SPINE Volume 26, Number 14, pp E330–E337 ©2001, Lippincott Williams & Wilkins, Inc. Prospective Dynamic Functional Evaluation of Gait and Spinal Balance Following Spinal Fusion in Adolescent Idiopathic Scoliosis Lawrence G. Lenke, MD,* Jack R. Engsberg, PhD,† Sandy A. Ross, MHS,† Angela Reitenbach, MHS,† Kathy Blanke, RN,* and Keith H. Bridwell, MD* Study Design. Prospective evaluation of gait and spinal–pelvic balance parameters in patients with adolescent idiopathic scoliosis undergoing a spinal fusion. Objective. To evaluate changes in gait and three-dimensional alignment and balance of the spine relative to the pelvis as a consequence of spinal fusion. Summary of Background Data. Preoperative and postoperative spinal radiographs have been the major forms of outcome analysis of adolescent idiopathic scoliosis fusions. The use of optoelectronic analysis for posture and gait has gained acceptance recently. However, there is a paucity of data quantifying, comparing, and correlating structural and functional changes in patients undergoing scoliosis fusion surgery including upright posture and gait. Methods. Thirty patients with adolescent idiopathic scoliosis undergoing an instrumented spinal fusion were prospectively evaluated. Coronal and sagittal vertical alignment was evaluated on radiographs (CVA-R, SVA-R), during upright posture (CVA-P and SVA-P), and during gait (CVA-G, SVA-G). Transverse plane alignment was evaluated by the acromion–pelvis angle during gait. Results. Gait speed was significantly decreased (P ⬍ 0.05) between preoperative (129 ⫾ 16 cm/sec) and 2-year postoperative (119 ⫾ 16 cm/sec) testing sessions. Decreasing gait speed was the result of significantly reduced cadence and decreased stride length. There were no significant differences for lower extremity kinematics over the entire gait cycle. Spinal–pelvic balance parameters showed significant improvement in mean CVA-R, CVA-G (P ⬍ 0.05), then unchanged CVA-P at 2 years postoperation. CVA-P was relatively unchanged while the mean CVA-G also showed significant improvement from preoperation (2.2 ⫾ 2.4 cm) to 2 years postoperation (1.3 ⫾ 1.3 cm)(P ⬍ 0.05). The mean SVA-R, SVA-P, and SVA-G were unchanged at 2 years postoperation (P ⬎ 0.05). The acromion–pelvis angle during gait at maximum shoulder rotation was statistically improved at 1 year (P ⫽ 0.002) and 2 years (P ⫽ 0.001) after surgery. Importantly, CVA-P correlated with CVA-G, and SVA-P correlated with SVA-G to the P ⬍ 0.05 level. Conclusions. Patients with adolescent idiopathic scoliosis undergoing spinal fusion show slightly decreased gait speed at 2 years postoperation without any change in From the *Department of Orthopaedic Surgery, Washington University School of Medicine, and †Human Performance Laboratory, Barnes-Jewish Hospital, St. Louis, Missouri. Presented at the Scoliosis Research Society 33rd Annual Meeting, New York, New York, September 1998. Acknowledgment date: October 10, 2000. First revision date: January 3, 2001. Acceptance date: March 29, 2001. Device status category: 11. Conflict of interest category: 15. E330 lower extremity kinematics. Spinal–pelvic balance parameters are improved in the coronal plane and unchanged in the sagittal plane radiographically and during standing posture and gait. Transverse plane parameters also are improved at maximum shoulder rotation during gait. [Key words: adolescent idiopathic scoliosis, coronal balance, sagittal balance, optoelectronic analysis, spinal–pelvic balance, gait] Spine 2001;26:E330 –E337 Spinal fusions are still the primary means of correcting a scoliosis deformity and thereby halting progression. The most common patient to have this kind of surgery is an adolescent who has idiopathic scoliosis12,17 and whose etiology is still unknown.5 Frequently, this fusion extends from the thoracic region into varying portions of the lumbar spine and can be performed via the posterior, anterior, or combined route. Although partial correction of the curvature is obtained, it is at the expense of removing normal intervertebral motions that exist in the scoliotic spine.19 The preoperative analysis and the postoperative documentation of surgical results are currently performed via long cassette radiographs of the spinal column in the coronal and sagittal projection. There are broad radiographic goals for scoliosis correction in the coronal plane and they include the following: 1) a balanced spine with a plumb line from C7 bisecting the sacrum; 2) a balanced thoracic and lumbar spine with the top and bottom of the instrumentation and fusion falling within the stable zone of Harrington; and 3) for fusions into the mid and lower lumbar spine (L2 and below) that the lowest instrumented vertebra (LIV) be horizontal to the pelvis, bisected by the center sacral line and neutrally rotated.13,18,21 In the sagittal plane the radiographic goals are as follows: 1) a balanced sagittal