointment (1), “not liking” the SMT (1). These withdrawals were evenly distributed among the MRI protocol groups (Fig 3). Because an important aspect of the study was to compare overall study outcomes at the M1 and M2 appointments (M2 after 2 weeks of care), MRI measurements were not taken from the M1 scans of withdrawn subjects. Enrollment was carefully tracked and intentionally slowed when only a few subjects were needed to reach the goal of 112 subjects. The study ended when the final subject completed the scheduled 2 weeks of treatment following the second MRI appointment.
First MRI Appointment
Subject Characteristics
Table 3 and the M1 (left) side of Table 4 show the subject characteristics for the 4 randomized protocol groups at M1. More males (n = 75) were enrolled than females (n = 37). With the possible exception of analgesic use (Table 4), subject characteristic differences among the protocol groups were not considered important to the outcome of the study. Subsequent analyses also found that analgesic use was not related to gapping (see Discussion).
Table 3Characteristics of subjects at first MRI appointment
Subjects were assigned to the protocol group opposite the protocol group to which they were randomized in the M1 appointment. The M1 protocol group is in parentheses.
13.8 (15.0)
14.4 (9.2)
4 (14%)
2
26.1 (15.7)
22.9 (10.9)
7 (25%)
2 (1)
17.1 (17.3)
13.8 (12.2)
7 (25%)
3
32.3 (22.0)
24.5 (11.7)
9 (32%)
3 (4)
15.7 (14.3)
12.4 (9.8)
8 (29%)
4
26.2 (13.6)
23.9 (11.1)
8 (29%)
4 (3)
18.9 (20.7)
13.2 (11.8)
7 (25%)
M1, initial presentation; M2, presentation after 2 weeks of chiropractic care.
a n = 28 for each protocol group, total N = 112.
b Visual analog scale scores (possible range, 0-100) at the beginning of the MRI appointment (mean [SD]).
c Bournemouth questionnaire of functional impairment scores (possible range, 0-70) at the beginning of the MRI appointment (mean [SD]).
d Subjects taking analgesic medications (subject count [percentage of subjects]).
e Subjects were assigned to the protocol group opposite the protocol group to which they were randomized in the M1 appointment. The M1 protocol group is in parentheses.
Analysis of the gapping differences for the 4 protocol groups at the first MRI appointment (n = 28 for each protocol group, N = 112) showed that there was a significant difference (P = .001; KW, 16.3). Protocol 1 (SPP) had greater gapping than the other 3 protocol groups (mean gapping differences: protocol 1, 1.09 ± 1.22 mm; protocol 2, 0.24 ± 0.40; protocol 3, 0.19 ± 1.23; protocol 4, 0.18 ± 0.71).
Plotting the MRI gapping difference data revealed several outliers in protocols 1 and 3. Consequently, an adjusted analysis was made that eliminated 2 values from protocols 1 and 3 (the highest and lowest values of each protocol group). No values were eliminated from protocols 2 and 4. The variability of protocol 1 remained high, with 4 remaining protocol 1 values markedly higher than the others. Therefore, secondary analyses (ANOVA) were performed with and without these 4 values in protocol 1 (n = 26 and n = 22, respectively). Both analyses of the gapping differences for the 4 protocol groups at the first MRI appointment showed that there was a significant difference (P = .0005 and KW, 13.2; P = .009 and KW, 17.9, for n = 108 and 104 subjects, respectively), with protocol 1 (SPP) having greater gapping than the other 3 protocol groups (mean gapping differences for 104 subjects: protocol 1, 0.66 ± 0.48 mm; protocol 2, 0.24 ± 0.41; protocol 3, 0.23 ± 0.86; protocol 4, 0.18 ± 0.71).
Greatest Gapping Differences of Males vs Females
No differences were found between males and females for GGDPTS at either the M1 (P = .81) or M2 (P = .91) appointments.
Verbal Numeric Pain Rating Scale at M1
Differences in VNPRS before and after the intervention were as follows: protocol 1, 0.18 ± 1.19; protocol 2, −0.04 ± 1.0; protocol 3, 0.79 ± 1.4; protocol 4, 0.18 ± 0.72. These VNPRS differences were significant (P = .04) with protocol 3 (SMT followed by SPP) showing the greatest difference (decreased pain after the intervention) and protocol 2 (SMT followed by supine positioning) showing the least difference. The difference between these 2 protocol groups was significant (P < .05). There were no significant differences between any other protocol groups.
