SPINE Volume 39, Number 24, pp E1481-E1487 ©2014, Lippincott Williams & Wilkins
CASE REPORT
Locomotor Biomechanics After Total Sacrectomy A Case Report Jo Armour Smith, PhD, PT,* Alexander Tuchman, MD,† Michael Huoh, MD,‡ Andreas M. Kaiser, MD,† Wesley G. Schooler, MD,† and Patrick C. Hsieh, MD†
Study Design. Biomechanical analysis of locomotion after total sacrectomy in a single patient case. Objective. To describe the biomechanics of locomotion after successful total sacrectomy and spinopelvic reconstruction. Summary of Background Data. Total sacrectomy is a complex surgery that has significant consequences for mobility after surgery due to loss of lower lumbar and sacral innervation to the lower extremities, and the anatomic dissociation of the spine from the pelvis. There is no existing literature quantifying locomotor biomechanics after total sacrectomy. Methods. A 22-year-old female with a sacral osteosarcoma underwent an en bloc sacrectomy with L3 to pelvis instrumented fusion. Neuromuscular function was tested 1 year after surgery using monopolar needle electromyography. Three-dimensional motion capture and surface electromyography were used to quantify spatiotemporal characteristics of locomotion and lower extremity kinematics, kinetics, and muscle function during locomotion at 6 months and 1 year after surgery. Results. Electrodiagnostic testing suggested partial preservation and reinnervation of S1 nerve root function on the right, resulting in greater than expected activity in the hamstrings, gluteus maximus, and triceps surae postsurgically. Unexpectedly on the left, there was residual activity in the hamstrings, despite the loss of sacral innervation and the sciatic nerve. At 1 year after surgery, the patient was able to walk independently. Kinematic and kinetic impairments and compensations were most evident in the sagittal and coronal planes.
From the *Division of Biokinesiology and Physical Therapy, and †Keck School of Medicine of USC, University of Southern California, Los Angeles, CA; and ‡Department of Physical Medicine and Rehabilitation, Kaiser Permanente Downey Medical Center, Downey, CA. Acknowledgment date: February 27, 2014. First revision date: June 9, 2014. Second revision date: July 12, 2014. Acceptance date: July 16, 2014. The manuscript submitted does not contain information about medical device(s)/drug(s). No funds were received in support of this work. Relevant financial activities outside the submitted work: consultancy, grants, employment, payment for lecture, royalties. Address correspondence and reprint requests to Jo Armour Smith, PhD, PT, Division of Biokinesiology and Physical Therapy, University of Southern California, 1540 East Alcazar St, CHP-155, Los Angeles, CA 90089; E-mail: [email protected] DOI: 10.1097/BRS.0000000000000594 Spine
Conclusion. Excellent locomotor outcomes are possible after total sacrectomy. Key words: sacrectomy, spinopelvic reconstruction, locomotion, kinetics, kinematics, electromyography, biomechanics, hip, knee, trunk. Level of Evidence: N/A Spine 2014;39:E1481–E1487
P
rimary tumors of the sacrum such as chordoma, chondrosarcoma, and osteogenic sarcoma are rare tumors that are locally invasive and can metastasize. Optimal management of these tumors is challenging due to the poor response to traditional chemotherapy and radiation. With evolution of surgical techniques and technology in spine surgery during recent decades, en bloc excision of these tumors with negative margins is now considered to be the standard treatment. Total en bloc sacrectomy is a technically difficult surgical procedure due to the complex anatomy and function of the sacrum. The sacrum is the cornerstone in the connection of the spine to the pelvis and lower extremities. With the excision of the tumor by total en bloc sacrectomy, the normal anatomical connection from the spinal column to the pelvis is disrupted. This anatomical connection must be reconstructed and stabilized via instrumented fusion to restore the ability for weight bearing on the legs. Patient mobility in the longterm is significantly affected by the necessary resection of the sacral and occasionally lower lumbar nerve roots to achieve negative tumor margins for optimal oncological outcomes. Although the oncological and functional outcomes after sacrectomy with spinopelvic fusions have been reported by various authors, due to the rarity of malignant primary sacral tumors, postsurgical locomotor biomechanics have not been described.1–5 The purpose of this study was to quantify neuromuscular function and locomotor biomechanics in a young patient after successful total en bloc sacrectomy for excision of an osteogenic sarcoma with spinopelvic reconstruction.
MATERIALS AND METHODS History A 22-year-old female presented with progressive low back pain and left lower extremity pain and weakness. She had a history www.spinejournal.com
E1481
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited. SPINE140307_LR E1481
22/10/14 10:53 AM
CASE REPORT of a pelvic yolk sac tumor successfully treated with surgical resection, chemotherapy, and radiation therapy in early childhood. She had lived free of tumor and with full function in the intervening years. Magnetic resonance imaging of the spine and pelvis demonstrated a large contrast-enhancing sacral lesion with involvement of the left ilium and sciatic notch (Figure 1A, B). A percutaneous biopsy of the lesion confirmed the diagnosis of osteogenic sarcoma involving the sacrum from the level of S1 down to the sciatic notch with involvement of the left sacroiliac joint and ilium. After undergoing neoadjuvant chemotherapy, the patient underwent a total en bloc sacrectomy for excision of the osteogenic sarcoma.
