Search

Unlabelled: The purpose of this study was to determine the efficacy of a cognitive-behavioral-based physical therapy (CBPT) program for improving outcomes in patients after lumbar spine surgery. A randomized controlled trial was conducted on 86 adults undergoing a laminectomy with or without arthrodesis for a lumbar degenerative condition. Patients were screened preoperatively for high fear of movement using the Tampa Scale for Kinesiophobia. Randomization to either CBPT or an education program occurred at 6 weeks after surgery. Assessments were completed pretreatment, posttreatment and at 3-month follow-up. The primary outcomes were pain and disability measured by the Brief Pain Inventory and Oswestry Disability Index. Secondary outcomes included general health (SF-12) and performance-based tests (5-Chair Stand, Timed Up and Go, 10-Meter Walk). Multivariable linear regression analyses found that CBPT participants had significantly greater decreases in pain and disability and increases in general health and physical performance compared with the education group at the 3-month follow-up. Results suggest a targeted CBPT program may result in significant and clinically meaningful improvement in postoperative outcomes. CBPT has the potential to be an evidence-based program that clinicians can recommend for patients at risk for poor recovery after spine surgery., Perspective: This study investigated a targeted cognitive-behavioral-based physical therapy program for patients after lumbar spine surgery. Findings lend support to the hypothesis that incorporating cognitive-behavioral strategies into postoperative physical therapy may address psychosocial risk factors and improve pain, disability, general health, and physical performance outcomes., (Copyright © 2016 American Pain Society. Published by Elsevier Inc. All rights reserved.)

Background and Purpose: Fear of movement is a risk factor for poor postoperative outcomes in patients following spine surgery. The purposes of this case series were: (1) to describe the effects of a cognitive-behavioral-based physical therapy (CBPT) intervention in patients with high fear of movement following lumbar spine surgery and (2) to assess the feasibility of physical therapists delivering cognitive-behavioral techniques over the telephone., Case Description: Eight patients who underwent surgery for a lumbar degenerative condition completed the 6-session CBPT intervention. The intervention included empirically supported behavioral self-management, problem solving, and cognitive restructuring and relaxation strategies and was conducted in person and then weekly over the phone. Patient-reported outcomes of pain and disability were assessed at baseline (6 weeks after surgery), postintervention (3 months after surgery), and at follow-up (6 months after surgery). Performance-based outcomes were tested at baseline and postintervention. The outcome measures were the Brief Pain Inventory, Oswestry Disability Index, 5-Chair Stand Test, and 10-Meter Walk Test., Outcomes: Seven of the patients demonstrated a clinically significant reduction in pain, and all 8 of the patients had a clinically significant reduction in disability at 6-month follow-up. Improvement on the performance-based tests also was noted postintervention, with 5 patients demonstrating clinically meaningful change on the 10-Meter Walk Test., Discussion: The findings suggest that physical therapists can feasibly implement cognitive-behavioral skills over the telephone and may positively affect outcomes after spine surgery. However, a randomized clinical trial is needed to confirm the results of this case series and the efficacy of the CBPT intervention. Clinical implications include broadening the availability of well-accepted cognitive-behavioral strategies by expanding implementation to physical therapists and through a telephone delivery model.

Study Design: A retrospective review of patient outcomes after lumbar spinal fusion., Objective: To determine whether patients with a fusion ending adjacent to a "degenerated disc" (DDD group) had worse clinical outcomes than patients with fusions ending adjacent to "normal" discs (NL group)., Summary of Background Data: Although it has been suggested that creating a rigid motion segment adjacent to a degenerated segment may negatively impact clinical outcomes after lumbar fusion, this question has not been addressed to our knowledge in the English literature., Methods: Twenty-five consecutive patients treated with lumbar fusion for degenerative instability who had preoperative lumbar spine magnetic resonance imaging, who completed health status questionnaire Short Form 36 (SF-36), and were seen in the office for radiographic follow-up at least 2 years following surgical treatment formed the study group. The magnetic resonance images were reviewed independently by two spine surgeons and rated for the presence of any degenerative changes. Statistical analysis of the SF-36 data was performed with chi2 and Mann-Whitney U testing., Results: Of the 25 patients, 20 were fused adjacent to at least one degenerated level (DDD group), whereas 5 were fused adjacent to a normal level (NL group). At follow-up, SF-36 scores were higher for the DDD group in all eight subgroups, contrary to the research hypothesis. A power analysis demonstrated with at least 98% certainty that if patients in the DDD group had even a 10% lower score in any of the 8 SF-36 subgroups, this study would have detected it., Conclusion: This retrospective review of patients who underwent lumbar fusion for degenerative instability demonstrated no adverse impact on clinical outcomes when the lumbar fusion ended adjacent to a degenerative motion segment. Although a power analysis validated these results with 98% certainty, larger prospective studies are needed to confirm that there is no benefit to include degenerated adjacent segments in a lumbar fusion for degenerative instability.

