journal of orthopaedic & sports physical therapy | volume 40 | number 3 | march 2010 | 133

[ research report ]

1Associate Professor, University of Puget Sound, School of Physical Therapy, Tacoma, WA. 2Assistant Professor, US Army-Baylor University Doctoral Program in Physical Therapy, Fort Sam Houston, TX. 3Physical Therapy Element Chief, United States Air Force, Sheppard Air Force Base, TX. 4Director of Rehabilitation Department, Northern Navajo Medical Center, Shiprock, NM. 5Associate Professor, Department of Physical Therapy, Texas State University, San Marcos, TX. This study was approved by The Institutional Review Boards of Brooke Army Medical Center and Wilford Hall Medical Center. The opinions or assertions contained herein are the private views of the authors and are not to be construed as official or as reflecting the views of the Departments of the Army, Air Force, or Defense. Address correspondence to Dr Robert E. Boyles, Associate Professor, University of Puget Sound, School of Physical Therapy, 1500 N Warner St, #1070, Tacoma, WA 98416. E-mail: bboyles@pugetsound.edu

RobeRt e. boyles, PT, DSc1 • Michael J. Walker, PT, DSc2 • Brian a. Young, PT, DSc3 Joseph B. strunce, PT, DSc4 • roBert s. Wainner, PT, PhD5

The Guide to Physical Therapist Prac-tice1 outlines several physical therapy interventions suitable for the manage-ment of patients with mechanical neck pain. These interventions include man-ual physical therapy, exercise, traction, physical agents, and mechanical and electrotherapeutic modalities.1 Despite their common use, existing research has produced insufficient evidence regarding the effectiveness of these interventions and the clinical decision-making strate-gies to guide their use.15-18 Current best evidence supports the multimodal use of manual physical therapy (MPT), includ-

Mechanical neck pain is defined as pain that can be provoked by neck movements or

provocative tests.2 Neck pain is a common musculoskeletal complaint, with a reported lifetime prevalence of 22% to 67%10 and a point prevalence of 13% to 22%.34 Up to 41% of patients with neck pain seek care from a general practitioner and 33% from a physical therapist.31

The Addition of Cervical Thrust Manipulations to a Manual Physical Therapy Approach in Patients Treated for Mechanical

Neck Pain: A Secondary Analysis

t stuDY Design: Secondary analysis of a randomized clinical trial (RCT).

t oBJectiVes: To perform a secondary analysis on the treatment arm of a larger RCT to determine differences in treatment outcomes, adverse reac-tions, and effect sizes between patients who re-ceived cervical thrust manipulation and those who received only nonthrust manipulation as part of an impairment-based, multimodal treatment program of manual physical therapy (MPT) and exercise for patients with mechanical neck pain.

t BackgrounD: A treatment regimen of MPT and exercise has been effective in patients with mechanical neck pain. Limited research has com-pared the effectiveness of cervical thrust manipu-lations and nonthrust mobilizations for this patient population, and no studies have investigated the added benefit of cervical thrust manipulations as part of an overall MPT treatment plan.

t MethoDs: Treatment outcomes from 47 patients in the treatment arm of a larger RCT, with a primary complaint of mechanical neck pain, were analyzed. Twenty-three patients (49%) received cervical thrust manipulations as part of their MPT treatment, and 24 patients (51%) received only cer-vical nonthrust mobilizations. All patients received up to 6 clinic sessions, twice weekly for 3 weeks, and a home exercise program. Primary outcome measures were the Neck Disability Index (NDI), 2

visual analog scales for cervical and upper extremity pain, and a 15-point global rating of change scale. Blinded outcome measurements were collected at baseline and at 3-, 6- and 52-week follow-ups.

t results: Consistent with the larger RCT, both subgroups in this secondary analysis demonstrat-ed improvement in short- and long-term pain and disability scores. Low statistical power (β.28) and the resultant small effect size indices (–0.21 to 0.17) preclude the identification of any between-group differences. No serious adverse reactions were reported by patients in either subgroup.

t conclusions: Clinically meaningful and statistically significant improvements in both sub-groups of patients over time suggest that cervical thrust manipulation, as part of the MPT treatment plan, did not influence the results of the treatment arm of the larger RCT from which this study was drawn. Although no between-group differences can be identified, the small observed effect sizes in this study may benefit future studies with sample size estimation for larger RCTs and indicate the need to incorporate clinical prediction rule criteria as a means to improve statistical power.

t leVel oF eViDence: Therapy, level 4. J Orthop Sports Phys Ther 2010;40(3):133-140. doi:10.2519/jospt.2010.3106

t keY WorDs: cervical spine, manual therapy, mobilization

ing cervical thrust and/or non-thrust manipulation, and exercise for patients with cervicogenic

headache and mechanical neck pain.5,17,18 Hoving et al21,22 and Korthals-de Bos et al28 reported

SUPPLEMENTAL VIDEO ONLINE

40-03 Boyles.indd 133 2/17/10 2:49:30 PM

134 | march 2010 | volume 40 | number 3 | journal of orthopaedic & sports physical therapy