Surgical Procedure The patient underwent a 2-staged procedure with anterior and posterior approach. An anterior approach was first employed to perform rectum excision with colostomy as a result of tumor involvement to the posterior rectum. In addition, anterior dissection of visceral and neural structures, ligation of the internal iliac vessels, and anterior L5–S1 discectomy
Locomotion After Sacrectomy • Smith et al
was performed to mobilize the normal tissues away from the tumor and delineate the anterior margins. An L4–L5 anterior discectomy with interbody fusion was also performed. In addition, we prepared the initial stages of vascularized flap reconstruction of the soft-tissue defect by harvesting an omental flap. A myocutaneous flap was prepared by mobilizing the left rectus abdominus inferiorly. Both vascularized grafts were rotated inferiorly and tucked into the pelvis anterior to the tumor with a Silastic sheet barrier (Dow Corning, Midland, MI). The anterior wound was closed and the patient was taken to intensive care unit for recovery. The posterior approach was performed with exposure of the lumbar spine, the sacrum, and the ilium to delineate the posterior and lateral tumor margins. We aimed for a wide margin resection to avoid violating the tumor capsule. The thecal sac was exposed with an L5 laminectomy and ligated below the L5 nerve roots to delineate the superior margin. On the right side, we performed an osteotomy from the top of the sacrum through the sacroiliac joint laterally and follow it inferiorly toward the sciatic notch. On the left side, an osteotomy was performed through the left ileum, lateral to the tumor and the left sacroiliac joint, from the top of the iliac crest to the sciatic notch. En bloc excision of the tumor with total sacrectomy was achieved with completion of L5–S1 discectomy, excision of the sacrospinous and sacrotuberous ligaments along with detachment of the pelvic muscles from the distal sacrum and coccyx. In addition, the left sciatic nerve was involved by the tumor and it was excised with the tumor to achieve negative margins. An L3 to pelvis instrumented fusion was then performed using segmental instrumentation with dual iliac screw fixation bilaterally and 4-rod construct across the lumbopelvic junction. In addition to the spinopelvic instrumented fusion, final soft-tissue reconstruction was performed with the omentum flap and the myocutaneous flap using the left rectus abdominus muscle by plastic surgery.
Postsurgical Course The patient tolerated the procedure without any complication and remained in intensive care unit with bed rest on an air-fluidized bed for her initial postsurgical recovery to protect her flap reconstruction of the wound (Clinitron Air Fluidized Therapy Bed; Hill-Rom, Batesville, IN) and remained on the same bed on transfer to the floor. She began gradual mobilization with full weight-bearing status and physical therapy 3 weeks after her surgery. Then, she was transferred to an inpatient rehabilitation facility where she underwent an intensive rehabilitation program.
Postsurgical Evaluation
Figure 1. Preoperative MRI showing axial (A) and coronal (B) views of an osteosarcoma arising from the sacrum and invading the left ilium. MRI indicates magnetic resonance imaging.
E1482
www.spinejournal.com
The patient was followed clinically with serial neurological examinations and radiographical studies at 3, 6, 9, 12, and 24 months postsurgically (Figure 2A, B). Monopolar needle electromyography (EMG) evaluation was conducted at approximately 1 year after surgery. She was also referred for biomechanical gait analysis at 6 months and 1 year after surgery. Data were acquired during multiple trials as the patient walked at self-selected speed along a walkway with embedded November 2014
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited. SPINE140307_LR E1482
22/10/14 10:53 AM
CASE REPORT
Locomotion After Sacrectomy • Smith et al
RESULTS
Figure 2. Postoperative standing AP (A) and lateral (B) 36-in. longcassette radiographs after en bloc sacrectomy followed by L3 to pelvis instrumented fusion (images from 2.5 yr after surgery).
force plates (AMTI, Watertown, MA). The patient wore a rigid lumbosacral orthosis, an ankle-foot orthosis (AFO) on the left lower limb, and used a single-point cane in the right hand during the gait trials. Three-dimensional motion capture data were collected using an 11-camera system (Qualisys AB, Gothenburg, Sweden). Optoreflective markers were placed bilaterally to define the anatomical landmarks of the upper trunk, pelvis, and lower limbs and rigid arrays of markers were used to track segment motion. Lower limb kinetics (presented as internal muscle moments) were calculated from the kinematics and force plate data using an inverse dynamics model (Visual 3D software; C-Motion Inc., Germantown, MD). Kinesiological EMG data were collected from bipolar surface electrodes placed on the medial gastrocnemius, biceps femoris long/short head, and vastus lateralis on the right, and on the biceps femoris and vastus lateralis on the left (Motion Lab Systems Inc., Baton Rouge, LA). Gait events were defined using the vertical trajectories of the heel markers. Patient findings were compared with data from 4 healthy females (mean age, 23.5 yr) from a pre-existing laboratory database, and with previously published norms.6 Spine