Two Weeks of Treatment
Subjects received 2 weeks of care between the M1 and M2 appointments. Based on each subject's clinical progress, the treating clinician determined the number of treatments and the modality (or modalities) of care given at each treatment. Excluding the MRI appointments, between 2 and 6 (mean, 4.3) treatments per subject were given during the 2 weeks of care, totaling 483 treatments. Only 3 of these 483 did not include a lumbar side-posture SMT (1 treatment for 3 different subjects who were be assigned to 3 different protocol groups at the M2 appointment [protocol groups 2, 3, and 4]). Table 5 provides detailed information of the other modalities used in the 2 weeks of treatments provided in the study. Only SMT was provided at the MRI appointments.
Table 5Treatment modalities used during 2 weeks of care between M1 and M2 appointments
Three subjects sought outside care during the course of the study; 2 subjects (1 from second [M2] MRI protocol group 3 and 1 from M2 MRI protocol group 4) received a massage during the 2 weeks of care. A third subject (from M2 MRI protocol group 3) made an appointment and was seen by an orthopedic surgeon for a consultation. This third subject was then referred for an MRI (outside the study) but received no additional treatment before the second MRI appointment and for at least the following 2 weeks (the duration of tracking in the study).
Second MRI Appointment
Changes in Subject Characteristics Between First and Second MRI Appointments
The right (M2) side of Table 4 shows VAS, BQ, and analgesic medication use at M2. The only notable difference among protocol groups was that the overall percentage of individuals using analgesic medication was reduced from 35% (n = 39) at M1 to 23% (n = 26) at M2. Protocol 1 had fewer subjects using analgesics than the other M2 protocol groups (14% vs 26%); however, this did not appear to influence the other variables.
Gapping Differences at M2
Data analysis for the second MRI appointment was initially performed on all 112 subjects. A significant difference (P = .005; KW, 12.9) existed, with protocol 3 (SMT) having greater gapping than the other 3 protocol groups (mean gapping differences: protocol 1, 0.65 ± 0.92 mm; protocol 2, 0.18 ± 0.51; protocol 3, 0.76 ± 0.85; protocol 4, 0.44 ± 0.46).
The values were plotted, and several outliers were identified as follows: protocol 1, no values; protocol 2, 2 values (highest and lowest); protocol 3, 2 values (lowest 2 values); protocol 4, 2 values (highest 2 values). The remaining values were well distributed following removal of the outliers. Consequently, the total number of subjects in the secondary analysis at the M2 appointment was 106. Differences remained significant (P = .0005, F = 6.5; protocol 1, 0.65 ± 0.92 mm; protocol 2, 0.17 ± 0.38; protocol 3, 0.89 ± 0.71; protocol 4, 0.35 ± 0.32). In addition, protocol 3 (SMT) was the only intervention that showed significantly more gapping than both control protocols (protocol 2: P < .001; 95% confidence interval [CI], −1.18 to −0.26; protocol 4: P < .05; 95% CI, 0.08-1.0). Protocol 1 (SPP) showed a significant difference with protocol 2 (P < .05; 95% CI, 0.02-0.93). Figure 6 shows the gapping differences for the 4 protocol groups, and Figure 7 shows examples of preintervention and postintervention scans for each protocol group.
Fig 6Gapping differences (in millimeters, Y-axis) between the A-P Z joint space measurements of the first and second MRI scans (value from the first scan was subtracted from the value of the second scan) at the M2 appointment. The GGDPTS are presented here. Protocol 3 (SMT protocol groups) showed more gapping (therapeutic gapping) than the other protocol groups, followed by protocol 1 (SPP protocol group).
Fig 7Preintervention and postintervention scans for each of the 4 study protocol groups. The box on each scan indicates the up-side Z joint during SSP or side-posture SMT. This was also the most painful side of LBP. L4 indicates the L4/L5 segmental level, and L5 indicates the L5/S1 segmental level. Notice the low signal line within the center of the protocol 3 “Post” R L4/L5 Z joint. This may be gas (most likely carbon dioxide) within the joint secondary to cavitation during SMT.