[ research report ]

examination of the cervical spine and upper quarter prior to randomization, where demographic information, self-report measures, and physical exam measurements (ie, cervical range of mo-tion, passive accessory mobility and pain provocation testing, cervical special test-ing, etc) were collected. The self-report scores from the 47 patients (31% female; mean SD age, 48.8 14.1 years) as-signed to the MPT and exercise group of the RCT were analyzed for the purposes of this secondary analysis.

interventionsPatients within the treatment arm re-ceived MPT and exercise directed at im-pairments of the cervical spine, thoracic spine, and adjacent ribs, as identified by the treating physical therapist. The spe-cific manual interventions applied (which included the use of thrust and nonthrust techniques) and the intervention param-eters used (grade, duration, and repeti-tions) were solely based on the clinical reasoning and decision making of the treating therapists. Although numerous cervical manipulative techniques exist, 2 techniques were predominantly used. Spinal stiffness in flexion, described as “opening restrictions,” was treated in flexion with a translatory thrust gapping manipulation applied to the contralater-al side of the restricted motion segment (Figure 1 , online ViDeo). Spinal stiffness in extension, described as “closing restric-tions,” was treated in extension with a caudally and slightly medially directed

an impairment-based, multimodal treat-ment program of MPT and exercise for patients with mechanical neck pain.

MethoDs

subjects

ninety-four patients referred to 3 outpatient military treatment facilities with a primary complaint

of neck pain were included in the initial multicenter RCT.38 Inclusion criteria for this RCT were the following: a primary complaint of neck pain, with or without unilateral upper extremity symptoms; age greater than 18 years; a Neck Disabil-ity Index (NDI) score equal to or greater than 10 points; and a composite visual analog scale (VAS) pain score equal to or greater than 30 mm, as derived from 3 separate 100-mm pain scales measuring the patient’s cervical, upper extremity, and average 24-hour pain scores. Patients were excluded if they had a whiplash in-jury within the past 6 weeks, a history of spinal tumors, spinal infection, cervical spine fracture, or previous neck surgery, a pending legal action regarding their neck pain, a diagnosis of central cervical spinal stenosis, bilateral upper extremity symp-toms, or 2 positive neurological findings at the same nerve root level.

Eligible patients were randomly as-signed to 2 treatment groups: MPT and exercise (n = 47) or minimal interven-tion (n = 47). Each patient underwent a standardized history and physical

superior patient perceived recovery and cost effectiveness when using cervical nonthrust manipulations as compared to general practitioner care or physical therapy that did not include nonthrust manipulations for patients with neck pain greater than 2 weeks in duration. Cervical thrust manipulation and exercise resulted in better patient outcomes and satisfaction levels when compared to the use of manipulation or exercise alone.3,12 In a recent randomized controlled trial, Walker et al38 compared the effectiveness of MPT and exercise to a minimal in-tervention approach in 94 patients with acute, subacute, or chronic mechanical neck pain with or without upper extrem-ity symptoms. Unlike previous studies, the MPT techniques included both thrust and nonthrust manipulations applied to the cervical spine, thoracic spine, and ad-jacent ribs. Walker et al33 demonstrated significant short-term (3- and 6-week) and long-term (1-year) improvements in pain, disability, and patient perceived recovery following up to 6 treatment ses-sions consisting of a multimodal use of MPT and exercise.

Three studies4,23,37 have directly com-pared the effectiveness of cervical thrust manipulation versus nonthrust manipu-lation for the treatment of patients with acute, subacute, and chronic neck pain. No differences in short- and long-term pain relief and disability were reported in these trials when comparing these MPT interventions.18 Despite these similar re-sults, Hurwitz et al24 reported that adverse reactions were more common in patients following cervical thrust manipulation, and that these adverse reactions negatively affected patient satisfaction, perceived im-provement, and pain and disability scores at subsequent follow-up visits.

The purpose of this study was to per-form a secondary analysis on the treat-ment arm of a larger RCT38 to determine differences in treatment outcomes, ad-verse reactions, and effect sizes between patients who received cervical thrust manipulation and those who received only nonthrust manipulation, as part of

Figure 1. Cervical thrust gapping manipulation in flexion for left C4-5 opening restriction (online ViDeo).

Figure 2. Cervical thrust manipulation in extension for right C4-5 closing restriction (online ViDeo).

40-03 Boyles.indd 134 2/17/10 2:49:31 PM

journal of orthopaedic & sports physical therapy | volume 40 | number 3 | march 2010 | 135

thrust to the ipsilateral dysfunctional seg-ment (Figure 2, online ViDeo). Additionally, and to a lesser extent, a nonphysiologic, or upslope, thrust manipulation tech-nique was used at the treating therapist’s discretion (Figure 3, online ViDeo).