It has been more than 4½ years since the patients underwent surgery. She has undergone routine follow-up positron emission tomography-computed tomography every 6 months since her surgery demonstrating no evidence of recurrent tumor. At her 1-year clinical evaluation she reported being free of pain on a regimen of gabapentin and occasional oxycodone. At more than 4½ years from surgery, she has minimal pain symptoms and is only rarely taking hydrocodone-acetaminophen to manage her symptoms. She is managing her bladder and bowel functions with urinary catheterization and a colostomy. Sexual intercourse is not possible due sensory deficits and pelvic scarring. She reports anesthesia in the perineal region and sacral dermatomes. However, since her surgery, she has remained free of tumor and her latest computed tomographic scan demonstrates stable spinopelvic instrumentation without evidence of instability (Figure 3A, B). Results from the 1-year monopolar EMG evaluation are summarized in Table 1. Motor unit activity was normal in muscles innervated by L2 through L5 on the right and by L2 through L4 on the left, with evidence of some reinnervation in the biceps femoris and medial gastrocnemius on the right. By 1 year after surgery, the patient walked independently for up to 20 minutes on level surfaces with a cane. She was also able to negotiate a 46% ramp and climb a single flight of stairs without assistance. At 6 months, her gait speed was an average of 0.64 m/s, and this had improved to 0.71 m/s, or 55% of normal gait speed for females, at 1 year after surgery.6 On average, the patient’s stride length was 19% less than that of healthy females, but improved from the 6-month to 1-year evaluation (1.00 m and 1.04 m, respectively).6 Step width was significantly decreased bilaterally, more profoundly on the left than on the right (−0.03 m and 0.04 m, respectively at 1 yr).6 In the sagittal plane, trunk motion was notable for rapid extension immediately after initial contact on the left (peak trunk extension during left loading response/midstance, 8.7° and that during right loading response/midstance, 4.2°). Selected time-series kinematics and kinetics of the lower limbs from the 1-year evaluation, with comparison with healthy females for reference, are presented in Figure 4. In particular, on the right, there was increased ankle dorsiflexion during loading response and midstance, increased knee flexion at initial contact and throughout stance phase, and increased stance phase hip abduction. On the left, there was knee hyperextension at initial contact and increased knee abduction and hip adduction throughout stance phase (Figure 4A). At 6 months, peak hip internal rotation during loading response on the left was 9.3° compared with 1.8° in the healthy subjects. However, at 1 year after surgery, peak internal rotation was normal while external rotation during midstance and terminal stance was increased. Stance phase kinetics (Figure 4B) were characterized by bilaterally decreased plantar flexor moments, increased knee extensor moments, and a reduction in the hip abductor moment on the left. Kinesiological EMG data from both evaluations indicated anomalous peaks of hamstring activity during preswing on the right and midswing on the left. Quadriceps activity was www.spinejournal.com
E1483
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited. SPINE140307_LR E1483
22/10/14 10:53 AM
CASE REPORT
Locomotion After Sacrectomy • Smith et al
Figure 3. CT follow-up imaging at 4 years after surgery; sagittal (A) and coronal (B). CT indicates computed tomography.
abnormally prolonged during stance on the right but this was less evident on the left. Activity in the right medial plantar flexors peaked during loading response and terminal stance.
DISCUSSION Total sacrectomy with sacral nerve root sacrifice and spinopelvic reconstruction is a rarely performed procedure that has significant consequences for locomotion. Mobility after total sacrectomy varies widely, ranging from independent walking with AFOs to wheelchair dependence.1–5,7,8 The variation in reported outcomes may in part be attributable to high rates of postoperative complications including nonunion and instrument failure. It has been suggested that preservation of the lumbar nerve roots permits independent locomotion after sacrectomy. This is due to maintenance of L5 innervation to the deep ankle plantar flexors and compensatory hypertrophy of the 2-joint hip extensors in response to impaired gluteus maximus E1484
www.spinejournal.com
function.7,9 In our patient, electrodiagnostic testing at 1 year after surgery indicated normal motor unit activity in muscles innervated by the L2, L3, L4, and L5 nerve roots on the right. In addition, both semimembranosus and gluteus maximus on the right had normal motor unit action potentials on electrodiagnostic evaluation, while the polyphasic motor unit potentials in biceps femoris and the medial gastrocnemius were suggestive of some reinnervation. These findings may be due to some partial sparing of S1 as a result of anomalous nerve root anatomy in this patient. On the left there was normal function in muscles innervated by the L2, L3, and L4 nerve roots. Motor unit action potentials were absent in muscles primarily innervated by S1 and S2 and those supplied by the sciatic nerve and its branches. Interestingly however, some semimembranosus function also seemed to be preserved on the left. At a relatively early stage postoperatively, this patient was able to walk independently for short distances with minimal November 2014
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited. SPINE140307_LR E1484
22/10/14 10:53 AM
CASE REPORT
Locomotion After Sacrectomy • Smith et al
TABLE 1. Electrodiagnostic EMG Evaluation at 1 yr After Surgery Right
Left
Tibialis anterior Insertional/spontaneous activity
Normal
Positive sharp waves
Voluntary activity
Normal
No motor unit activity
Insertional/spontaneous activity
Normal
Positive sharp waves
Voluntary activity
Normal
No motor unit activity
Insertional/spontaneous activity
Positive sharp waves
Fibrillations/positive sharp waves
Voluntary activity
Small amplitude, polyphasic MUAPs
No motor unit activity
Insertional/spontaneous activity
Normal
Normal
Voluntary activity
Normal
Normal
Insertional/spontaneous activity
Positive sharp waves
Positive sharp waves
Voluntary activity
One 9-phase polyphasic MUAP
No motor unit activity
Insertional/spontaneous activity
Normal
Normal
Voluntary activity
Normal
Normal
Insertional/spontaneous activity
Normal
Reduced insertional activity
Voluntary activity
Normal
No motor unit activity
Insertional/spontaneous activity
Reduced insertional activity
Reduced insertional activity
Voluntary activity
Small amplitude MUAPs
No motor unity activity
Tibialis posterior
Medial gastrocnemius
Vastus medialis
Biceps femoris
Semimembranosus
Gluteus maximus
Gluteus medius
MUAP indicates motor unit action potential; EMG, electromyography.