Removal of outliers did not change the outcomes for M2 VAS, BQ, or VNPRS; consequently, the data here represent analysis of all 112 subjects. Visual analog scale changed from 28.8 ± 18.16 mm (range, 1-82 mm) at the first MRI appointment to 16.4 ± 16.88 mm (range, 0-81 mm) at the second MRI appointment, a difference of 12.4 mm (P < .0001). Bournemouth questionnaire changed from 23.8 ± 11.56 (range, 3-54) at the first MRI appointment to 13.4 ± 10.69 (range, 0-48) at the second MRI appointment, a difference of 10.4 (P < .0001).
Verbal Numeric Pain Rating Scale at M2
Verbal numeric pain rating scale preintervention values were low (little pain reported) at M2, and differences in VNPRS before and after the intervention at M2 were not significant (P = .41).
No changes were made to the protocols, and no adverse events, harms, or unintended effects were reported during the course of the study.
Discussion
Approximately twice as many males (n = 75) were enrolled as females (n = 37). This representation is different than is usually found in chiropractic practices, which have a slightly higher percentage of female patients.
Characterization of health status and modifiable risk behavior among United States adults using chiropractic care as compared with general medical care.
The higher number of males in this study was a reflection of the numbers of males (n = 798) and females (n = 497) who responded to the recruiting advertisements (12 subjects declined before any data were recorded). Although recruitment for this study was based on methods used successfully in previous clinical trials,
future studies will include more advertisements that target females and run on media and programming with a higher female demographic. Because no difference was found in gapping between male and female subjects at either the M1 or M2 appointments, the higher percentage of males most likely did not affect the outcome of the study.
First MRI Appointment
Subjects were allowed to take analgesic medications, both over-the-counter and previously prescribed prescription medication, and medication use between appointments was documented at every visit. Although a much higher percentage of M1 protocol 1 subjects were taking analgesic medication, those protocol 1 subjects taking analgesic medications had only 0.16 mm more gapping than those not taking analgesic medications; the GGDPTS for patients not taking and taking analgesic medications was 1.00 and 1.16 mm, respectively. Removal of protocol group 1 subjects taking analgesic medication did not alter the results.
First MRI appointment protocol 1 (SPP only) subjects showed more gapping than the other protocol groups, including protocol 3 (side-posture SMT followed by SPP). Paraspinal muscles may have relaxed more during prolonged SPP in protocol 1, whereas SMT may have resulted in transient increased muscle tightness in protocol 3. This relationship reversed at the second MRI appointment (M2) with protocol 3 showing more gapping than protocol 1. The increased gapping of M2 protocol 3 subjects may indicate that the paraspinal muscles were more relaxed after 2 weeks of treatment (including SMT) at the M2 appointment allowing more Z joint gapping with SMT.
Protocol 3 (SMT followed by SPP) was the only protocol group to show significant improvement in pain (VNPRS), whereas those subjects receiving side-posture SMT and then placed on their backs (protocol 2) had almost no change in pain following the intervention. These results indicate that the lumbar side-posture position may have therapeutic benefit in acute LBP, increasing gapping in patients in acute pain (protocol 1) and enhancing pain reduction following SMT (protocol 3). This is consistent with the common recommendation by clinicians and the literature that lying on the side is of benefit for LBP patients.
The increased gapping that occurred with prolonged SPP (ie, SPP of approximately 12 minutes) could conceivably promote the break-up of intra-articular Z joint adhesions.
Stimulation of such Z joint mechanoreceptors has been hypothesized as a mechanism of reducing pain via a gating mechanism in the spinal cord. This hypothesis is supported by animal studies showing that pressure on the Z joint (including the capsule) decreases activity of spinal cord dorsal horn neurons responding to nociceptive stimulation.
The treatments in this study could be described as “structured pragmatic” in nature. Because the purpose of the study was to assess Z joint gapping at initial LBP presentation and after clinical improvement at 2 weeks, the purpose of the treatment was to provide usual chiropractic care that would result in the maximum improvement following 2 weeks of treatment. Spinal manipulative therapy was provided in almost all treatment appointments (only 3 of 483 treatments did not receive SMT). The study clinician was given the latitude to determine the number of treatments per week, within the parameters of 1 to 3 treatments, and choose from a “menu” of other modalities that could be used.