In addition to MPT, all patients re-ceived 3 basic clinic and home exercises: cervical rotation range of motion, cervi-cal retraction (chin tucks), and deep neck

flexor strengthening. The therapist could add additional exercises as needed to tar-get specific impairments or reinforce the manual interventions. Patients received up to 6 sessions, twice weekly for 3 weeks. Treatment sessions were limited to 1 hour for the initial examination and treatment and to 30 minutes for subsequent follow-up sessions. The 8 physical therapists providing treatment were either faculty or fellows in the US Army-Baylor University Post-Professional Doctoral Program for Orthopaedic Manual Physical Therapy.

For this secondary analysis, patient outcome data from these 47 patients were separated into 2 groups based on the type of cervical MPT provided: (1) patients who received cervical thrust manipula-tion as part of their treatment (thrust group, n = 23 [49%]) and (2) patients who received only cervical nonthrust manipulations (nonthrust group, n = 24 [51%]). Patients within this treatment arm were not randomized with respect

to the inclusion or exclusion of cervical thrust manipulations. All patients receiv-ing cervical thrust manipulation also re-ceived cervical nonthrust manipulations during their treatment duration.

Baseline comparisons (taBle 1) of clin-ical and demographic characteristics re-vealed statistically significant differences between the 2 groups in age (nonthrust group being older [P =.02]) and base-line NDI scores. The difference in NDI scores was statistically significant but not clinically meaningful (thrust group hav-ing a higher NDI score [P =.05]). The difference in symptom duration (mean difference, 1166 days), while clinically significant, was not statistically signifi-cant. Although the distribution of pa-tients with acute, subacute, and chronic neck pain was similar between groups, the nonthrust group had 6 patients with symptoms greater than 5 years in du-ration as compared to 1 patient in the thrust group.

taBle 1 Baseline Demographic and Clinical Variables*

Abbreviations: NDI, neck disability index; ROM, range of motion; UE, upper extremity; VAS, visual analog scale.* Values expressed as mean SD unless otherwise noted.† Indicates significant (P.05) difference between groups using independent t tests.

Demographic and clinical Variables all patients (n = 47) thrust (n = 23) nonthrust only (n = 24) significance (P Value)

Age (y) 48.8 14.1 44.0 11.1 53.5 15.3 .02†

Gender, female (n) 31 15 16 .92

Symptom duration (d) 1082.4 2256.7 487.0 447.2 1653.0 3049.5 .08

Acute, 30 d (n) 7 3 4 .73

Subacute, 30-90 d (n) 9 5 4 .66

Chronic, 90 d (n) 31 15 16 .92

Imaging performed (n) 31 13 18 .18

Headaches (n) 27 14 13 .64

UE symptoms (n) 31 13 18 .18

Cumulative ROM (deg) 255.3 56.8 263.6 59.2 247.3 54.5 .35

Flexion 44.9 14.1 46.3 13.9 43.6 14.5 .51

Extension 42.7 14.3 45.1 16.7 40.3 11.5 .26

Right rotation 53.0 14.2 55.8 12.7 50.4 15.3 .20

Left rotation 48.8 14.8 50.6 15.6 47.1 14.1 .42

Right sidebending 32.7 12.8 32.8 13.3 32.6 12.5 .95

Left sidebending 33.1 12.5 33.0 13.6 33.3 11.7 .93

Initial NDI (0-50 points) 15.6 4.5 16.9 5.2 14.4 3.3 .05†

Cervical VAS (0-100 mm) 53.5 20.7 57.4 23.1 49.7 17.7 .21

UE VAS (0-100 mm) 25.3 24.9 23.4 27.1 27.2 23.0 .61

Figure 3. Cervical thrust manipulation using nonphysiologic (upslope) techniques to right C4-5 (online ViDeo).

40-03 Boyles.indd 135 2/17/10 2:49:32 PM

136 | march 2010 | volume 40 | number 3 | journal of orthopaedic & sports physical therapy

[ research report ]

outcome MeasuresPrimary outcome measures consisted of the NDI, a VAS for cervical pain, a VAS for upper extremity pain, and the patient-perceived global rating of change (GRC). The NDI and VAS measures were com-pleted at baseline, upon treatment com-pletion at 3 weeks, and at 6-week and 1-year follow-ups. The GRC was com-pleted at the 3 follow-up periods. Recent clinical trials22,28 and a systematic review18 have used these 3 common outcome mea-sures to assess pain intensity, disability, and perceived recovery in patients with mechanical neck pain.

The NDI was selected to assess the pa-tient’s self-reported disability due to me-chanical neck pain.29,36 The NDI has been found to have high test-retest reliability, internal consistency,32,3629 and good con-current validity with the McGill Pain Questionnaire and patient-perceived im-provement.36 Stratford et al32 analyzed the NDI in relation to patient decision mak-ing and found both the minimal clinically important difference (MCID), the small-est change in a scale that is meaningful to patients, and the minimal detectable

change (MDC), the smallest amount of change that represents a change beyond measurement error, to be 5 raw points or 10 percentage points. Cleland et al8 compared the psychometric properties of the NDI and the Numeric Pain Rating Scale (NPRS) in patients with mechani-cal neck pain and found them both to be responsive and to display fair to moder-ate test-retest reliability. They also found the NDI to have a higher MDC score than previously reported (19%). Young et al40 reported similar results to Cleland et al,8 with the MDC found to be 10 raw score points on the NDI.