assistance from a cane. Her gait speed was reduced compared with that of healthy young females. Normally, the speed of forward progression during locomotion is maximized by propulsive forces during preswing phase, smooth forward motion of the body over a series of pivots or “rockers” during stance phase, and the forward momentum of the body.6 In this patient, all 3 of these factors were affected as a result of both denervation and gait compensations. First, as expected, the plantar flexor moment prior to swing phase was substantially decreased bilaterally in this patient compared with healthy subjects. Although passive tension in the triceps surae may contribute to push off at preswing in healthy subjects,6 unusual biceps femoris EMG activity at preswing on the right in this patient suggested that as a result of decreased active plantar flexion, a compensatory mechanism was required to drive the trailing limb into knee flexion. Second, the AFO reduced the normal ability of the left ankle, forefoot, and toes Spine
to act as pivots for the forward advancement of the center of mass. Finally, the abnormal trunk extension that we observed in this patient during left stance phase may have served to limit the hip extensor moment on the left side by moving the center of mass posterior to the hip joint, but also reduced forward momentum of the center of mass.6 Several gait characteristics after sacrectomy were related to impaired function of the hip musculature. It was anticipated that the loss of the eccentric action of gluteus maximus at initial contact would result in an increase in hip adduction and internal rotation on the left side during loading response. This was observed at the 6-month evaluation, but to a lesser extent at the 1-year evaluation. In addition, the unopposed action of the adductors on the left was likely responsible for the increased hip adduction throughout the stride cycle and decreased left step width in this patient. This was not so apparent on the right due to residual function www.spinejournal.com
E1485
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited. SPINE140307_LR E1485
22/10/14 10:53 AM
CASE REPORT
Locomotion After Sacrectomy • Smith et al
Figure 4. Selected time-series kinematics (A) and kinetics (B) of the ankle, knee, and hip compared with healthy subjects (n = 4). Positive values indicate (A) ankle dorsiflexion, knee flexion, and hip abduction joint angles and (B) ankle plantar flexor, knee extensor, and hip abductor moments. Data are time normalized to 100% stride cycle (kinematics) and stance phase (kinetics). Kinetic data are normalized to body weight.
in the hip abductors and the relatively normal hip abductor moment. However, despite impaired left hip abductor strength, a Trendelenburg gait pattern was not evident during left stance phase, likely due to the use of the cane in the contralateral hand. Partial (right) and total (left) loss of the innervation to triceps surae resulted in impaired deceleration of the tibia over the foot and increased ankle dorsiflexion and knee flexion during the stance phase of gait. This was more pronounced on the right than on the left, due to the stabilizing effect of the AFO. As a result, stance phase knee extensor moments, and duration of quadriceps EMG activity, were substantially increased on the right side compared with healthy subjects. On the left, a different compensation was observed, with knee hyperextension at initial contact and reduced flexion throughout stance phase helping to compensate for the absent plantar flexors and reduced quadriceps strength. As anticipated, hip extensor moments during loading response were reduced bilaterally compared with controls. Although we hypothesized that an increased hip flexor moment during pre- and initial-swing phases would be evident as a compensation for decreased push off at the ankle,10 this was not observed in this patient, potentially due to her reduced gait speed compared with the controls. E1486
www.spinejournal.com
It should be noted that although our data support the subjective and anecdotal locomotor outcomes that have been presented in earlier manuscripts describing patients with similar patterns of nerve root sacrifice after sacrectomy,7,9 and previously hypothesized patterns of compensation, it is possible that other individuals will display different patterns of kinetic and kinematic compensation after this surgery.
CONCLUSION In this patient, substantial independent locomotion was possible as early as 6 months after successful total sacrectomy and spinopelvic reconstruction, and the spatiotemporal and kinematic aspects of locomotion continued to improve at 1 year after surgery. Despite sacrifice of the sacral nerve roots a functional locomotor pattern may be achieved with kinematic and kinetic compensations.
➢ Key Points Locomotor biomechanics were assessed in a young patient after successful total en bloc sacrectomy and spinopelvic reconstruction with resection of the sacral nerve roots bilaterally. November 2014
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited. SPINE140307_LR E1486
22/10/14 10:53 AM
CASE REPORT The patient achieved a functional independent locomotor pattern relatively early after surgery, and the spatiotemporal characteristics of her gait pattern continued to improve at 1-year follow-up. Kinematic and kinetic compensations were evident in the sagittal and coronal planes in the trunk, hips, knees, and ankles.
References
1. Singh R, Custodio C, Lidestri P, et al. Functional outcome in a patient with radical sacrectomy and sacral cord resection: a case report. 2004 Annual Academy Assembly Abstracts. Arch Phys Med Rehabil 2004;85:E17–8. 2. Spiegel DA, Richardson WJ, Scully SP, et al. Long-term survival following total sacrectomy with reconstruction for the treatment of primary osteosarcoma of the sacrum. J Bone Jt Surg 1999;81-A:848–5.
Spine
Locomotion After Sacrectomy • Smith et al
3. Zileli M, Hoscoskun C, Brastianos P, et al. Surgical treatment of primary sacral tumors: complications associated with sacrectomy. Neurosurg Focus 2003;15:1–8. 4. Sahakitrungruang C, Chantra K, Dusitanond N, et al. Sacrectomy for primary sacral tumors. Dis Colon Rectum 2009;52:913–8. 5. Newman CB, Keshavarzi S, Aryan HE. En bloc sacrectomy and reconstruction: technique modification for pelvic fixation. Surg Neurol 2009;72:752–6. 6. Perry J, Burnfield J. eds. Gait Analysis: Normal and Pathological Function. 2nd ed. Thorofare, NJ: Slack Inc; 2010. 7. Ohata N, Ozaki T, Kunisada T, et al. Extended total sacrectomy and reconstruction for sacral tumor. Spine 2004;29;E123–6. 8. Santi MD, Mitsunaga MM, Lockett JL. Total sacrectomy for a giant sacral schwannoma: a case report. Clin Orthop Relat R 1993;294:285–9. 9. Wuisman P, Lieshout O, Sugihara S, et al. Total sacrectomy and reconstruction: oncologic and functional outcome. Clin Orthop Relat R 2000;381:192–203. 10. Lewis CL, Ferris DP. Walking with increased ankle push off decreases hip muscle moments. J Biomech 2008;41:2082–9.