Spontaneous resorption of herniated nucleus pulposus (HNP) is commonly observed when there is substantial contact of the disc with the spinal canal. We already demonstrated the expression of matrix metalloproteinase (MMP)-3 (stromelysin-1) in the granulation tissues of HNP, suggesting its role in the resorption process of HNP. Recent studies of osteoarthritic cartilages reported an up-regulated expression of metalloproteinases including MMP-7 (matrilysin) and MMP-8 (neutrophil collagenase), suggesting their roles in the matrix degradation. To clarify the expression of MMP-7 and MMP-8 in HNP, immunohistological analysis of various types of HNP was performed. We found MMP-7 was expressed in infiltrated mononuclear cells and chondrocytes, whereas MMP-8 was specifically expressed in chondrocytes. The positive rate for both MMP-7 and MMP-8 significantly increased when HNP was exposed to the epidural space (p < 0.01). Our data suggest that not only MMP-3 but also MMP-7 and MMP-8 may play a role in the resorption process of HNP.

We compared the outcomes from lumbar discectomy for patients who were workers' compensation claimants and/or who were involved in active litigation with patients who underwent elective lumbar discectomy, but who were not involved with either compensation or litigation. Eighty-two consecutive patients who underwent elective lumbar discectomy by the senior author were identified from 1989 through 1994. Those patients who underwent a primary discectomy with a minimum of 6 months' follow-up were studied. Patients were excluded if a spinal fusion was performed or if a multilevel laminectomy procedure was required. Patients were classified as compensation patients if they were involved in either worker's compensation claims or active litigation at the time of the lumbar discectomy. The compensation group was further divided into three subsets of patients: those involved in active litigation without compensation, those involved in both compensation and litigation, and those pursuing workers' compensation claims without litigation. The control group was comprised of patients who were not in any way involved with compensation or litigation. Outcome assessment and ratings were determined independently by the coauthors, not the primary surgeon. Outcome was based on pain, employment status, analgesic use, and level of activity. Fifty-four patients met the inclusion criteria. Average follow-up for the compensation patients was 40 weeks. Follow-up for the noncompensation patients averaged 51 weeks. Eighty-one percent of our patients in the noncompensation group achieved a good result. Only 1 of 27 patients was categorized as having a poor outcome. Conversely, patients who were actively involved in the compensation and/or litigation process had significantly poorer outcomes, with only 29% of the patients receiving a good outcome evaluation (p = < 0.0002). Legal involvement was associated with poorer outcome in compensation patients (p = < 0.001).

The biomechanics of a new spinal implant were evaluated and its dependence on an anterior strut graft was determined. Six fresh-frozen adult porcine spines were used. An L3 corpectomy was created. The spinal nail was inserted intraosseously into the middle column of the L2-L4 vertebral bodies in a trough. Implant drill holes were made for proximal and distal locking bolts in L2 and L4. A locking plate bridged the L2 and L4 trough and anchored the exposed ends of the locking bolts. Testing was performed in axial compression, torsion, and flexion. The axial and torsional stability of the spine instrumented with the spinal nail are similar to results with other devices when used with anterior strut grafting. The axial and flexural stiffnesses of the instrumented spine are independent of strut grafting. This study suggests that this device may be useful for reconstruction of the anterior spine because of instability.

A quantitative serial sectioning technique and a video-imaging procedure were used to obtain precise (20-microns voxels, in a 5 mm x 6 mm x 7 mm test volume) digital images of lumbar vertebral cancellous bone specimens. Conventional stereological image analyses were performed on this data to determine multiplanar (bulk) and surface planar variations in cancellous structural properties. Based upon an error analysis of subgroups of the complete serial planar images, our findings suggest that, for a plane resolution of 20 microns, sectioning at an increment < or = 100 microns along an axis is necessary to obtain accurate data of bulk structural properties of cancellous bone. Planar structural variations obtained from the serial sections of cancellous bone revealed a high degree of complexity and heterogeneity of the bone architecture. Bone area centers deviated from the section centroid and showed a helical variation along the primary or superior--inferior loading axis of samples. In comparison with the base planes (perpendicular to the superior-inferior axis), the lateral planes (parallel to the superior-inferior axis) of spine samples showed smaller mean values of structural indices and a more oriented structure. This structural anisotropy may be related to the functional mechanical anisotropy of the samples.