The 100-mm VAS, where 0 represent-ed “no pain” and 100 represented “worst pain imaginable,” was used to assess cer-vical and upper extremity pain intensi-ties. The VAS has reported test-retest reliability between 0.95 to 0.9730 and an MCID of 12 mm (3 mm at a 95% CI),27 regardless of the severity of pain initially reported.

The patient-perceived GRC25,26 was used to assess the patient’s perception of change in their condition. The GRC is a 15-point scale ranging from –7 to +7,

where 0 represents no change, –7 indi-cates that the patient is “a very great deal worse,” and +7 indicates that the patient is “a very great deal better.” Juniper et al26 proposed the following classifications based on a patient’s GRC score: 0, 1, or –1 had no change; 2 to 3 had minimal change; 4 to 5 had moderate change; and 6 to 7 had a large change in their condition.

Data analysisBased on the manual interventions used, patients were divided into 2 groups for data analysis. Twenty-three patients received cervical high-velocity thrust (thrust group) as part of their treatment, and 24 patients received only nonthrust techniques (nonthrust group). NDI and VAS variables were analyzed using a 2-by-4 mixed-model multivariate analysis of covariance (MANCOVA) and univariate analyses of covariance (ANCOVA), with covariates of age and duration of symp-toms (α = .05). A separate 2-by-3 mixed-model MANCOVA and ANCOVA were performed to include baseline NDI scores as an additional covariate in the model.

taBle 2Adjusted Mean Scores, Mean Differences, and

Effect Size Indices for Primary Outcome Measures*

Abbreviations: CI, confidence interval; NDI, neck disability index; UE, upper extremity; VAS, visual analog scale.* Values are means (95% CIs) unless otherwise noted. Adjusted values based on covariates of age, symptom duration, and baseline NDI scores.† Effect sizes of 0.2, 0.5, and 0.8 correspond to small, medium, and large differences, respectively. Positive mean differences and effect sizes denote an advantage towards the thrust group.

Measure/Follow-up thrust Mean nonthrust-only Mean Mean Difference (95% ci) effect size† (95% ci)

NDI (0-50)

Baseline 17.1 (15.2 to 19.0) 14.1 (12.3 to 16.0) 3.0 (0.3 to 5.6) 0.66 (0.06 to 1.23)

3 wk 6.2 (4.0 to 8.5) 6.4 (4.2 to 8.6) –0.2 (–3.3 to 3.0) –0.04 (–0.61 to 0.54)

6 wk 6.3 (4.3 to 8.4) 5.5 (3.5 to 7.5) 0.8 (–2.1 to 3.7) 0.17 (–0.41 to 0.74)

1 y 5.2 (2.8 to 7.6) 6.4 (4.1 to 8.8) –1.2 (–4.6 to 2.2) –0.21 (–0.78 to 0.37)

Cervical VAS (0-100 mm)

Baseline 56.5 (47.3 to 65.6) 50.6 (41.7 to 59.5) 5.9 (–6.9 to 18.6) 0.27 (–0.31 to 0.84)

3 wk 14.7 (6.4 to 22.9) 13.7 (5.7 to 21.8) 0.9 (–10.6 to 12.5) 0.05 (–0.52 to 0.62)

6 wk 16.3 (8.6 to 24.1) 15.2 (7.6 to 22.8) 1.1 (–9.7 to 12.0) 0.06 (–0.51 to 0.63)

1 y 17.2 (7.9 to 26.6) 19.8 (10.6 to 28.9) –2.5 (–15.6 to 10.5) –0.11 (–0.68 to 0.46)

UE VAS (0-100 mm)

Baseline 21.5 (10.5 to 32.5) 29.0 (18.2 to 39.8) –7.5 (–22.9 to 7.9) –0.29 (–0.86 to 0.29)

3 wk 6.9 (0.9 to 12.8) 7.6 (1.8 to 13.4) –0.7 (–9.0 to 7.6) –0.05 (–0.62 to 0.52)

6 wk 8.5 (1.7 to 15.4) 6.7 (0.1 to 13.6) 1.7 (–7.9 to 11.3) 0.10 (–0.47 to 0.67)

1 y 8.6 (–0.4 to 17.6) 11.5 (2.7 to 20.3) –2.9 (–15.5 to 9.7) –0.14 (–0.71 to 0.44)

40-03 Boyles.indd 136 2/17/10 2:49:32 PM

journal of orthopaedic & sports physical therapy | volume 40 | number 3 | march 2010 | 137

Dichotomized GRC scores were used to classify patients as a success or nonsuc-cess at each follow-up interval. Patients that rated their improvement at or above 6 (“a great deal better”) were considered a success, while patients that rated their change at 5 (“quite a bit better”) or below were a nonsuccess. These success rates were compared using the chi-square sta-tistic (α = .05).

results

there were no significant in-teractions for either the multivari-ate or univariate analyses, but there

was a significant main effect with respect to time for all variables in both analyses (P.05). None of the covariate variables was significant (α = .05), indicating that these variables did not significantly add to the model and resulted in similar ad-justed and unadjusted means (differences of 0.5 points for NDI means and 2.5 mm for VAS scores). Adjusted mean dif-ferences are listed in taBle 2, along with the effect size index for all between-group comparisons, which was small. Observed power was low for both the MANCOVA (β.28) and ANCOVA (β.44) proce-dures. Figure 4 depicts the improvement in cervical pain over time for both inter-vention groups and the nonsignificant interaction effect between groups. This graph is representative of the changes observed in the NDI and upper extrem-ity VAS pain scores.