www.spinejournal.com
E1487
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited. SPINE140307_LR E1487
22/10/14 10:53 AM
CASE REPORT
Locomotor Biomechanics After Total Sacrectomy A Case Report Jo Armour Smith, PhD, PT,* Alexander Tuchman, MD,† Michael Huoh, MD,‡ Andreas M. Kaiser, MD,† Wesley G. Schooler, MD,† and Patrick C. Hsieh, MD†
Study Design. Biomechanical analysis of locomotion after total sacrectomy in a single patient case. Objective. To describe the biomechanics of locomotion after successful total sacrectomy and spinopelvic reconstruction. Summary of Background Data. Total sacrectomy is a complex surgery that has significant consequences for mobility after surgery due to loss of lower lumbar and sacral innervation to the lower extremities, and the anatomic dissociation of the spine from the pelvis. There is no existing literature quantifying locomotor biomechanics after total sacrectomy. Methods. A 22-year-old female with a sacral osteosarcoma underwent an en bloc sacrectomy with L3 to pelvis instrumented fusion. Neuromuscular function was tested 1 year after surgery using monopolar needle electromyography. Three-dimensional motion capture and surface electromyography were used to quantify spatiotemporal characteristics of locomotion and lower extremity kinematics, kinetics, and muscle function during locomotion at 6 months and 1 year after surgery. Results. Electrodiagnostic testing suggested partial preservation and reinnervation of S1 nerve root function on the right, resulting in greater than expected activity in the hamstrings, gluteus maximus, and triceps surae postsurgically. Unexpectedly on the left, there was residual activity in the hamstrings, despite the loss of sacral innervation and the sciatic nerve. At 1 year after surgery, the patient was able to walk independently. Kinematic and kinetic impairments and compensations were most evident in the sagittal and coronal planes.
From the *Division of Biokinesiology and Physical Therapy, and †Keck School of Medicine of USC, University of Southern California, Los Angeles, CA; and ‡Department of Physical Medicine and Rehabilitation, Kaiser Permanente Downey Medical Center, Downey, CA. Acknowledgment date: February 27, 2014. First revision date: June 9, 2014. Second revision date: July 12, 2014. Acceptance date: July 16, 2014. The manuscript submitted does not contain information about medical device(s)/drug(s). No funds were received in support of this work. Relevant financial activities outside the submitted work: consultancy, grants, employment, payment for lecture, royalties. Address correspondence and reprint requests to Jo Armour Smith, PhD, PT, Division of Biokinesiology and Physical Therapy, University of Southern California, 1540 East Alcazar St, CHP-155, Los Angeles, CA 90089; E-mail: [email protected] DOI: 10.1097/BRS.0000000000000594 Spine
Conclusion. Excellent locomotor outcomes are possible after total sacrectomy. Key words: sacrectomy, spinopelvic reconstruction, locomotion, kinetics, kinematics, electromyography, biomechanics, hip, knee, trunk. Level of Evidence: N/A Spine 2014;39:E1481–E1487
P
rimary tumors of the sacrum such as chordoma, chondrosarcoma, and osteogenic sarcoma are rare tumors that are locally invasive and can metastasize. Optimal management of these tumors is challenging due to the poor response to traditional chemotherapy and radiation. With evolution of surgical techniques and technology in spine surgery during recent decades, en bloc excision of these tumors with negative margins is now considered to be the standard treatment. Total en bloc sacrectomy is a technically difficult surgical procedure due to the complex anatomy and function of the sacrum. The sacrum is the cornerstone in the connection of the spine to the pelvis and lower extremities. With the excision of the tumor by total en bloc sacrectomy, the normal anatomical connection from the spinal column to the pelvis is disrupted. This anatomical connection must be reconstructed and stabilized via instrumented fusion to restore the ability for weight bearing on the legs. Patient mobility in the longterm is significantly affected by the necessary resection of the sacral and occasionally lower lumbar nerve roots to achieve negative tumor margins for optimal oncological outcomes. Although the oncological and functional outcomes after sacrectomy with spinopelvic fusions have been reported by various authors, due to the rarity of malignant primary sacral tumors, postsurgical locomotor biomechanics have not been described.1–5 The purpose of this study was to quantify neuromuscular function and locomotor biomechanics in a young patient after successful total en bloc sacrectomy for excision of an osteogenic sarcoma with spinopelvic reconstruction.
MATERIALS AND METHODS History A 22-year-old female presented with progressive low back pain and left lower extremity pain and weakness. She had a history www.spinejournal.com
E1481
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited. SPINE140307_LR E1481
22/10/14 10:53 AM
CASE REPORT of a pelvic yolk sac tumor successfully treated with surgical resection, chemotherapy, and radiation therapy in early childhood. She had lived free of tumor and with full function in the intervening years. Magnetic resonance imaging of the spine and pelvis demonstrated a large contrast-enhancing sacral lesion with involvement of the left ilium and sciatic notch (Figure 1A, B). A percutaneous biopsy of the lesion confirmed the diagnosis of osteogenic sarcoma involving the sacrum from the level of S1 down to the sciatic notch with involvement of the left sacroiliac joint and ilium. After undergoing neoadjuvant chemotherapy, the patient underwent a total en bloc sacrectomy for excision of the osteogenic sarcoma.