plane with the plumb line from C7 falling at/or behind the lumbosacral disc; 2) normal thoracic, thoracolumbar, and lumbar sagittal plane alignment within the instrumented and uninstrumented regions; and 3) to have junctions between the regional segments of thoracic kyphosis and lumbar lordosis neutral while avoiding a junctional kyphosis.21 Since the advent of spinal instrumentation as an adjunct to operative correction and fusion, radiographs have been the major forms of outcome analysis of scoliosis fusions. With the current use of segmental spinal instrumentation systems, three-dimensional analysis of the deformity preoperation and postoperation has become important. This has improved radiographic results, Adolescent Idiopathic Scoliosis • Lenke et al E331 Table 1. Demographic and Surgical Data for 30 Patients With AIS Patient Number Patient Age 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 16 18 ⫹ 8 12 ⫹ 8 18 ⫹ 1 12 ⫹ 7 12 ⫹ 8 15 ⫹ 7 13 ⫹ 3 13 ⫹ 7 17 ⫹ 7 14 ⫹ 11 13 ⫹ 2 15 ⫹ 2 17 ⫹ 11 13 ⫹ 11 14 ⫹ 6 11 ⫹ 6 15 ⫹ 2 13 ⫹ 9 14 ⫹ 9 14 ⫹ 11 14 ⫹ 2 17 ⫹ 1 14 ⫹ 6 13 ⫹ 2 18 ⫹ 5 13 ⫹ 7 18 ⫹ 8 13 ⫹ 2 16 ⫹ 9 Sex Type of Fusion ASF/PSF/Combo Total Number of Fusion Levels LIV F F F F F M F F M F F F F M F F F F F F F F F F F F F F F F PSF PSF PSF PSF PSF PSF PSF PSF PSF PSF PSF PSF PSF PSF PSF PSF ASF ASF ASF ASF ASF ASF ASF ASF ASF ASF ASF ASF ASF/PSF ASF/PSF T4–L2 (10) T2–L1 (11) T2–L1 (11) T4–L2 (10) T4–T12 (8) T4–L4 (12) T3–L3 (12) T4–L1 (9) T3–L2 (11) T2–L2 (12) T5–L4 (13) T4–T12 (8) T4–L3 (11) T4–L1 (9) T3–L3 (12) T3–T12 (9) T5–T11 (6) T4–T11 (7) T5–T12 (7) T11–L3 (4) T5–T11 (6) T5–L1 (8) T5–T12 (7) T10–L3 (5) T6–L1 (7) T5–T11 (6) T5–T13 (8) T5–T11 (6) T2–L3 (13) T12–L2 (2) L2 L1 L1 L2 T12 L4 L3 L1 L2 L2 L4 T12 L3 L1 L3 T12 T11 T11 T12 L3 T11 L1 T12 L3 L1 T11 T13 T11 L3 L2 LIV ⫽ lowest instrumented vertebrae. especially with increased emphasis placed on the sagittal plane.13,21 Results for correction of axial plane deformity have been less successful.1 Nevertheless, although changes in static alignment of the spinal column as assessed by standard radiographs do quantify changes in structure, they do not quantify changes in function from a dynamic perspective. A paucity of data exists quantifying, comparing, and correlating structural and functional changes in patients undergoing scoliosis fusion surgery. The purpose of this study is twofold: 1) to evaluate changes in gait as a consequence of spinal fusion in adolescents with idiopathic scoliosis and 2) to evaluate changes in the three-dimensional alignment and balance of the spine relative to the pelvis during standing posture and gait. To this end we hope to gain valuable insight as to the functional effects of a scoliosis fusion from a more dynamic perspective. Materials and Methods This study details 30 patients (28 female and two male) with adolescent idiopathic scoliosis (AIS) who underwent an instrumented spinal fusion to treat their scoliosis. The mean age of the patients was 14 years range 12–18 years), and all were treated with one of three types of surgical procedures involving a posterior spinal fusion (PSF, n ⫽ 16), an anterior spinal fusion (ASF, n ⫽ 12), or a combined ASF/PSF (n ⫽ 2). All patients treated with a PSF had posterior segmental spinal instrumentation placed for correction of deformity and performance of a spinal fusion using autogenous bone harvested from the iliac crest or the convex ribs to improve any associated rib prominence. Patients with an ASF had a single anterior screw rod instrumentation augmented with rib autograft placed in the disc spaces for the anterior fusion. Patients with a combined ASF/PSF had anterior release and autogenous rib fusion performed followed by a PSF procedure with instrumentation as described above. The demographic patient data along with the type of surgery performed, total amount of fusion levels, and the lowest instrumented vertebrae are provided for each patient in Table 1. The mean number of levels fused was 10.5 (range 8 –13). The LIV was T11 (n ⫽ 5), T12 (n ⫽ 5), T13 (n ⫽ 1), L1 (n ⫽ 6), L2 (n ⫽ 5), L3 (n ⫽ 6), and L4 (n ⫽ 2). All patients had preoperative, 1-year, and 2-year postoperative long cassette spinal radiographs performed in the coronal and sagittal planes. All coronal curves were measured by the Cobb method, and radiographic coronal balance (i.e., coronal vertical alignment [CVA-R])18 was measured as a C7 plumb line distance in centimeters from the midsacral line. Cobb measurements also were made in the sagittal plane measuring thoracic kyphosis (T5–T12), lumbar lordosis (T12–S1), and overall radiographic sagittal balance (i.e., sagittal vertical alignment [SVA-R])18 as a C7 plumb line measured in centimeters as the distance from the posterior–superior edge of S1. A positive value indicated that the C7 plumb line fell in front of the posterior superior aspect of S1, whereas a negative value indicated that this line