There were no significant differences between the 2 groups for treatment suc-cess rates based on patient-perceived GRC scores (P.28) (Figure 5).

Patients reported no adverse treat-ment effects, regardless of which manual physical therapy technique was used.

Discussion

this study is a secondary analy-sis of the treatment arm of a larger RCT,38 in which patients with me-

chanical neck pain achieved significant short- and long-term improvements in

pain and disability following treatment with MPT and exercise. Consistent with these findings, both the thrust and non-thrust subgroups in this analysis dem-onstrated similar improvements over time in short- and long-term clinical outcomes, with no statistical or clini-cally significant differences between the groups (taBle 2). At first glance, a similar outcome between the 2 groups might in-dicate that there was a lack of influence by a subset of patients for whom thrust manipulation was indicated, or that, if such a subgroup existed, the influence was minimal. However, an equally plau-sible explanation is that the nonsignifi-cant results were due to a relatively small, heterogeneous sample.

The mean differences between groups at each follow-up interval (taBle 2) are smaller than the established MCID for each outcome measure (12 mm for the VAS and 5 points for the NDI). Addi-tionally, the observed effect size indices were small (–0.21 to 0.17), with large 95% confidence intervals that include null values for each outcome measure and time interval. Effect size indices less than 0.2 typically indicate small clinical differences that are neither clinically nor statistically significant.

Caution, however, must be exercised when interpreting these data based on study limitations. This is an underpow-ered secondary analysis (β.28 and β.44) that prohibits any definitive state-

57.4

14.0 13.7 14.2

49.7

14.417.8

22.6

0.0

10.0

20.0

30.0

40.0

50.0

60.0

70.0

Cerv

ical

VAS

Pai

n Sc

ores

(mm

)

Thrust Nonthrust only

Baseline 3 wk 6 wk 1 yr

Figure 4. Cervical visual analog scale (VAS) pain scores. Values expressed in millimeters (scale, 0-100 mm). Nonsignificant interaction effect with analysis of covariance ANCOVA (P = .73). Significant main effect for time (P.05) demonstrates improvement in cervical pain for both groups.

0

10

20

30

40

50

60

70

80

Trea

tmen

t Suc

cess

(%)

Thrust Nonthrust only

3 wk 6 wk 1 yr

Figure 5. Global rating of change treatment success rates (where GRC 6). *Values expressed as a percentage. †No observed differences with chi-square analysis (P.28) for all time intervals.

40-03 Boyles.indd 137 2/17/10 2:49:34 PM

138 | march 2010 | volume 40 | number 3 | journal of orthopaedic & sports physical therapy

[ research report ]ment regarding the presence or absence of a treatment advantage of one approach over the other. Given the results of previ-ous studies demonstrating the effective-ness of both thrust3,12 and nonthrust21,22 procedures, a small effect size is not to be unexpected when comparing differences between cervical thrust and nonthrust manipulation as part of a larger treatment plan. If this is indeed the case, a larger and possibly more homogenous sample would be necessary to determine the clinical and statistical relevance in observed dif-ferences between these interventions. For example, Hurwitz et al23 compared thrust versus nonthrust manipulation in a large RCT involving 336 patients with neck pain. Even with this larger sample size, they were unable to detect clinical and statistically meaningful differences be-tween the 2 interventions. Their findings are consistent with our own, including similar mean differences and confidence intervals for pain and disability. Taken collectively, these trials underscore the need for future studies to incorporate criteria associating thrust manipulation with successful outcomes6,34 rather than simply increasing sample size. Due to the rare but potentially serious complications that have been reported following cervical thrust manipulation, as well as perceived harm,19,20 a stronger case might be made for the use of nonthrust procedures if they are shown to be equally effective, even though nonthrust procedures are still not without associated complications.11

In a recent RCT, Gonzalez-Iglesias et al14 reported that patients with neck pain who received thoracic spine thrust manipulations had significantly great-er improvements in pain, motion, and disability than a control group up to 4 weeks following treatment. Cleland et al9 reported that thrust manipulation of the thoracic spine was significantly more ef-fective than nonthrust manipulation for reducing pain and disability in patients with mechanical neck pain. Although similar to our study in sample size (n = 60) and patient presentation (mean age, 43.3 years; 55% female), Cleland et al7

had sufficient power and effect sizes to detect these between-group differences. High-velocity, low-amplitude thrust techniques appear to be more effective than low-velocity, variable-amplitude nonthrust manipulation techniques in overcoming the relative stiffness/im-mobility found in the thoracic spine. In contrast, cervical thrust and nonthrust manipulation techniques appear to have similar treatment effects when applied as part of an ongoing treatment program to the smaller, more mobile facet joints within the cervical spine.23 However, when used as a single-session interven-tion, Vernon et al35 concluded in a recent systematic review that thrust manipula-tion, as compared to nonthrust manual therapy, demonstrates superior changes in 100-mm VAS change scores and larger effect sizes in patients with nonradicular chronic neck pain.