The goals of the treatments were successfully achieved. The results show that the subjects all received approximately the same number of treatments, SMT, and similar numbers and types of “other modalities.” In addition, after 2 weeks of care, the subjects showed approximately the same improvement as measured by pain and functional impairment. Therefore, the study was able to measure Z joint gapping in a homogenous cohort at initial presentation and after similar improvement of LBP following 2 weeks of care. Recall, this was not a study assessing effectiveness of care (although all subjects significantly improved following 2 weeks of care) but was designed to assess Z joint gapping in acute LBP patients.
Effects of side-posture positioning and side-posture adjusting on the lumbar zygapophysial joints as evaluated by magnetic resonance imaging: a before and after study with randomization.
the Z joints resumed their normal spacing once they received the load of the supine position, explaining why protocol 4 (brief SPP followed by neutral/supine position [SPP control group]) and protocol 2 (SMT followed by neutral/supine position [SMT control group]) showed little gapping.
Spinal manipulative therapy followed by SPP (protocol 3) resulted in the greatest amount of Z joint gapping in LBP subjects at the M2 appointment, followed by SPP (protocol 1). These data indicate that SMT produced more gapping after subjects received 2 weeks of treatment. One would anticipate even greater differences between protocol 1 (SPP) and protocol 3 (SMT) mean gapping differences if the subjects had been assessed with a third MRI appointment after 4 weeks of care. The subjects would then have been more similar in pain and function to the healthy subjects assessed in previous studies, where protocol 3 showed 0.7 mm more gapping than protocol 1.
Effects of side-posture positioning and side-posture adjusting on the lumbar zygapophysial joints as evaluated by magnetic resonance imaging: a before and after study with randomization.
The increased gapping after 2 weeks of care in this study could have been due to reduction of intra-articular adhesions (and potentially other connective tissue adhesions, including adhesions within the fascia) and reduced muscle tension of the paraspinal muscles surrounding the Z joints. Future studies should assess changes in muscle activity, as measured by electromyography, at the M1 and M2 appointments.
The gapping changes were also accompanied with an overall reduction of pain and improved functional impairment. The protocol 3 (SMT followed by SPP) findings also indicate that keeping a person in the side-posture position for several minutes following SMT may have therapeutic benefit. Future work in animals and humans should further assess the unique effects of SPP alone and SMT followed by SPP.
This study provides additional evidence that normal Z joints (ie, Z joints within normal anatomical limits; recall that subjects with anomalous Z joints were excluded from this study) gap with SMT and SPP, which is different from conclusions of previous authors
who believed that Z joints that were within normal anatomical limits do not gap. These previous authors strongly indicated that chiropractors were misinforming their patients when describing Z joint gapping as a mechanism of SMT. The study conducted by the other investigators placed the cadaveric spines in a more extended posture, which significantly reduces Z joint rotation, and, consequently, Z joint gapping. The standard SMT of this study is administered with the Z joints in a flexed position, which allows for rotation
Effects of side-posture positioning and side-posture adjusting on the lumbar zygapophysial joints as evaluated by magnetic resonance imaging: a before and after study with randomization.
indicate that typical Z joints do gap with SMT and SPP. The evidence that Z joints do gap can lead to a different approach to patient care than an assumption that they do not gap. When combined with other studies showing that adhesions develop in hypomobile Z joints
the results of this study further buttress the theory provided in Figure 1. This theory begins with the a priori assumption that Z joints become hypomobile for a variety of reasons (eg, sedentary lifestyle, injury, repetitive asymmetrical tasks at work) and that hypomobile Z joints develop adhesions, which further reduces motion; SMT gaps the Z joint surfaces, thus breaking up Z joint adhesions and reestablishing spinal motion.
Limitations
This study was conducted on the lumbar spine. Additional research assessing the cervical and thoracic regions is needed to determine the effects of positioning and gapping of the Z joints in these regions of the vertebral column.
As discussed previously, future studies should target female subjects for recruitment to obtain a more equal distribution of male and female subjects.