fell in back of this point.24 To prepare the subject for the videographic gait analysis, 36 retroreflective surface markers were placed at selected locations on segments of the lower extremities, trunk, and head of each patient. A subset of 24 markers were used in the analysis of the present investigation (Table 2). For the collection of the gait data, each subject walked barefoot along a 9-m walkway and had video data from a 6 camera HiRes Motion Analysis Corporation System (Santa Rosa, CA) collected during the middle 2 m (Figure 1). Five to seven trials of data were collected from each subject. In addition, one trial of posture data was collected while the patient stood in a relaxed manner. For the posture and gait trials of each subject, the surface marker location data were converted to three-dimensional coordinates as a function of time, then tracked and edited to Table 2. Surface Markers Used to Represent Body Segments or Specific Vertebrae Segment or Vertebrae Shoulders Vertebrae Pelvis Thighs Legs Feet 1st Surface Marker 2nd Surface Marker Right acromion process C7 T10 Right ASIS Greater trochanter Fibular head Superior navicular Left acromion process S2 L4 Left ASIS Mid anterior portion of femur Mid anterior portion of tibia Posterior calcaneus 3rd Surface Marker S2 Lateral epicondyle of femur Lateral malleolus 5th metatarsal head 332 Spine • Volume 26 • Number 14 • 2001 Figure 3. Mean acromion pelvis angle (APA-G) during the gait cycle showing minimum degrees of difference occurring just after the 20% of the gait cycle, with the maximum greatest range of motion seen just past the halfway point of the gait cycle. Figure 1. Patient with preoperative scoliosis undergoing the optoelectronic video gait analysis. produce quantified measurements of standing posture and selected gait parameters. The standing posture measurements were coronal vertical alignment (CVA-P) and sagittal vertical alignment (SVA-P). These quantify the horizontal distance from S2 to a vertical line dropped from C7 in the coronal and sagittal planes, respectively. They were designed to be similar to the previously described CVA-R and SVA-R radiographic parameters. The gait data were analyzed to quantify trunk alignment, limb motion, and basic gait parameters. Seven distinct analyses were performed. The first four were the coronal vertical alignment (CVA-G) and the sagittal vertical alignment (SVA-G), for both right and left steps. Specifically, the CVA-G and SVA-G were defined as above, but the values were from initial contact during gait of the right and left feet. The initial contact time was selected to represent the entire gait cycle because little change was observed in the values over the cycle (Figure 2). The final three gait parameters in the first set were related to the transverse plane motion of the two acromion process markers relative to the two pelvis ASIS markers. Specifically, the acromion–pelvis angle during gait (APA-G) was the transverse plane angle created by the line formed from the right and left acromion process markers, relative to the line created by the two ASIS markers. A positive APA-G value was calculated if the right acromion process marker was oriented anterior to the right ASIS marker (i.e., protracted right shoulder), and a negative value was calculated if the right acromion process was oriented posterior to the right ASIS marker (i.e., retracted right shoulder). The first of the three parameters was the maximum APA-G (APA-G Max)(Figure 3). The second was the minimum APA-G (APA-G Min). The third parameter was the range of motion over the entire gait cycle in the transverse plane and was defined to be the difference between the maximum and minimum APA values (APA-G ROM). We also analyzed basic gait parameters including gait speed, stride length, cadence, and step width.7 These were analyzed from a subset of 10 of the 30 original patients. This analysis first included a visual inspection of the ankle, knee, and hip joints as a function of the gait cycle in the sagittal, coronal, and Figure 2. Mean coronal (CVA-G) and sagittal (SVA-G) spinal balance during the entire gait cycle. Adolescent Idiopathic Scoliosis • Lenke et al E333 Figure 4. Lower extremity kinematic data including right knee flexion– extension, right angle inversion– eversion, right ankle dorsiflexion– plantarflexion, and right foot rotation during the entire gait cycle showing minimal changes from preoperation to 1–2 years postoperation. There are slight differences in the able bodied (AB) patient data. transverse planes for the three test sessions (preoperative and 12 and 24 months postoperative). Because little variation was observed across test sessions, four representative joint angles over the gait cycle in a single plane were chosen to represent the lower extremity gait changes (Figure 4). These were as follows: 1) right sagittal plane knee angle, 2) right sagittal plane