Clinical decision making regarding the use of thrust manipulation is of in-terest. Another significant limitation of this study is the inherent selection bias in this nonrandomized post hoc analysis. Although subjects were randomly as-signed to the MPT and exercise group in the larger RCT,38 the inclusion of cervical thrust manipulation was at the discretion of the treating physical therapist. Physi-cal therapists in our study chose to ma-nipulate patients who were younger (P = .02) and had more reported disability (P = .05) and less symptom chronicity (P = .08) than patients receiving only non-thrust manipulations. Although adjusting for these covariates (age, duration, and baseline NDI scores) did not add to our statistical model, observed differences in these variables suggest that they were considered as part of the physical thera-pist’s clinical decision-making process for when to administer cervical thrust manipulation. It may be that physical therapists in this study performed cervi-cal thrust techniques on patients whom they perceived to be more responsive or presented with less relative risk (“safer”) towards this intervention.

Clinical prediction rules (CPR) have

been developed to assist in the clinical decision making associated with the di-agnosis and treatment of several muscu-loskeletal conditions. A CPR consisting of clinical examination items has been reported to identify patients with LBP who are more likely to respond to lum-bar manipulation.6,13 Similarly, Cleland and colleagues7 have reported a develop-mental CPR for patients with neck pain who benefit from thrust manipulation to the thoracic spine. Tseng and colleagues33 reported on predicted responders to cer-vical manipulation, but this has yet to be validated. Identifying a relevant sub-group of patients with neck pain who re-spond to cervical thrust manipulation, if one exists, would not only be useful for clinical decision making but would also help adequately plan future clinical trials. Future studies that compare the relative effectiveness of cervical thrust manipula-tion to nonthrust manipulation to treat patients with neck pain must consider the lack of significance and small effect sizes reported to date.23 Incorporating ex-isting7,39 and future developmental CPR criteria that identify a subset of patients who would benefit from cervical thrust manipulation would not only help deter-mine the validity of developmental CPRs but may help determine if there is a clini-cally meaningful difference between the 2 interventions for a subgroup of patients.

conclusion

this secondary analysis of 2 subgroups of patients with me-chanical neck pain demonstrated

short- and long-term improvements in pain and disability following a treatment regimen of MPT and exercise. The inclu-sion or exclusion of cervical thrust ma-nipulation into the MPT treatment plan did not influence the results of the treat-ment arm of the larger RCT from which this study was drawn. While the methods used in this study limits our conclusions, our small observed effect sizes may ben-efit future researchers with sample size estimation for larger RCTs and suggest

40-03 Boyles.indd 138 2/17/10 2:49:35 PM

journal of orthopaedic & sports physical therapy | volume 40 | number 3 | march 2010 | 139

the need to incorporate developmental CPR criteria as a means to improve sta-tistical power. t

keY pointsFinDings: Patients who received thrust manipulation to the cervical spine had similar short- and long-term results in all outcome scores of NDI, VAS, and GRC compared to those patients who received cervical nonthrust manipula-tions. Neither group reported adverse effects from either thrust or nonthrust cervical manipulations.iMplications: Both approaches appear to be equally effective and safe in this pa-tient population.caution: This is a secondary analysis performed on a relatively small number of patients from a larger RCT.

reFerences

1. Guide to Physical Therapist Practice. Second Edition. American Physical Therapy Association. Phys Ther. 2001;81:9-746.

2. Bogduk N. Neck pain. Aust Fam Physician. 1984;13:26-30.

3. Bronfort G, Evans R, Nelson B, Aker PD, Gold-smith CH, Vernon H. A randomized clinical trial of exercise and spinal manipulation for patients with chronic neck pain. Spine (Phila Pa 1976). 2001;26:788-797; discussion 798-789.

4. Cassidy JD, Lopes AA, Yong-Hing K. The imme-diate effect of manipulation versus mobilization on pain and range of motion in the cervical spine: a randomized controlled trial. J Manipula-tive Physiol Ther. 1992;15:570-575.

5. Childs JD, Cleland JA, Elliott JM, et al. Neck pain: Clinical practice guidelines linked to the International Classification of Functioning, Dis-ability, and Health from the Orthopedic Section of the American Physical Therapy Association. J Orthop Sports Phys Ther. 2008;38:A1-A34.

6. Childs JD, Fritz JM, Flynn TW, et al. A clinical prediction rule to identify patients with low back pain most likely to benefit from spinal ma-nipulation: a validation study. Ann Intern Med. 2004;141:920-928.