Surgical Procedure The patient underwent a 2-staged procedure with anterior and posterior approach. An anterior approach was first employed to perform rectum excision with colostomy as a result of tumor involvement to the posterior rectum. In addition, anterior dissection of visceral and neural structures, ligation of the internal iliac vessels, and anterior L5–S1 discectomy
Locomotion After Sacrectomy • Smith et al
was performed to mobilize the normal tissues away from the tumor and delineate the anterior margins. An L4–L5 anterior discectomy with interbody fusion was also performed. In addition, we prepared the initial stages of vascularized flap reconstruction of the soft-tissue defect by harvesting an omental flap. A myocutaneous flap was prepared by mobilizing the left rectus abdominus inferiorly. Both vascularized grafts were rotated inferiorly and tucked into the pelvis anterior to the tumor with a Silastic sheet barrier (Dow Corning, Midland, MI). The anterior wound was closed and the patient was taken to intensive care unit for recovery. The posterior approach was performed with exposure of the lumbar spine, the sacrum, and the ilium to delineate the posterior and lateral tumor margins. We aimed for a wide margin resection to avoid violating the tumor capsule. The thecal sac was exposed with an L5 laminectomy and ligated below the L5 nerve roots to delineate the superior margin. On the right side, we performed an osteotomy from the top of the sacrum through the sacroiliac joint laterally and follow it inferiorly toward the sciatic notch. On the left side, an osteotomy was performed through the left ileum, lateral to the tumor and the left sacroiliac joint, from the top of the iliac crest to the sciatic notch. En bloc excision of the tumor with total sacrectomy was achieved with completion of L5–S1 discectomy, excision of the sacrospinous and sacrotuberous ligaments along with detachment of the pelvic muscles from the distal sacrum and coccyx. In addition, the left sciatic nerve was involved by the tumor and it was excised with the tumor to achieve negative margins. An L3 to pelvis instrumented fusion was then performed using segmental instrumentation with dual iliac screw fixation bilaterally and 4-rod construct across the lumbopelvic junction. In addition to the spinopelvic instrumented fusion, final soft-tissue reconstruction was performed with the omentum flap and the myocutaneous flap using the left rectus abdominus muscle by plastic surgery.
Postsurgical Course The patient tolerated the procedure without any complication and remained in intensive care unit with bed rest on an air-fluidized bed for her initial postsurgical recovery to protect her flap reconstruction of the wound (Clinitron Air Fluidized Therapy Bed; Hill-Rom, Batesville, IN) and remained on the same bed on transfer to the floor. She began gradual mobilization with full weight-bearing status and physical therapy 3 weeks after her surgery. Then, she was transferred to an inpatient rehabilitation facility where she underwent an intensive rehabilitation program.
Postsurgical Evaluation
Figure 1. Preoperative MRI showing axial (A) and coronal (B) views of an osteosarcoma arising from the sacrum and invading the left ilium. MRI indicates magnetic resonance imaging.
E1482
www.spinejournal.com
The patient was followed clinically with serial neurological examinations and radiographical studies at 3, 6, 9, 12, and 24 months postsurgically (Figure 2A, B). Monopolar needle electromyography (EMG) evaluation was conducted at approximately 1 year after surgery. She was also referred for biomechanical gait analysis at 6 months and 1 year after surgery. Data were acquired during multiple trials as the patient walked at self-selected speed along a walkway with embedded November 2014
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited. SPINE140307_LR E1482
22/10/14 10:53 AM
CASE REPORT
Locomotion After Sacrectomy • Smith et al
RESULTS
Figure 2. Postoperative standing AP (A) and lateral (B) 36-in. longcassette radiographs after en bloc sacrectomy followed by L3 to pelvis instrumented fusion (images from 2.5 yr after surgery).
force plates (AMTI, Watertown, MA). The patient wore a rigid lumbosacral orthosis, an ankle-foot orthosis (AFO) on the left lower limb, and used a single-point cane in the right hand during the gait trials. Three-dimensional motion capture data were collected using an 11-camera system (Qualisys AB, Gothenburg, Sweden). Optoreflective markers were placed bilaterally to define the anatomical landmarks of the upper trunk, pelvis, and lower limbs and rigid arrays of markers were used to track segment motion. Lower limb kinetics (presented as internal muscle moments) were calculated from the kinematics and force plate data using an inverse dynamics model (Visual 3D software; C-Motion Inc., Germantown, MD). Kinesiological EMG data were collected from bipolar surface electrodes placed on the medial gastrocnemius, biceps femoris long/short head, and vastus lateralis on the right, and on the biceps femoris and vastus lateralis on the left (Motion Lab Systems Inc., Baton Rouge, LA). Gait events were defined using the vertical trajectories of the heel markers. Patient findings were compared with data from 4 healthy females (mean age, 23.5 yr) from a pre-existing laboratory database, and with previously published norms.6 Spine