ankle angle, 3) right coronal plane ankle angle (eversion–inversion), and 4) right transverse plane foot orientation (foot progression angle). From each curve the numerical value at initial contact was selected for use in the statistical analysis. The statistical analysis included a multiple analysis of variance to determine any differences among the preoperative and two postoperative test sessions. In addition, a Pearson Product Moment Correlation Coefficient (r) was used to determine whether respective relationships existed among the CVA and SVA variables. Results Radiographic The radiographic results are listed in Table 3. The mean preoperative thoracic Cobb measurement was 57° (range 48 –77°), decreasing to a mean 27° (range 13– 40°) postoperation, for an average 53% instrumented correction (range 19 –77%). The radiographic coronal plumb line (CVA-R) improved from a mean 0.76 cm off balance preoperative (range 0 –5.9 cm) to 0.01 cm (range 0 –3.3 cm) at 1 year postoperation, and 0.21 cm (range 0 –3.0 cm) at 2 years postoperation (P ⬍ 0.05). There were no significant changes noted in the radiographic and postural sagittal data at 2 years postoperation (P ⬎ 0.05). Table 3. Radiographic Data for All 30 Patients Sagittal Alignment Cobb Patient Number 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 CVA-R T5–T12 L1–Sac SVA-R Pre 2 yr Pre 1 yr 2 yr Pre 2 yr Pre 2 yr Pre 1 yr 2 yr 53 55 77 57 57 48T/58L 45T/61L 56 62 60 55T/62L 61 53 52 53 48 59 63 69 34 57 82 40 22T/49TL 66 50 66 52 82 42TL 57° 19 22 40 15 35 33T/30L 18T/23L 25 28 47 36T/27L 42 13 27 38 39 33 35 35 16 18 37 23 26T/28TL 23 38 35 25 28 19 27° MEAN 0.8R 1.8L 2.5L 2.3R 4L 2.6R 1.9R 0.4L 1.2R 1.1R 0.9R 2.6L 1.6R 2.6L 2.3R 0 1.5R 1.6R 0.4R 2.9R 0.7L 0.9R 5.9R 4.7R 0.9R 3.9R 1.6L 2R 2R 2.4R 0.75926 0.4L 0.31L 1.5L 1L 1.6L 2.1R 0 1.1L 0.1R 0.1R 0.3R 3.3L 3R 0.1R 0.1R 2R 3.2L 1.5L 2.3R 2.2R 1.0R 0.1R 2.9R 2.3R 0.2L 2.4L 2.5L 2.2L 0.6R 1.4R 0.0107 0.9L 0.3R 0 0.3R 3L 0.8R 1.2R 1.5L 0.2R 2.2R 0 2.2L 1.8R 0.8L 0.2R 0 1.3L 1.2L 0 0.4R 0.3L 1.9R 1R 0.1R 0.2R 1.2L 1.4L 3L 0.2R 0.9 0.2077 36 30 43 25 39 18 23 29 24 37 36 34 41 30 46 16 28 13 19 14 4 20 30 22 9 45 58 ⫺5 35 48 28.071 20 28 29 7 24 22 22 23 16 43 40 18 30 15 28 20 31 20 17 29 23 31 27 22 27 46 55 10 27 46 26.852 ⫺61 ⫺65 ⫺60 ⫺62 ⫺64 ⫺42 ⫺55 ⫺79 ⫺57 ⫺58 ⫺81 ⫺48 ⫺84 ⫺82 ⫺71 ⫺54 ⫺76 ⫺66 ⫺63 ⫺55 ⫺66 ⫺72 ⫺74 ⫺59 ⫺58 ⫺59 ⫺66 ⫺59 ⫺73 ⫺65 ⫺64.46667 ⫺52 ⫺62 ⫺53 ⫺56 ⫺61 ⫺58 ⫺70 ⫺71 ⫺47 ⫺68 ⫺74 ⫺52 ⫺68 ⫺72 ⫺72 ⫺58 ⫺60 ⫺72 ⫺57 ⫺70 ⫺63 ⫺69 ⫺68 ⫺59 ⫺56 ⫺70 ⫺68 ⫺50 ⫺60 ⫺69 ⫺62.407 ⫺8.1 ⫺0.9 ⫺2.2 ⫺7.2 ⫹1.9 ⫺4.4 ⫺1.1 ⫺3 ⫹4.5 ⫹.4 ⫺9.3 ⫹4.9 ⫺2.3 ⫹0.8 ⫺1.3 ⫹0.3 ⫺8.4 ⫹3.3 ⫺4.8 ⫺0.7 0.6 ⫺5.2 ⫺0.3 ⫺2.1 ⫹0.9 ⫹4.1 ⫺5.1 ⫺2.7 ⫺3.3 ⫺2.1 ⫺3.519 ⫺0.2 ⫺0.9 ⫹3.3 ⫺6 ⫺0.8 ⫺3.8 ⫺1.2 ⫺2.5 ⫺3.6 ⫺4.5 ⫺4.1 ⫹0.9 ⫹0.4 ⫹3.6 ⫺7.7 ⫺2.8 ⫹1.7 ⫺3.4 ⫹3.4 ⫺4.2 ⫺4.1 ⫺7.6 ⫺1.1 ⫺0.9 ⫹1.3 ⫺2.6 ⫺1 ⫺1.6 ⫺6 ⫺2.7 ⫺3.145 ⫹3.2 ⫹0.5 ⫹1.2 ⫺2.5 ⫺1 ⫹1.7 ⫺0.7 ⫺5.6 ⫹1.7 ⫺4.8 0 ⫹1.7 ⫹1.1 ⫹1.6 ⫺5.6 ⫺4.1 ⫺6.4 ⫹1.1 0 ⫺4.2 ⫺0.4 ⫺1.6 ⫹0.1 ⫹1.5 ⫹1.7 ⫹1.5 1.6 ⫺0.4 ⫺6.3 ⫺3.2 ⫺3.0 E334 Spine • Volume 26 • Number 14 • 2001 Gait Analysis Gait speed was significantly decreased (Table 4) between the preoperative test session (129 ⫾ 16 cm/sec) and the two postoperative sessions (12 months, 120 ⫾ 16 cm/ sec; 24 months, 119 ⫾ 16 cm/sec). The decrease in gait speed was a result of a significantly reduced cadence for both comparisons (Table 4) and a significantly decreased stride length for the preoperative to 24-month postoperative test sessions. There were no differences between the two postoperative sessions for speed, cadence, and stride length. There was a significant decrease in stride width between the preoperative condition and the 12-month postoperative condition, but not the 2-year postoperative condition. There was no stride width difference between the two postoperative conditions, and no difference in hip joint/pelvic motion. Using the value for initial contact (0%) of the gait cycle for each joint measurement (Figure 4) for the subgroup of 10 subjects indicated there were no significant differences between test sessions preoperation, 1 year, and 2 years postoperation (Table 5). Thus, the results showed no differences in the lower extremity kinematic data for these 10 patients, and no further analysis was performed. Spinal–Pelvic Alignment and Balance Coronal plane spinal–pelvic balance during upright posture (CVA-P) averaged ⫺0.5 ⫾ 2.3 cm preoperation, 0.4 ⫾ 1.1 at 1 year postoperation, and 0.0 ⫾ 1.3 cm at 2 years postoperation. One-year but not 2-year postoperative values were statistically different from the preoperative values (P ⬍ 0.05). CVA-G at right initial foot contact had a mean preoperative value of 2.3 ⫾ 2.4 cm, decreasing to 1.4 ⫾ 1.4 cm at 1 year postoperation, and decreasing slightly again to 1.3 ⫾ 1.3 cm at 2 years postoperation. The differences between preoperative and 1 year (P ⫽ 0.028) and between the preoperative and 2-year postoperative values (P ⫽ 0.009) were both statistically