7. Cleland JA, Childs JD, Fritz JM, Whitman JM, Eberhart SL. Development of a clinical predic-tion rule for guiding treatment of a subgroup of patients with neck pain: use of thoracic spine manipulation, exercise, and patient educa-tion. Phys Ther. 2007;87:9-23. http://dx.doi.org/10.2522/ptj.20060155

8. Cleland JA, Childs JD, Whitman JM. Psycho-metric properties of the Neck Disability Index

and Numeric Pain Rating Scale in patients with mechanical neck pain. Arch Phys Med Rehabil. 2008;89:69-74. http://dx.doi.org/10.1016/j.apmr.2007.08.126

9. Cleland JA, Glynn P, Whitman JM, Eberhart SL, MacDonald C, Childs JD. Short-term ef-fects of thrust versus nonthrust mobilization/manipulation directed at the thoracic spine in patients with neck pain: a randomized clinical trial. Phys Ther. 2007;87:431-440. http://dx.doi.org/10.2522/ptj.20060217

10. Cote P, Cassidy JD, Carroll L. The Saskatchewan Health and Back Pain Survey. The prevalence of neck pain and related disability in Saskatchewan adults. Spine (Phila Pa 1976). 1998;23:1689-1698.

11. Di Fabio RP. Manipulation of the cervical spine: risks and benefits. Phys Ther. 1999;79:50-65.

12. Evans R, Bronfort G, Nelson B, Goldsmith CH. Two-year follow-up of a randomized clinical trial of spinal manipulation and two types of exercise for patients with chronic neck pain. Spine (Phila Pa 1976). 2002;27:2383-2389. http://dx.doi.org/10.1097/01.BRS.0000030192.39326.FF

13. Flynn T, Fritz J, Whitman J, et al. A clinical prediction rule for classifying patients with low back pain who demonstrate short-term improve-ment with spinal manipulation. Spine (Phila Pa 1976). 2002;27:2835-2843. http://dx.doi.org/10.1097/01.BRS.0000035681.33747.8D

14. Gonzalez-Iglesias J, Fernandez-de-las-Penas C, Cleland JA, Gutierrez-Vega Mdel R. Thoracic spine manipulation for the management of pa-tients with neck pain: a randomized clinical trial. J Orthop Sports Phys Ther. 2009;39:20-27.

15. Gross AR, Aker PD, Goldsmith CH, Peloso P. Pa-tient education for mechanical neck disorders. Cochrane Database Syst Rev. 2000;CD000962. http://dx.doi.org/10.1002/14651858.CD000962

16. Gross AR, Aker PD, Goldsmith CH, Peloso P. Physical medicine modalities for me-chanical neck disorders. Cochrane Database Syst Rev. 2000;CD000961. http://dx.doi.org/10.1002/14651858.CD000961

17. Gross AR, Goldsmith C, Hoving JL, et al. Con-servative management of mechanical neck disorders: a systematic review. J Rheumatol. 2007;34:1083-1102.

18. Gross AR, Hoving JL, Haines TA, et al. A Co-chrane review of manipulation and mobilization for mechanical neck disorders. Spine (Phila Pa 1976). 2004;29:1541-1548.

19. Haldeman S, Kohlbeck FJ, McGregor M. Risk fac-tors and precipitating neck movements causing vertebrobasilar artery dissection after cervical trauma and spinal manipulation. Spine (Phila Pa 1976). 1999;24:785-794.

20. Haldeman S, Kohlbeck FJ, McGregor M. Un-predictability of cerebrovascular ischemia associated with cervical spine manipulation therapy: a review of sixty-four cases after cervi-cal spine manipulation. Spine (Phila Pa 1976). 2002;27:49-55.

21. Hoving JL, de Vet HC, Koes BW, et al. Manual therapy, physical therapy, or continued care

by the general practitioner for patients with neck pain: long-term results from a prag-matic randomized clinical trial. Clin J Pain. 2006;22:370-377. http://dx.doi.org/10.1097/01.ajp.0000180185.79382.3f

22. Hoving JL, Koes BW, de Vet HC, et al. Manual therapy, physical therapy, or continued care by a general practitioner for patients with neck pain. A randomized, controlled trial. Ann Intern Med. 2002;136:713-722.

23. Hurwitz EL, Morgenstern H, Harber P, Komin-ski GF, Yu F, Adams AH. A randomized trial of chiropractic manipulation and mobilization for patients with neck pain: clinical outcomes from the UCLA neck-pain study. Am J Public Health. 2002;92:1634-1641.

24. Hurwitz EL, Morgenstern H, Vassilaki M, Chiang LM. Adverse reactions to chiropractic treatment and their effects on satisfaction and clinical outcomes among patients enrolled in the UCLA Neck Pain Study. J Manipulative Physiol Ther. 2004;27:16-25. http://dx.doi.org/10.1016/j.jmpt.2003.11.002

25. Jaeschke R, Singer J, Guyatt GH. Measurement of health status. Ascertaining the minimal clini-cally important difference. Control Clin Trials. 1989;10:407-415.