The influence of load-bearing surface layers on Young's modulus was determined for cancellous bone and polyurethane foam specimens of three different heights cut from each material: 16, 9, and 5 mm for bone specimens (cross-sectional area A = 81 mm2) ranging in density from 0.04 to 0.32 g cm-3; 20, 10, and 5 mm for foam specimens (A = 100 mm2) ranging in density from 0.064 to 0.64 g cm-3. The compressive elastic modulus of the 5 mm bone specimens was significantly smaller (42%) than that of the 16 mm bone specimens. A similar change in elastic modulus was also observed for the foam specimens, but the height effect diminished in a relatively linear manner as the foam density increased. For a height change from 20 to 5 mm, the foam modulus difference ranged from a 41% decrease for the lowest density specimens (p = 0.064 g cm-3) to an insignificant change (< 1.8%) for the highest foam density specimens (p = 0.64 g cm-3). The specimen height effect on elastic modulus was hypothesized to result from a higher deformation (compliance) of the cellular materials in the load-bearing or contacting surface layer than in the bulk. A mathematical formula was derived to predict the variation in measured elastic modulus (E) caused by the contacting surface layer modulus (Ecs) to bulk modulus (Eb) ratio (Y = Ecs/Eb) and contacting surface layer height (hcs) to sample height (h) ratio (t = hcs/h): E/Eb = Y/[Y(1-t)+t]. Predicted differences in the measured modulus associated with graduated specimen heights were verified experimentally by regional strain measurements of the foam and bone samples using an optical microscope. The influence of free-surface layers was also determined for the foam specimens with cross-sectional areas 1.21 and 10 times the load-bearing area of a load platen, but the measured differences were small (< 10%) compared to that of the load-bearing surface layer influence. In future compressive mechanical testing of cellular materials, specimens < 10 mm in height should be avoided, particularly specimens with low bulk density or high porosity.

Burst fractures of the lumbar spine that are located below the thoracolumbar junction present a challenge when operative management is indicated. Short-segment instrumentation offers the advantage of incorporating fewer motion segments in the fusion, but may not provide adequate long-term stabilization. The goal of this study was to assess the axial stiffness and torsional rigidity of several short-segment instrumentation procedures. Compressive axial stiffness and torsional rigidity were measured in six intact porcine lumbar spines (L1-L5). A corpectomy was performed to simulate a burst fracture injury and decompression. Posterior instrumentation, posterior instrumentation with an anterior strut (a wood block), and anterior instrumentation with an anterior strut one level above and one level below the fracture site were applied as treatment strategies. VSP plates (Acromed, Cleveland, OH) for posterior instrumentation and the Kaneda system (Acromed, Cleveland, OH) for anterior instrumentation were used. Load-displacement and torque-angle plots were generated and used to calculate 144 estimates of axial stiffness and 144 estimates of torsional rigidity for these constructs. These analyses showed that, in comparison with the intact spine, posterior instrumentation alone was an average of 76% less stiff axially, posterior instrumentation with an anterior strut was 3% more stiff (not significantly different from intact), and anterior instrumentation with an anterior strut was 15% more stiff. Posterior instrumentation alone was an average of 30% less rigid in torsion, posterior instrumentation with an anterior strut was 26% less rigid, and anterior instrumentation with an anterior strut was 24% less rigid than the intact spine.(ABSTRACT TRUNCATED AT 250 WORDS)

A nondestructive mechanical testing technique was used to assess midsagittal plane variations in the mechanical properties of human lumbar vertebral bone adjacent to the intervertebral discs of normal and degenerated spines. Bone mechanical properties were compared with midsagittal plane measurements of intervertebral disc (IVD) physical-chemical properties. In the normal spine a heterogeneous midsagittal plane distribution of subdiscal bone stiffness, IVD hydration and fixed charge density was found. Segments with degenerated discs exhibited a lower amount of midsagittal variation in subdiscal bone and IVD properties. A positive linear correlation between subdiscal bone stiffness and IVD fixed charge density was also found, suggesting that an interdependence of IVD and bone properties exists. These results support previous findings and are hypothesized to be an adaptive response to changes in the stress environment in the IVD.