It has been more than 4½ years since the patients underwent surgery. She has undergone routine follow-up positron emission tomography-computed tomography every 6 months since her surgery demonstrating no evidence of recurrent tumor. At her 1-year clinical evaluation she reported being free of pain on a regimen of gabapentin and occasional oxycodone. At more than 4½ years from surgery, she has minimal pain symptoms and is only rarely taking hydrocodone-acetaminophen to manage her symptoms. She is managing her bladder and bowel functions with urinary catheterization and a colostomy. Sexual intercourse is not possible due sensory deficits and pelvic scarring. She reports anesthesia in the perineal region and sacral dermatomes. However, since her surgery, she has remained free of tumor and her latest computed tomographic scan demonstrates stable spinopelvic instrumentation without evidence of instability (Figure 3A, B). Results from the 1-year monopolar EMG evaluation are summarized in Table 1. Motor unit activity was normal in muscles innervated by L2 through L5 on the right and by L2 through L4 on the left, with evidence of some reinnervation in the biceps femoris and medial gastrocnemius on the right. By 1 year after surgery, the patient walked independently for up to 20 minutes on level surfaces with a cane. She was also able to negotiate a 46% ramp and climb a single flight of stairs without assistance. At 6 months, her gait speed was an average of 0.64 m/s, and this had improved to 0.71 m/s, or 55% of normal gait speed for females, at 1 year after surgery.6 On average, the patient’s stride length was 19% less than that of healthy females, but improved from the 6-month to 1-year evaluation (1.00 m and 1.04 m, respectively).6 Step width was significantly decreased bilaterally, more profoundly on the left than on the right (−0.03 m and 0.04 m, respectively at 1 yr).6 In the sagittal plane, trunk motion was notable for rapid extension immediately after initial contact on the left (peak trunk extension during left loading response/midstance, 8.7° and that during right loading response/midstance, 4.2°). Selected time-series kinematics and kinetics of the lower limbs from the 1-year evaluation, with comparison with healthy females for reference, are presented in Figure 4. In particular, on the right, there was increased ankle dorsiflexion during loading response and midstance, increased knee flexion at initial contact and throughout stance phase, and increased stance phase hip abduction. On the left, there was knee hyperextension at initial contact and increased knee abduction and hip adduction throughout stance phase (Figure 4A). At 6 months, peak hip internal rotation during loading response on the left was 9.3° compared with 1.8° in the healthy subjects. However, at 1 year after surgery, peak internal rotation was normal while external rotation during midstance and terminal stance was increased. Stance phase kinetics (Figure 4B) were characterized by bilaterally decreased plantar flexor moments, increased knee extensor moments, and a reduction in the hip abductor moment on the left. Kinesiological EMG data from both evaluations indicated anomalous peaks of hamstring activity during preswing on the right and midswing on the left. Quadriceps activity was www.spinejournal.com
E1483
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited. SPINE140307_LR E1483
22/10/14 10:53 AM
CASE REPORT
Locomotion After Sacrectomy • Smith et al
Figure 3. CT follow-up imaging at 4 years after surgery; sagittal (A) and coronal (B). CT indicates computed tomography.
abnormally prolonged during stance on the right but this was less evident on the left. Activity in the right medial plantar flexors peaked during loading response and terminal stance.
DISCUSSION Total sacrectomy with sacral nerve root sacrifice and spinopelvic reconstruction is a rarely performed procedure that has significant consequences for locomotion. Mobility after total sacrectomy varies widely, ranging from independent walking with AFOs to wheelchair dependence.1–5,7,8 The variation in reported outcomes may in part be attributable to high rates of postoperative complications including nonunion and instrument failure. It has been suggested that preservation of the lumbar nerve roots permits independent locomotion after sacrectomy. This is due to maintenance of L5 innervation to the deep ankle plantar flexors and compensatory hypertrophy of the 2-joint hip extensors in response to impaired gluteus maximus E1484
www.spinejournal.com
function.7,9 In our patient, electrodiagnostic testing at 1 year after surgery indicated normal motor unit activity in muscles innervated by the L2, L3, L4, and L5 nerve roots on the right. In addition, both semimembranosus and gluteus maximus on the right had normal motor unit action potentials on electrodiagnostic evaluation, while the polyphasic motor unit potentials in biceps femoris and the medial gastrocnemius were suggestive of some reinnervation. These findings may be due to some partial sparing of S1 as a result of anomalous nerve root anatomy in this patient. On the left there was normal function in muscles innervated by the L2, L3, and L4 nerve roots. Motor unit action potentials were absent in muscles primarily innervated by S1 and S2 and those supplied by the sciatic nerve and its branches. Interestingly however, some semimembranosus function also seemed to be preserved on the left. At a relatively early stage postoperatively, this patient was able to walk independently for short distances with minimal November 2014
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited. SPINE140307_LR E1484
22/10/14 10:53 AM
CASE REPORT
Locomotion After Sacrectomy • Smith et al
TABLE 1. Electrodiagnostic EMG Evaluation at 1 yr After Surgery Right
Left
Tibialis anterior Insertional/spontaneous activity
Normal
Positive sharp waves
Voluntary activity
Normal
No motor unit activity
Insertional/spontaneous activity
Normal
Positive sharp waves
Voluntary activity
Normal
No motor unit activity
Insertional/spontaneous activity
Positive sharp waves
Fibrillations/positive sharp waves
Voluntary activity
Small amplitude, polyphasic MUAPs
No motor unit activity
Insertional/spontaneous activity
Normal
Normal
Voluntary activity
Normal
Normal
Insertional/spontaneous activity
Positive sharp waves
Positive sharp waves
Voluntary activity
One 9-phase polyphasic MUAP
No motor unit activity
Insertional/spontaneous activity
Normal
Normal
Voluntary activity
Normal
Normal
Insertional/spontaneous activity
Normal
Reduced insertional activity
Voluntary activity
Normal
No motor unit activity
Insertional/spontaneous activity
Reduced insertional activity
Reduced insertional activity
Voluntary activity
Small amplitude MUAPs
No motor unity activity
Tibialis posterior
Medial gastrocnemius
Vastus medialis
Biceps femoris
Semimembranosus
Gluteus maximus
Gluteus medius
MUAP indicates motor unit action potential; EMG, electromyography.