significant. Interestingly, right CVA-G was significantly improved while left CVA-G was not. This may reflect the fact that all patients included in this series have a curvature that was convex to the right (right thoracic). There is no statistical difference between the SVA-P preoperation at 1 year and 2 years postoperation (P ⬎ 0.05). Similarly, for both SVA-G at right and left initial foot contact, there were no significant differences between the preoperative, 1-year, and 2-year postoperative values (P ⬎ 0.05) (Table 6). In the transverse plane APA-G Min showed no significant differences between preoperative and 2-year postoperative values (P ⬎ 0.05). The APA was also calculated at maximum shoulder rotation (APA-G Max) with a preoperative mean of ⫺12.5 ⫾ 5.1°, decreasing to ⫺8.9 ⫾ 5.1 at 1 year postoperation, and decreasing slightly to ⫺7.1 ⫾ 3.6 at 2 years postoperation. The preoperative APA at maximum shoulder rotation was statistically significant at the 1-year and 2-year postoperative values (P ⫽ 0.002 and 0.001, respectively). Lastly, the APA-G ROM preoperative showed a mean value of 14.2 ⫾ 4.1°, decreasing to 10.2 ⫾ 2.9° at 1 year postoperation and decreasing slightly again to 9.3 ⫾ 3.5° at 2 years postoperation. The values between the preoperative and 1-year preoperative and 2-year postoperative measurements were both highly statistically significant (P ⬍ 0.0001). In summary, the SVA-G did not show any differences between preoperative and postoperative values during both right and left initial foot contact. However, the CVA-G at right initial foot contact showed statistically significant differences with improved coronal balance at both 1 year and 2-year postoperation compared with preoperation. Interestingly, at left initial foot contact, these changes were not statistically significant. Finally, using the transverse position of the shoulders relative to the pelvis as determined by the APA-G, highly significant changes were noted between preoperative and postoperative values. Specifically, there was an improved angular symmetry found in the maximum transverse angles (APA-G Max) during the gait cycle. We believe this indicated an improvement occurred because the shoulders rotated toward a more equal angular alignment with the pelvis postoperative. However, the decline in transverse plane range of motion (APA-G ROM) could be a limitation of the surgery. There was no correlation between the radiographic data and the postural data (CVA-P and SVA-P) as well as the gait data (CVA-G and SVA-G) for the coronal or sagittal plane alignment assessment. However, a positive correlation was seen between the sagittal standing postural data (SVA-P) to the gait data (SVA-G) (P ⬍ 0.05). In addition, there was an extremely strong correlation between the coronal standing postural data (CVA-P) and the gait data (CVA-G) (P ⬍ 0.01). This strong correlation was also noted between the right (CVA-G and SVA-G) and left (CVA-G and SVA-G) initial foot contact coronal and sagittal plane gait data (P ⬍ 0.01). Discussion Table 4. Linear Gait Data (n ⴝ 30) Speed (cm/s) Cadence (steps/min) Stride length (cm) Stride width (cm) Preop 12 Month Postop 24 Month Postop 129 ⫾ 16 120 ⫾ 8 128 ⫾ 11 8.1 ⫾ 3.0 120 ⫾ 16* 115 ⫾ 8* 125 ⫾ 11 7.2 ⫾ 2.7* 119 ⫾ 16* 114 ⫾ 9* 124 ⫾ 12* 7.4 ⫾ 2.2 * Significantly different from preop (P ⬍ 0.05). Standing long cassette coronal and sagittal radiographs are the standard means of preoperative analysis and postoperative assessment of the surgical results of scoliosis fusions.12 Although changes in the static alignment of the spinal column, as is assessed by standard radiographs, do quantify changes in structure, they do not address concomitant changes in patient function that may occur. Currently, a paucity of data exists quantify- Adolescent Idiopathic Scoliosis • Lenke et al E335 Table 5. Kinematic Gait Data (n ⴝ 10) Sagittal knee angle at initial contact Sagittal ankle angle at initial contact Coronal ankle angle at initial contact Transverse foot angle at initial contact Preop Degrees 12 Month Postop Degrees 24 Month Postop Degrees 3.3 ⫾ 1.0 ⫺3.0 ⫾ 3.2 3.1 ⫾ 2.4 ⫺11.7 ⫾ 4.6 3.8 ⫾ 1.1 ⫺2.0 ⫾ 3.2 3.4 ⫾ 2.9 ⫺10.8 ⫾ 4.5 2.8 ⫾ 0.9 ⫺3.0 ⫾ 2.3 3.1 ⫾ 2.8 ⫺11.5 ⫾ 2.7 ing structural and functional changes in patients undergoing scoliosis fusion surgery.11 This study attempts to address how scoliosis fusion affects gait and spinal– pelvic balance alignment parameters after surgery using optoelectronic techniques. There have been a few studies analyzing various gait parameters in scoliotic patients. Thurston and Harris were the first to analyze normal kinematics of the lumbar spine and pelvis using gait analysis in 48 male volunteers ranging in age from 16 to 74 years.23 They presented results for both range of motion in each plane as well as a wave pattern for movement in each plane over time. Analysis of the wave pattern showed that movements of the pelvis