26. Juniper EF, Guyatt GH, Willan A, Griffith LE. Determining a minimal important change in a disease-specific Quality of Life Questionnaire. J Clin Epidemiol. 1994;47:81-87.

27. Kelly AM. The minimum clinically significant difference in visual analogue scale pain score does not differ with severity of pain. Emerg Med J. 2001;18:205-207.

28. Korthals-de Bos IB, Hoving JL, van Tulder MW, et al. Cost effectiveness of physiotherapy, manual therapy, and general practitioner care for neck pain: economic evaluation alongside a randomised controlled trial. Bmj. 2003;326:911. http://dx.doi.org/10.1136/bmj.326.7395.911

29. MacDermid JC, Walton DM, Avery S, et al. Mea-surement properties of the neck disability index: a systematic review. J Orthop Sports Phys Ther. 2009;39:400-417. http://dx.doi.org/10.2519/jospt.2009.2930

30. McCormack HM, Horne DJ, Sheather S. Clinical applications of visual analogue scales: a critical review. Psychol Med. 1988;18:1007-1019.

31. Picavet HS, Schouten JS. Musculoskeletal pain in the Netherlands: prevalences, consequences and risk groups, the DMC(3)-study. Pain. 2003;102:167-178.

32. Stratford PW, Riddle DL, Binkley JM, Spadoni G, Westaway MD, Padfield B. Using the Neck Disability Index to make decisions concern-ing individual patients. Physiother Canada. 1999;51:107-112.

33. Tseng YL, Wang WT, Chen WY, Hou TJ, Chen TC, Lieu FK. Predictors for the immediate respond-ers to cervical manipulation in patients with neck pain. Man Ther. 2006;11:306-315. http://dx.doi.org/10.1016/j.math.2005.08.009

34. van der Donk J, Schouten JS, Passchier J, van Romunde LK, Valkenburg HA. The associations

40-03 Boyles.indd 139 2/17/10 2:49:36 PM

140 | march 2010 | volume 40 | number 3 | journal of orthopaedic & sports physical therapy

[ research report ]

@ More inForMationwww.jospt.org

gorithm. Am J Phys Med Rehabil. 2003;82:203-218; quiz 219-221. http://dx.doi.org/10.1097/01.PHM.0000052700.48757.CF

40. Young BA, Walker MJ, Strunce JB, Boyles RE, Whitman JM, Childs JD. Responsiveness of the Neck Disability Index in patients with mechani-cal neck disorders. Spine J. 2009;9:802-808. http://dx.doi.org/10.1016/j.spinee.2009.06.002

fect of spinal manipulation in the treatment of chronic neck pain: a pilot study. J Manipulative Physiol Ther. 1990;13:13-16.

38. Walker MJ, Boyles RE, Young BA, et al. The effectiveness of manual physical therapy and exercise for mechanical neck pain: a ran-domized clinical trial. Spine (Phila Pa 1976). 2008;33:2371-2378. http://dx.doi.org/10.1097/BRS.0b013e318183391e

39. Wang WT, Olson SL, Campbell AH, Hanten WP, Gleeson PB. Effectiveness of physical therapy for patients with neck pain: an individualized approach using a clinical decision-making al-

of neck pain with radiological abnormalities of the cervical spine and personality traits in a gen-eral population. J Rheumatol. 1991;18:1884-1889.

35. Vernon H, Humphreys BK. Chronic mechanical neck pain in adults treated by manual therapy: a systematic review of change scores in random-ized controlled trials of a single session. J Man Manip Ther. 2008;16:E42-52.

36. Vernon H, Mior S. The Neck Disability Index: a study of reliability and validity. J Manipulative Physiol Ther. 1991;14:409-415.

37. Vernon HT, Aker P, Burns S, Viljakaanen S, Short L. Pressure pain threshold evaluation of the ef-

EARN CEUs With JOSPT’s Read for Credit ProgramJOSPT’s Read for Credit (RFC) program invites Journal readers to study and analyze selected JOSPT articles and successfully complete online quizzes about them for continuing education credit. To participate in the program:

1. Go to www.jospt.org and click on “Read for Credit” in the left-hand navigation column that runs throughout the site or on the link in the “Read for Credit” box in the right-hand column of the home page. 2. Choose an article to study and when ready, click “Take Exam” for that article. 3. Login and pay for the quiz by credit card. 4. Take the quiz. 5. Evaluate the RFC experience and receive a personalized certificate of continuing education credits.

The RFC program o�ers you 2 opportunities to pass the quiz. You may review all of your answers—including the questions you missed. You receive 0.2 CEUs, or 2 contact hours, for each quiz passed. The Journal website maintains a history of the quizzes you have taken and the credits and certificates you have been awarded in the “My CEUs” section of your “My JOSPT” account.

40-03 Boyles.indd 140 2/17/10 2:49:38 PM

133JOSPTmar10.p1134JOSPTmar10.p1135JOSPTmar10.p1136JOSPTmar10.p1137JOSPTmar10.p1138JOSPTmar10.p1139JOSPTmar10.p1140JOSPTmar10.p1