A new technique incorporating a motion analysis system and a materials testing machine was used to investigate regional differences in the tensile mechanical properties of the lumbar spine anterior longitudinal ligament (ALL). Bone-ALL-bone specimens were prepared from young human cadaveric motion segments with no disc or bony pathology. Each specimen was distracted until failure at a constant crosshead displacement rate of 2.5 mm s-1 (approximately 1.0% strain per second). Strains were evaluated from digitized video recordings of markers attached to the ALL at 12 sites along its length and width, including the ligament substance and insertions. The 'overall' strain in the ligament was calculated from the outermost pairs of markers along the ligament length. The average tensile strength, the 'overall' tensile modulus and the 'overall' strain of the ALL at failure were 27.4 MPa (S.D. 5.9), 759 MPa (S.D. 336) and 4.95% (S.D. 1.51), respectively. Large and significant variations in the strains were present along the width and length of the ALL. Peak substance strains were over twofold greater than peak strains at the ligament insertion sites, whereas across the ligament width, peak strains in the outer portion of the ligament were over 40% greater than in the central region. Failure consistently occurred in the ligament mid-substance and ultimate strains at the ligament failure site averaged 12.1% (S.D. 2.3). These results indicate that the strains are highly nonuniform in the normal ALL.

This article describes regional variations in cancellous bone morphology within the human lumbar spine of a 60-year-old woman. An automated imaging technique, based upon the method of directed secants, was used to quantify variations in bone porosity and trabecular size, distribution, and orientation within serial transverse histological sections of the L1 and L2 vertebral centrum. Several morphologically distinct levels of organization were observed along the axis of the centrum. The superior and inferior sections of the centrum were made up of an open-celled network of trabeculae, while the translational and middle sections consisted primarily of plate-like trabeculae forming a closed-cell structure. The densest and most oriented bone was found in the superior and inferior sections of the posterior-lateral centrum, whereas least dense and least oriented bone was found in the central third of the anterior centrum. Average trabecular thickness did not vary appreciably from level to level, but varied considerably within a given level, particularly in the inferior sections of the centrum. The number of trabeculae was lowest in the central sections of the centrum, where plate-like trabeculae were most prevalent. The topological architecture of different levels within the centrum exhibited a remarkable degree of organization, which suggests that the spine, like other skeletal structures, has adapted to functional stresses produced during daily activities. Detailed morphological data are tabulated and should be of use to investigators interested in modeling the three-dimensional structural and physical properties of the lumbar spine.

In using the finite element method to examine certain aspects of the mechanical behavior of the human lumbar spine, most investigators have made numerous simplifying assumptions regarding the geometric and material data used to build a model of the spine. Since there are no specific rules for choosing geometric and material data for a "normal" human lumbar spine, considerably different types of models have been used by investigators in their finite element studies. In this study, variations in model geometric and material properties are shown to significantly affect the finite element results of an axisymmetric model of the human lumbar spine under axial compression. The Young's moduli of the cancellous bone and intervertebral disc annulus, Poisson's ratios of the cartilaginous end-plate and disc annulus, the width of the disc annulus, the height of the disc, and the ratio of the disc nucleus pressure and axial pressure are recognized as the parametric variables that most significantly affect the finite element solution.

Flexion-distraction injuries to the lumbar spine frequently occur with concomitant abdominal viscus injury. We encountered 16 patients with flexion-distraction lumbar spine injuries. Half of them also suffered abdominal viscus rupture. Lap safety belts were most frequently associated with these injuries; however, the incorrect, underarm use of the shoulder harness was common in our series. The diagnosis of the spinal injury was frequently delayed when abdominal viscus injury occurred together with a flexion-distraction spinal injury. Thorough physical and radiographic examination of the spine as well as a detailed history, including seat belt use, is necessary to diagnose these injuries properly.