and lumbar spine relate directly to identifiable events in the gait cycle. Giakas et al analyzed gait patterns between healthy and patients with scoliosis via analysis of ground reaction forces.9 They concluded that patients with scoliosis exhibited balance problems during the stance phase of gait and have significant asymmetry in the frequency characteristics of gait. No kinematics were reported. Hopf et al studied idiopathic scoliosis patients before and after surgery to evaluate muscle activation patterns.14 They analyzed 23 patients before and after posterior instrumentation and fusion with truncal and lower extremity EMGs. The results supported the hypothesis that activity asymmetries observed in the paravertebral musculature in patients with idiopathic scoliosis are the result of the scoliotic body deformities with consequent asymmetries in the biomechanical force patterns of body Table 6. Spinal-Pelvic Balance Parameters CVA-R (cm) SVA-R (cm) CVA-P (cm) SVA-P (cm) Right CVA-G (cm) Left CVA-G (cm) Right SVA-G (cm) Left SVA-G (cm) APA-G Max (degrees) APA-G Min (degrees) APA-G ROM (degrees) Preop 1 Yr Postop 2 Yr Postop 0.8 ⫾ 2.4 ⫺3.52 ⫾ 3.8 ⫺0.5 ⫾ 2.3 4.6 ⫾ 2.6 2.3 ⫾ 2.4 ⫺0.0 ⫾ 1.9 6.1 ⫾ 2.6 6.2 ⫾ 2.6 12.5 ⫾ 5.1 ⫺1.7 ⫾ 4.5 14.2 ⫾ 4.1 0.0 ⫾ 1.8 ⫺3.15 ⫾ 3.0 0.4 ⫾ 1.1* 4.5 ⫾ 2.4 1.4 ⫾ 1.4* ⫺0.3 ⫾ 1.3 5.8 ⫾ 2.6 5.8 ⫾ 2.6 8.9 ⫾ 5.1* ⫺1.3 ⫾ 4.6 10.2 ⫾ 2.9* ⫺0.2 ⫾ 1.3* ⫺3.0 ⫾ 2.9 0.0 ⫾ 1.3 5.0 ⫾ 2.6 1.3 ⫾ 1.3* ⫺0.4 ⫾ 1.3 6.4 ⫾ 3.1† 6.4 ⫾ 2.8† 7.1 ⫾ 3.6* † ⫺2.2 ⫾ 4.3 9.3 ⫾ 3.5* * Significantly different from preop (P ⬍ 0.05). † Significantly different from 12 month postop (P ⬍ 0.05). CVA-R ⫽ coronal vertical alignment–radiographic; SVA-R ⫽ sagittal vertical alignment–radiographic; CVA-P ⫽ coronal vertical alignment–postural; SVAP ⫽ sagittal vertical alignment–postural; CVA-G ⫽ coronal vertical alignment– gait; SVA-C ⫽ sagittal vertical alignment– gait; APA-G ⫽ acromion-pelvic angle– gait; APA-G MAX ⫽ acromion-pelvic angle– gait maximum; APA-G MIN ⫽ acromion-pelvic angle– gait minimum; APA-G ROM ⫽ acromion-pelvic angle– gait range of motion. postures and body motions. Schizas et al evaluated gait asymmetries in patients with idiopathic scoliosis using vertical force measurements only.22 They utilized published gait data in normal subjects as a control group.8,15 In 20 of the 21 subjects, an asymmetry of at least one gait parameter was noted. However, multiple regression analysis showed that no association between the gait asymmetry and curve direction, curve magnitude, or vertebral rotation existed. There are several studies that have assessed the gait patterns of patients after Harrington rod instrumentation and fusion.2– 4,6,10,16,25 These primarily investigated the negative aspects of Harrington distraction instrumentation into the lumbar spine with subsequent loss of lumbar lordosis and a creation of a flatback syndrome.6,26 This syndrome is associated with straightening of the lumbar spine, forward sagittal imbalance of the entire spine on the pelvis, and flexion of the knees as a compensatory mechanism to maintain as upright sagittal posture as possible. However, no study to date has analyzed these patients objectively using gait and motion analysis evaluations, although we have initiated this project and are currently collecting patient data. Certainly, prevention of these flatback syndrome deformities is currently the best means of treatment. Results from our study demonstrate that lower extremity kinematics during gait are not changed 2 years after spinal fusion surgery in patients with idiopathic scoliosis. However, we did see a decrease in gait speed secondary to a slower cadence and decreased stride length in these patients after the fusion surgery. It is unclear whether this is due to restriction from the spinal fusion placed on the trunk and torso or a possible deconditioning effect in these patients who potentially are less active after their spinal fusion surgery than before it. The body is not asked to perform at its maximum limits during gait and the surgical changes may have reduced the limits, but not to gait limit thresholds. Masso and Gorton were the first to evaluate standing postural changes in scoliosis patients using optoelectronic measurements.20 They evaluated 33 subjects with idiopathic scoliosis at approximately 13 months after posterior instrumentation and fusion. They evaluated these patients both in the coronal and transverse planes and found consistent improvement in both planes. They concluded that it was a useful, noninvasive technique to evaluate patients with scoliosis and their deformity be- E336 Spine • Volume 26 • Number 14 • 2001 fore and after surgery. Sagittal plane evaluation was not included, nor were any gait data. In this study sagittal spinal–pelvic parameters did not show any differences after surgery on the radiographic, standing postural, and gait analysis. This may be related to the fact that none