The ideal design would have been 4 MRI appointments: 1 before commencement of treatment (M1 in this study), 1 after 1 week of treatment, 1 after 2 weeks of treatment (M2 in this study), and 1 after 4 weeks of treatment. This would not only have allowed for assessment of gapping at additional time points in the LBP continuum but would also have allowed each subject to be in each of the 4 protocol groups. However, 4 MRI appointments could not be justified from cost and patient burden standpoints. In addition, future research assessing gapping differences in subjects with chronic LBP should be performed. Future work could also include a medication only control treatment protocol group.
Conclusion
In this study of acute LBP subjects, SPP subjects (protocol 1) showed the greatest Z joint gapping at the baseline MRI appointment. After 2 weeks of standard chiropractic treatment, SMT followed by SPP (protocol 3) resulted in the greatest amount of Z joint gapping, followed by SPP alone (protocol 1); these results are consistent with those of previous studies on healthy subjects.
Effects of side-posture positioning and side-posture adjusting on the lumbar zygapophysial joints as evaluated by magnetic resonance imaging: a before and after study with randomization.
The side-posture position appeared to have additive therapeutic benefit to SMT, with acute LBP subjects receiving SMT and remaining in side-posture experiencing the greatest reduction of pain, independent of Z joint gapping, at the first appointment and the greatest amount of Z joint gapping after 2 weeks of care.
Funding Sources and Potential Conflicts of Interest
This project was made possible by grant number R01-AT000123 from the National Center for Complementary and Alternative Medicine. Its contents are solely the responsibility of the authors and do not necessarily represent the official views of the National Center for Complementary and Alternative Medicine or the National Institutes of Health. No conflict of interest was reported by any of the authors.
Practical Applications
•
Zygapophyseal joint gapping is hypothesized to be related to a therapeutic benefit of SMT (“therapeutic gapping”).
•
Previous studies of healthy subjects found that Z joints receiving SMT gapped more than those receiving SPP alone.
•
In this study of acute LBP subjects, SPP showed the greatest Z joint gapping at the baseline MRI appointment.
•
After 2 weeks of standard chiropractic treatment, SMT followed by SPP resulted in the greatest amount of Z joint gapping, followed by SPP alone.
•
The side-posture position appeared to have additive benefit to SMT regarding pain reduction and Z joint gapping.
Acknowledgment
We gratefully acknowledge the technical support of Joshua Healy, DC, and the observers who made the morphometric measurements: Frank Balester, MSOM, LAc; Tyra Horner, DC; and Derek Simpson, DC, ND.
References
Janse J
Hildebrandt RW Principles and practice of chiropractic: an anthology. Kjellberg & Sons,
Wheaton1976: 326
Effects of side-posture positioning and side-posture adjusting on the lumbar zygapophysial joints as evaluated by magnetic resonance imaging: a before and after study with randomization.
A practice-based study of patients with acute and chronic low back pain attending primary care and chiropractic physicians: two-week to 48-month follow-up.
Quebec Task Force on Spinal Disorders. Scientific approach to the assessment and management of activity-related spinal disorders: a monograph for clinicians.
Reliability of zygapophysial joint space measurements made from magnetic resonance imaging scans of acute low back pain subjects: comparison of 2 statistical methods.
Chiropractic chronic low back pain sufferers and self-report assessment methods. Part I. A reliability study of the Visual Analogue Scale, the Pain Drawing, and the McGill Pain Questionnaire.
The effects of side-posture adjusting on the lumbar zygapophysial joints of low back pain patients as evaluated by magnetic resonance imaging: a preliminary study.
Characterization of health status and modifiable risk behavior among United States adults using chiropractic care as compared with general medical care.
This project was made possible by grant number 2R01-AT000123 from the National Center for Complementary and Alternative Medicine. Its contents are solely the responsibility of the authors and do not necessarily represent the official views of the National Center for Complementary and Alternative Medicine or the National Institutes of Health.
00055-9&id=gr1.jpg){kind=link}
00055-9&id=gr2.jpg){kind=link}
00055-9&id=gr3.jpg){kind=link}
00055-9&id=gr4.jpg){kind=link}
00055-9&id=gr5.jpg){kind=link}
00055-9&id=gr6.jpg){kind=link}
00055-9&id=gr7.jpg){kind=link}