assistance from a cane. Her gait speed was reduced compared with that of healthy young females. Normally, the speed of forward progression during locomotion is maximized by propulsive forces during preswing phase, smooth forward motion of the body over a series of pivots or “rockers” during stance phase, and the forward momentum of the body.6 In this patient, all 3 of these factors were affected as a result of both denervation and gait compensations. First, as expected, the plantar flexor moment prior to swing phase was substantially decreased bilaterally in this patient compared with healthy subjects. Although passive tension in the triceps surae may contribute to push off at preswing in healthy subjects,6 unusual biceps femoris EMG activity at preswing on the right in this patient suggested that as a result of decreased active plantar flexion, a compensatory mechanism was required to drive the trailing limb into knee flexion. Second, the AFO reduced the normal ability of the left ankle, forefoot, and toes Spine
to act as pivots for the forward advancement of the center of mass. Finally, the abnormal trunk extension that we observed in this patient during left stance phase may have served to limit the hip extensor moment on the left side by moving the center of mass posterior to the hip joint, but also reduced forward momentum of the center of mass.6 Several gait characteristics after sacrectomy were related to impaired function of the hip musculature. It was anticipated that the loss of the eccentric action of gluteus maximus at initial contact would result in an increase in hip adduction and internal rotation on the left side during loading response. This was observed at the 6-month evaluation, but to a lesser extent at the 1-year evaluation. In addition, the unopposed action of the adductors on the left was likely responsible for the increased hip adduction throughout the stride cycle and decreased left step width in this patient. This was not so apparent on the right due to residual function www.spinejournal.com
E1485
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited. SPINE140307_LR E1485
22/10/14 10:53 AM
CASE REPORT
Locomotion After Sacrectomy • Smith et al
Figure 4. Selected time-series kinematics (A) and kinetics (B) of the ankle, knee, and hip compared with healthy subjects (n = 4). Positive values indicate (A) ankle dorsiflexion, knee flexion, and hip abduction joint angles and (B) ankle plantar flexor, knee extensor, and hip abductor moments. Data are time normalized to 100% stride cycle (kinematics) and stance phase (kinetics). Kinetic data are normalized to body weight.
in the hip abductors and the relatively normal hip abductor moment. However, despite impaired left hip abductor strength, a Trendelenburg gait pattern was not evident during left stance phase, likely due to the use of the cane in the contralateral hand. Partial (right) and total (left) loss of the innervation to triceps surae resulted in impaired deceleration of the tibia over the foot and increased ankle dorsiflexion and knee flexion during the stance phase of gait. This was more pronounced on the right than on the left, due to the stabilizing effect of the AFO. As a result, stance phase knee extensor moments, and duration of quadriceps EMG activity, were substantially increased on the right side compared with healthy subjects. On the left, a different compensation was observed, with knee hyperextension at initial contact and reduced flexion throughout stance phase helping to compensate for the absent plantar flexors and reduced quadriceps strength. As anticipated, hip extensor moments during loading response were reduced bilaterally compared with controls. Although we hypothesized that an increased hip flexor moment during pre- and initial-swing phases would be evident as a compensation for decreased push off at the ankle,10 this was not observed in this patient, potentially due to her reduced gait speed compared with the controls. E1486
www.spinejournal.com
It should be noted that although our data support the subjective and anecdotal locomotor outcomes that have been presented in earlier manuscripts describing patients with similar patterns of nerve root sacrifice after sacrectomy,7,9 and previously hypothesized patterns of compensation, it is possible that other individuals will display different patterns of kinetic and kinematic compensation after this surgery.
CONCLUSION In this patient, substantial independent locomotion was possible as early as 6 months after successful total sacrectomy and spinopelvic reconstruction, and the spatiotemporal and kinematic aspects of locomotion continued to improve at 1 year after surgery. Despite sacrifice of the sacral nerve roots a functional locomotor pattern may be achieved with kinematic and kinetic compensations.
➢ Key Points Locomotor biomechanics were assessed in a young patient after successful total en bloc sacrectomy and spinopelvic reconstruction with resection of the sacral nerve roots bilaterally. November 2014
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited. SPINE140307_LR E1486
22/10/14 10:53 AM
CASE REPORT The patient achieved a functional independent locomotor pattern relatively early after surgery, and the spatiotemporal characteristics of her gait pattern continued to improve at 1-year follow-up. Kinematic and kinetic compensations were evident in the sagittal and coronal planes in the trunk, hips, knees, and ankles.
References
1. Singh R, Custodio C, Lidestri P, et al. Functional outcome in a patient with radical sacrectomy and sacral cord resection: a case report. 2004 Annual Academy Assembly Abstracts. Arch Phys Med Rehabil 2004;85:E17–8. 2. Spiegel DA, Richardson WJ, Scully SP, et al. Long-term survival following total sacrectomy with reconstruction for the treatment of primary osteosarcoma of the sacrum. J Bone Jt Surg 1999;81-A:848–5.
Spine
Locomotion After Sacrectomy • Smith et al
3. Zileli M, Hoscoskun C, Brastianos P, et al. Surgical treatment of primary sacral tumors: complications associated with sacrectomy. Neurosurg Focus 2003;15:1–8. 4. Sahakitrungruang C, Chantra K, Dusitanond N, et al. Sacrectomy for primary sacral tumors. Dis Colon Rectum 2009;52:913–8. 5. Newman CB, Keshavarzi S, Aryan HE. En bloc sacrectomy and reconstruction: technique modification for pelvic fixation. Surg Neurol 2009;72:752–6. 6. Perry J, Burnfield J. eds. Gait Analysis: Normal and Pathological Function. 2nd ed. Thorofare, NJ: Slack Inc; 2010. 7. Ohata N, Ozaki T, Kunisada T, et al. Extended total sacrectomy and reconstruction for sacral tumor. Spine 2004;29;E123–6. 8. Santi MD, Mitsunaga MM, Lockett JL. Total sacrectomy for a giant sacral schwannoma: a case report. Clin Orthop Relat R 1993;294:285–9. 9. Wuisman P, Lieshout O, Sugihara S, et al. Total sacrectomy and reconstruction: oncologic and functional outcome. Clin Orthop Relat R 2000;381:192–203. 10. Lewis CL, Ferris DP. Walking with increased ankle push off decreases hip muscle moments. J Biomech 2008;41:2082–9.
www.spinejournal.com
E1487
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited. SPINE140307_LR E1487
22/10/14 10:53 AM