of these patients had a significant preoperative sagittal malalignment to begin with, as is the typical situation for patients with AIS treated with segmental spinal instrumentation techniques.17 This is not the case in patients treated with previous instrumentation systems (i.e., Harrington rods) that tended to remove lumbar lordosis and create long-term sagittal plane problems.6,16 The most significant changes (improvements) were seen in the transverse plane where postoperative results demonstrated improved angular symmetry of the shoulders with respect to the pelvis at initial contact. It was interesting that no correlation existed between the radiographic CVA and SVA variables and the postural and gait variables. This may reflect that the changes before and after surgery being evaluated were small (⬍1 cm) in both planes. This could also be a function of instructions and also the other differences that may occur in standing posture when one is obtaining a standing radiograph versus standing for a video electronic evaluation, as well as the position of the surface markers. Vedantam et al have studied the importance of arm alignment and positioning in obtaining a standing long cassette spinal radiograph to evaluate sagittal balance.24 They found that positioning the arms at 30° was best compared with 90° to the vertical. The most significant difference was the increased negative sagittal vertical line alignment as the arms were elevated from 30° to 90°. Certainly, several limitations are present in this study, probably the most significant being the lack of normal controls. We elected to use the patient’s preoperative data as their own control because our goal was to assess the effects of spinal fusion surgery on these patients’ gait and posture. There are several articles that have looked at various gait parameters between healthy and scoliotic patients previously.8,9 In addition, we would assume that postural data for normal controls would be symmetric in the coronal plane because of the lack of any deformity. In the sagittal plane previous radiographic alignment parameters have shown similar sagittal balance values for normal controls as was seen in the current study of scoliotic patients.24 In addition, there are certainly many other confounding variables and technical procedures performed on this study group that we were unable to separate out from the scoliosis fusion surgery itself. These included any effect that iliac crest bone grafting would have on the pelvic parameters and separating the effects that a formal open thoracotomy would have on those patients that had an anterior spinal fusion. Also included was the effect that prior bracing had on patients who subsequently underwent a spinal fusion. Lastly, there are several variables that we attempted to analyze, such as differences in the lowest instrumented vertebrae, and also the total number of fusion levels. However, in our current study group there are not enough patients in each one of these separate groups to make a viable statistical comparison. In addition, some of the measured differences are small in magnitude, which makes clinical interpretation of these differences to be analyzed with caution. In conclusion, patients with AIS undergoing a spinal fusion show slight decreased gait speed at 2 years postoperation without any change in lower extremity kinematics. This decrease is due to a combination of a slower cadence as well as a diminished stride length. Spinal and pelvic balance parameters are improved in the coronal plane radiographically, and also during standing posture and gait, whereas sagittal balance parameters are unchanged radiographically and during standing posture and gait. Lastly, significant improvements were seen in the transverse plane where improved angular symmetry of the shoulders with respect to the pelvis occurred during gait at the expense of a loss of transverse plane range of motion. This type of analysis will further our understanding of the functional ramifications of performing spinal fusion on these patients. Additional functional analysis testing of trunk range of motion, trunk and abdominal strength, as well as aerobic analysis will also complement the radiographic assessment to form a more complete functional evaluation of these patients, providing important information for the patient, family, and treating surgeon. Key Points ● Lower extremity kinematics are unaffected by spinal fusion in patients with adolescent idiopathic scoliosis. ● Gait speed is decreased at 2 years postoperation as a result of significantly reduced cadence and stride length. ● Coronal spinal balance parameters are improved radiographically and during standing posture and gait. ● Sagittal balance parameters are unchanged on radiographs and during standing posture and gait. ● Transverse plane parameters demonstrated improved angular symmetry during gait. References 1. Aaro S, Dahlborn M. The effect of Harrington instrumentation on the longitudinal axis rotation of the apical vertebra and on the spinal and rib-cage deformity in idiopathic scoliosis studied by computer tomography. Spine 1982;7:456 – 62. 