The purpose of this study was to determine whether there was any difference in the clinical outcome between groups of patients treated with lumbar discectomy and vertebral endplate curettage as compared with disc fragment excision without endplate curettage. Eighty-three patients requiring lumbar disc excision for herniated nucleus pulposus were evaluated retrospectively at minimum 2-year follow-up. Forty-three patients had undergone fragment excision and disc space curettage at one center, whereas 40 patients underwent fragment excision without curettage at two other centers. There was no increased rate of reherniation or reoperation in the excision-only group. Vertebral endplate curettage carries a risk of annular penetration and damage to the great vessels. This study shows that this potentially dangerous step of the operation is unwarranted. In addition, patients who underwent endplate curettage had a higher incidence of low-back pain on follow-up.

Commonly used clinical measurements of spinal flexibility in the sagittal and frontal planes were examined as predictors of future back pain reports within industry. The study sample consisted of 3,020 aircraft manufacturing employees who were examined and tracked for more than 4 years for reports of back pain. Modified Schober, sit-and-reach, and lateral bending measurements were not significantly associated with risk of future back pain reporting, nor were any trends present. There was a statistically significant relationship between decreased flexibility and reports of current or previous back problems. However, the differences in flexibility between subjects with and without a history of back problems were too small to be of practical significance.

In vivo creep-recovery and disc pressure measurements were performed on the lumbar spine of immature and mature swine. The creep-recovery measurements were performed using a custom materials testing apparatus designed to apply static or dynamic loads to the spine of anesthetized animals. A series of three separate experiments were performed to assess the effects of: (I) animal death, (II) graded injury to the disc anulus, and (III) respiratory mechanics on the biomechanical response of the porcine L1-L3 vertebral unit (VU). In Experiments I and II, creep rate, modulus, and viscosity parameters were computed using a three-parameter solid rheological analysis of the displacement-time response recorded during the application of a 300-N load. In Experiment III, the effects of respiratory volume and frequency changes on disc pressure were assessed in the unloaded, statically loaded, and immobilized porcine VU. Our results indicated that the adult VU tended to be stiffer, deform or creep more slowly, and had a significantly higher viscosity than the VU of immature pigs. The results of Experiment I demonstrated that the biomechanical response for the VU was significantly altered by the death of the animal; the VU of the living animal (adolescent or mature) was more compliant and deformed at a faster rate than the VU of the same animal after death. Disc injury produced changes in stiffness, viscosity, and creep rate analogous to that of aging, and on the basis of the graded injuries created in this study, it appears that a small defect in the annulus is just as deleterious as removing a large section of anular material. The results of Experiment III indicated that respiration plays an important role in the normal, in vivo mechanical and nutritional behavior of the porcine VU. Altogether, these results demonstrate that, in the absence of normal physiologic conditions, one may not be able to reliably predict the mechanical response of the lumbar spine, and suggest that standards for the testing, handling, and storage of biologic tissue should be established.

The in vitro viscoelastic "creep" behavior was examined in 18 cadaveric human lumbar motion segments subjected to static axial compressive loads. Axial deformation was followed for 30 min under constant applied load. Compressive material constants (moduli and viscosity coefficients) were then determined for each intervertebral disc using a linearization method based on a Taylor series expansion of experimental data for the "three parameter" viscoelastic creep model. The degree of disc degeneration and bone mineral content (BMC) were also assessed. Good correlation between the experimentally determined and model predicted strain values were found, with the average error less than 1%. We found that motion segments from older and more degenerated lumbar discs were less stable and had lower material constants than segments from younger and less degenerated discs. Material constants and BMC correlated closely, suggesting that an interdependency of disc and vertebral body properties exists. No correlation between the creep characteristics and disc height, disc area, segment level, or sex were noted.

The viscoelastic properties of normal and surgically altered lumbar intervertebral disks from eight immature swine were examined in vivo. Rheological models were used to mathematically characterize the compressive creep-relation behavior of the disks before and after alteration by either chemonucleolysis [chymopapain (Chymodyactin) injection] or denucleation. In the normal disks, a significant modulating effect of respiration was observed that tended to increase the creep deformation response in comparison to that observed during similar in vitro and in situ studies. These results suggest that, in terms of assessing the absolute magnitude of the viscoelastic properties of lumbar disks, the influence of normal physiological function on adjacent vertebrae and surrounding tissues cannot be neglected. Preliminary results obtained from the experimentally altered disks indicated that partial denucleation primarily affected the initial stiffness behavior of the disk (24% decrease in elastic modulus), whereas chemonucleolysis caused changes in both the time-dependent (15% increase in creep rate) and instantaneous (23% decrease in elastic modulus) properties of the disk. Both partial denucleation and acute chemonucleolysis produced biomechanical changes that were comparable to grade II or slightly degenerative, age-related changes found in human lumbar intervertebral disks.

This study examines different morphologic measurements in the evaluation of patients with lumbar spinal stenosis. Preoperative CT scans from 24 patients who underwent surgery for central lumbar stenosis were analyzed. No correlation was observed between the size of the bony spinal canal and the size of the dural sac. A new measurement, the transverse area of the dural sac, is introduced. Normal values are provided. Correlation between the cross-sectional area of the dural sac and the anteroposterior diameter of the dural sac was excellent.

A retrospective review was made of 30 patients who had failed multiple traditional surgical procedures for low-back pain, sciatic pain, or both. The most common cause of the poor results appeared to be failure of initial selection, even though all patients appeared to meet traditional indications for operative intervention. Psychosocial problems (eg, drug abuse, alcoholism, marital discord, personality disturbances) were prevalent, but were not initially recognized. Thoughtful assessment of patients being considered for initial lumbar spine surgery, which includes objective psychological testing, may reduce the incidence of unsuccessful lumbar spine surgery.

Reports of back discomfort and even back injuries during isometric strength testing in specific lifting positions indicated an analysis of the loads on the lumbar spine during this type of testing. A biomechanical analysis, which has been validated against EMG and intravital disc pressure measurements, was used for the calculations of the loads in four test persons. The calculations indicated compressive loads on L3 ranging from 5000-11,000 N during squat and torso lifting. Such loads in vitro have been found to cause structural failures of the vertebral endplates. Similar loads also may result in damage to the spine in vivo.

Seventeen cadaveric human lumbar motion segments from eight spines were cyclically loaded in vitro under axial compression. Loading frequency and magnitude were chosen to simulate rigorous activity within an in vivo physiological level. The load magnitude was determined as a percentage of the ultimate compressive load, the latter estimated from the bone mineral content (BMC) of lumbar vertebrae determined by dual-photon absorptiometry. Following testing, the degree of macroscopic disc degeneration was assessed and the type of fracture in each specimen was determined from serial sagittal sections. Fractures were found in all but one specimen. Three types of fractures were formed: the node of Schmorl and Junghanns (type I), central endplate fracture (type II), and a crush or burst fracture (type III). The results suggested that type I fractures were predominantly associated with segments with normal discs, type II fractures were found primarily in segments with moderately degenerated discs, and type III fractures were associated with segments that failed on the first cycle. Segment stiffness and fatigue strength (cycles to failure) were correlated with disc degeneration, age, and segment BMC, the latter an in vivo measure of bone density. Fatigue strength also decreased in proportion to a power coefficient with increasing relative stress (cyclic stress range/ultimate stress).

The compressive mechanical properties of human lumbar vertebral trabeculae were examined on the basis of anatomic origin, bone density, and intervertebral disc properties. Trabecular bone compressive strength and stiffness increased with increasing bone density, the latter proportional to strength and stiffness to the one-half power. Regional variations within each segment were found, the most prevalent differences occurring in regions of bone overlying the disc nucleus in comparison with bone overlying the disc anulus. For normal discs, the ratio of strength of bone overlying the disc nucleus to bone overlying the disc anulus was 1.25, decreasing to 1.0 for moderately degenerated discs. These results suggest that an interdependency of trabecular bone properties and intervertebral disc properties may exist.

Between January 1, 1975 and November 1, 1986, 77 patients with acute unstable thoracic or lumbar spine fractures underwent reduction, posterior stabilization with dual Harrington distraction rods, and fusion with autogenous iliac crest bone graft at Vanderbilt University Medical Center. Beginning March 1985, in 25 patients, segmental interspinous wires were employed, in addition to the Harrington rods, to augment the surgical construct. Clinical and radiographic analysis was performed to determine if differences existed between the two groups. Average time to surgery, hospital stay, and time to brace discontinuance were lower in the segmental wire group. Pain and work status at final follow-up were similar in the two groups. There was a higher incidence of superior and inferior hook migration and reoperation for this complication in patients treated with Harrington rods alone. There was no significant difference in postoperative correction of deformity, although at follow-up, there was greater recurrence of deformity in the Harrington rod group. In summary, the use of segmental wires to supplement Harrington rod fixation appears to offer advantages over Harrington rods alone, with minimal or no increased risk.