2. Aaro S, Ohlen G. The effect of Harrington instrumentation on the sagittal configuration and mobility of the spine in scoliosis. Spine 1983;8:570 –5. 3. Balderston RA, Winter RB, Moe JH, Bradford DS, Lonstein JE. Fusion to the sacrum for non-paralytic scoliosis in the adult. Spine 1986;11:824 –9. 4. Boachie-Adjei O, Dendrinos GK, Ogilvie JW, et al. Fusion to the sacrum for non-paralytic scoliosis in the adult. J Spinal Disord 1991;4:131– 41. 5. Burwell RG, Cole AA, Cook TA, et al. Pathogenesis of idiopathic scoliosis: the Nottingham concept. Acta Orthop Belg 1992;58:33–58. 6. Cochran T, Irstam L, Nachemson A. Long-term anatomic and functional Adolescent Idiopathic Scoliosis • Lenke et al E337 changes in patients with adolescent idiopathic scoliosis treated by Harrington rod fusion. Spine 1983;8:576 – 84. 7. Engsberg JR, Andrews JG. An analysis of the combined talocalcaneal/ talocrural joint during running support. Med Sci Sports Exerc 1987;19:275– 84. 8. Engsberg JR, Lee AG, Tedford KG, et al. Normative ground reaction force data for walking. J Pediatr Orthop 1993;13:167–73. 9. Giakas G, Baltzopoulos V, Dangerfield PH, et al. Comparison of gait patterns between healthy and scoliotic patients using time and frequency domain analysis of ground reaction forces. Spine 1996;21:2235– 42. 10. Ginsburg HH, Goldstein L, Haake PW, et al. Longitudinal study of back pain in postoperative idiopathic scoliosis: long-term follow-up, phase IV. Presented at the Scoliosis Research Society 30th Annual Meeting, Asheville, NC, 1995. 11. Gracovetsky S. A hypothesis for the role of the spine in human locomotion: a challenge to current thinking. J Biomed Eng 1985;7:205–16. 12. Haher TR, Merola A, Zipnick RI, et al. Meta-analysis of surgical outcome in adolescent idiopathic scoliosis: a 35-year English literature review of 11,000 patients. Spine 1995;20:1575– 84. 13. Hamill C, Lenke LG, Bridwell KH, et al. The use of pedicle screws to improve correction in the lumbar spine of adolescent idiopathic scoliosis: is it warranted? Spine 1996;21:1241–9. 14. Hopf C, Scheidecker M, Steffan K, et al. Gait analysis in idiopathic scoliosis before and after surgery: a comparison of pre- and postoperative muscle activation pattern. Eur Spine J 1998;7:6 –11. 15. Khoo BC, Goh JC, Bose K. A biomechanical model to determine lumbosacral loads during single stance phase in normal gait. Med Eng Phys 1995;17:27– 35. 16. LaGrone MO, Bradford DS, Moe JH, et al. Treatment of symptomatic flatback after spinal fusion. J Bone Joint Surg [Am] 1988;70:569 – 80. 17. Lenke LG, Bridwell KH, Baldus C, et al. Cotrel-Dubousset instrumentation for adolescent idiopathic scoliosis. J Bone Joint Surg [Am] 1992;74:1056 – 67. 18. Lenke LG, Bridwell KH, Baldus C, et al. Ability of Cotrel-Dubousset instrumentation to preserve distal lumbar motion segments in adolescent idiopathic scoliosis. J Spinal Disord 1993;6:339 –50. 19. Lenke LG, Engsberg JR, Olree KS, et al. Trunk range of motion and gait analysis following adolescent idiopathic scoliosis fusion. Poster presentation at the Scoliosis Research Society Annual Meeting, St. Louis, MO, September 1997. 20. Masso PD, Gorton GE III. Quantifying changes in standing body segment alignment following spinal instrumentation and fusion in idiopathic scoliosis using an optoelectronic measurement system. Spine 2000;25:457– 62. 21. Richards BS, Birch JG, Hering JA, et al. Frontal plane and sagittal plane balance following Cotrel-Dubousset instrumentation for idiopathic scoliosis. Spine 1989;17:733–7. 22. Schizas CG, Kramers-de Quervain IA, Stussi E, et al. Gait asymmetries in patients with idiopathic scoliosis using vertical forces measurement only. Eur Spine J 1998;7:95– 8. 23. Thurston AJ, Harris JD. Normal kinematics of the lumbar spine and pelvis. Spine 1983;8:199 –205. 24. Vendantam R, Lenke LG, Keeney JA, et al. Comparison of standing sagittal spinal alignment in asymptomatic adolescents and adults. Spine 1998;23: 211–5. 25. Vendantam R, Lenke LG, Linville D, et al. The effect on variation in arm position on sagittal spinal alignment. Spine 2000;25:2204 –9. 26. Winter RB, Lonstein JE. Adult idiopathic scoliosis treated with Luque or Harrington rods and sublaminar wiring. J Bone Joint Surg Am 1989;71: 1308 –13. Address reprint requests to Lawrence G. Lenke, MD Department of Orthopaedic Surgery Washington University School of Medicine One Barnes-Jewish Plaza, Suite 11300 St. Louis, MO 63110 E-mail: lenkel@msnotes.wustl.edu
0
You can add this document to your study collection(s)
Sign in Available only to authorized usersYou can add this document to your saved list
Sign in Available only to authorized users(For complaints, use another form )