Radiographic and Neurological Outcome After Surgical Treatment of Traumatic Fractures of the Ankylotic Thoracic Spine: A Retrospective Case Series

Radiographic and Neurological Outcome After Surgical Treatment of Traumatic Fractures of the... Abstract BACKGROUND Spontaneous thoracic ankylosis is a progressive degenerative process that predisposes patients to potentially highly unstable traumatic injuries. Acute hyperextension fractures result in dynamic instability putting the spinal cord at risk. OBJECTIVE To describe preoperative radiographic characteristics of fractures of the ankylotic thoracic spine and relate findings to early postoperative radiographic and clinical outcomes. METHODS A single center, retrospective review was performed of 28 surgically treated patients with fractures of the ankylotic thoracic spine. Radiographic assessment included preoperative fracture angulation (FA) and fracture displacement (FD), and postoperative change in sagittal alignment. Early clinical outcomes included preoperative and postoperative American Spinal Injury Association (ASIA) grade and perioperative complications. RESULTS Seven patients (25%) presented with poor neurological grade (ASIA A-C) compared to 21 (75%) with good grade (ASIA D, E). At presentation, poor grade patients had a mean FA of 16.4° (range 0°-34.5°), and FD of 7.76 mm (range 0.8-9.2). Good grade patients had a mean FA of 18.2° (range 0°-43.3°), and FD of 4.77 mm (range 0-25.1). There was no statistically significant difference in FA or FD between groups (P = .70 and .20 respectively). All underwent posterior pedicle screw fixation for stabilization. Fifty per cent of patients presenting with ASIA C or D spinal cord injury improved 1 or more ASIA grades. There were no perioperative complications. Early postoperative sagittal alignment was maintained with a mean change of –2.6°. CONCLUSION Presenting fracture alignment does not significantly correlate with pre- or postoperative neurological status. Early posterior stabilization preserved neurological function, with neurological recovery occurring in a portion of individuals. Ankylosis, Extension-distraction injury, Fracture angulation, Fracture displacement, Hyperextension-spondylolysis, Posterior stabilization, Spinal cord injury ABBREVIATIONS ABBREVIATIONS ASIA American Spinal Injury Association CI confidence interval CT computed tomography FA fracture angulation FD fracture displacement FL fracture level IRB Institutional Review Board MVC motor vehicle collision SCI spinal cord injury SD standard deviation Spontaneous ankylosis of the spine is a chronic, degenerative disorder characterized by progressive autofusion across multiple motion segments of the vertebral column. Ankylosing spondylitis (AS) is one etiology of spinal ankylosis due to progressive enthesitis at the intervertebral disc space leading to syndesmophyte formation and eventual fusion;1-3 however, degenerative ankylosis can occur in the absence of AS secondary to calcification of the anterior or posterior longitudinal ligaments or disc space. Regardless of underlying pathogenesis, multilevel vertebral ankylosis, particularly at kyphotic segments, puts affected individuals at risk for severe hyperextension injury with spinal trauma.4 Fractures of the ankylotic cervical spine have been previously well described, mostly in patients with underlying AS.5 Increased risk of spinal cord injury (SCI) after AS-related cervical fractures is attributed to severe instability that occurs with hyperextension through the ankylosed “bamboo” spine. Immediate immobilization and/or stabilization of AS-related cervical spine fractures is critical to prevent further neurological injury from unstable fracture dislocation and dynamic mechanical compression of the spinal cord.6-8 While much has been published about cervical injuries, little exists in the literature describing treatment of traumatic fractures of the ankylotic thoracic spine and neurological outcomes. The thoracic spine is normally kyphotic, and therefore like AS-related cervical fracture, is at risk for worsening injury with hyperextension.9 In addition, because of the regional thoracic kyphosis, fracture in the setting of multilevel ankylosis may pose challenges in maintaining anatomic alignment and protection of the spinal cord when transferring an acute trauma patient. Paradoxically, positioning a patient supine (eg, flat on a trauma board) may worsen hyperextension injury through the fracture segment of an ankylotic thoracic spine, underscoring the importance of fracture stabilization. A recent thoracolumbar spine classification system characterized the potential instability of this hyperextension fracture in ankylotic disorders,10 with the recommendation that these fractures generally be treated surgically.11 To date, the existing literature reporting this fracture type in the setting of thoracic ankylosis or AS involves only small case series (<10 patients).12,13 Therefore, we report the largest known single center, case series of surgically treated fractures of the ankylotic thoracic spine with early radiographic and neurological outcomes. METHODS Study Design A single-center retrospective chart review was performed of surgically treated patients with traumatic fracture of the ankylotic thoracic spine. All patients underwent treatment by the neurosurgery department between 2010 and 2015. Institutional Review Board (IRB) approval was obtained prior to study initiation (IRB#: I201500874). In accordance with IRB policy, patient consent did not need to be obtained for this study. Patients were identified from operative case logs of the senior spine neurosurgeons. Medical records and radiological imaging studies were reviewed to select those with ankylosis of the thoracic spine, as defined by computed tomography (CT) evidence of continuous bridging calcification across ≥3 intervertebral discs. Imaging of the sacroiliac joints and HLA-B27 gene testing was not routinely performed to determine if AS was the underlying cause of ankylosis. Admission CT imaging was reviewed by a single observer to describe presenting fracture characteristics including fracture level (FL), AO classification fracture type,11,14 fracture angulation (FA), and anteroposterior fracture displacement (FD). All patients presented as a trauma alert and underwent a contrasted CT body scan, which included thin-slice, multiplanar imaging of the full thoracic and lumbar spine. FA was determined by measuring the degree of angulation of the fracture distraction (Figure 1A). FD was determined by measuring the horizontal distance (mm) between vertical lines drawn along the dorsal vertebral body of the rostral and caudal vertebra (Figure 1B). All radiological measurements were made using computer imaging software (Visage Imaging v7.1.6, Richmond, Australia). Change in sagittal alignment was determined by comparing the preoperative and postoperative Cobb angle formed by the inferior endplate of the adjacent vertebra cranial to the fracture level and the superior endplate of the adjacent caudal vertebra. FIGURE 1. View largeDownload slide Measurements of A, fracture angulation (FA) and B, fracture displacement (FD). The figure demonstrates measurements of FA (left) and FD (right) for this cohort of patients. FIGURE 1. View largeDownload slide Measurements of A, fracture angulation (FA) and B, fracture displacement (FD). The figure demonstrates measurements of FA (left) and FD (right) for this cohort of patients. Clinical data were obtained from individual patient medical charts. American Spinal Injury Association (ASIA) grade was determined by documentation of motor and sensory score at admission, postoperative day 1, and last follow-up. Surgical procedure performed including type of stabilization, and number of levels was determined from operative reports. Complications reviewed included postoperative neurological deterioration, reoperation, surgical site infection, postoperative hematoma, instrumentation failure, and mortality. Statistical Methods Association between spinal fracture radiographic characteristics and clinical outcomes was examined using descriptive statistics. Patients were grouped for analysis by either “poor ASIA grade” (ASIA A-C) or “good ASIA grade” (ASIA D, E). The relationship between average FA and FD to poor vs good ASIA grade at presentation and last follow-up were analyzed using Wilcoxon rank-sum tests to generate P values. This approach was used to account for variable length of follow-up in determining a relationship between fracture alignment and clinical outcome. All P values were considered significant if ≤.05. Interval-censored exponential survival regression was used to estimate the average rate of neurological improvement in ASIA grade from initial presentation to last follow-up. This improvement rate was then tested for association with FA and FD, with a confidence interval (CI) set at 95%. RESULTS Twenty-eight patients with traumatic fracture of the ankylotic thoracic spine were identified. Individual patient clinical characteristics are listed in Table. The mean age was 69.1 ± 9.95 yr (standard deviation [SD]) and 18 (64%) were male. All fractures were classified as type B3 hyperextension injuries with disruption of the anterior longitudinal ligament and through the disc space or vertebral endplate. TABLE. General Patient Characteristics of the 28 Individuals in the Study Cohort Patient number  Age /sex  Time to OR (d)  Length of stay  Fracture level  Other trauma present  Medical morbidities  Fracture angulation (°)  Fracture displacement (mm)  Procedure levels and (total no. of ankylosed levels)  No. of levels stabilized (change in sagittal alignment degrees)  Follow-up (d)  ASIA grade presentation  ASIA grade last follow-up  Mortality (Y/N)  1  72/M  2  13  T12-L1  N  DM  23.5  4.4  T8-L5 (9)  6 (–2.6)  13  A  A  N  2  64/M  0  5  T11-12  N  CAD, CHF, Afib on coumadin, DM  19.4  0  T10-L1 (11)  3 (–12)  158  E  E  N  3  64/M  0  1  T6-7  Pneumothorax  Afib on coumadin, Pacemaker  17.3  25.1  T3-11 (14)  8 (–0.6)  24  D  D  N  4  79/M  0  28  T8-9  Rib fx  HTN, ASA 325  19  1.6  T7-10 (13)  3 (–6.0)  58  E  E  N  5  81/M  7  7  T7  Rib fx, pulmonary contusion  HTN, Afib, CAD, bladder cancer  16.6  19.2  T5-10 (9)  5 (–2.8)  49  A  A  Y  6  65/F  2  13  T10-L1  N  HTN, obesity  23.3  7.1  T7-L4 (14)  9 (0)  178  E  E  N  7  68/M  2  6  T6  Pneumothorax  ASA 325 mechanical aortic valve  8.6  4.5  T4-8 (11)  4 (–5.6)  107  E  E  N  8  83/M  0  6  T4  N  Afib on coumadin, Pacer, CAD  6.9  0.8  T2-6 (13)  4 (–0.3)  19  A  A  Y  9  59/F  4  9  T9  Falcine SDH, Pelvic fx  Sickle cell anemia, DM, HTN  16.8  3.8  T7-11 (10)  4 (1.2)  104  E  E  N  10  67/M  4  12  T11-12  Pelvic fx, R tibia/fibula fx  Prostate cancer  24.7  8.5  T10-L1 (12)  3 (4.5)  8  C  C  N  11  62/M  0  4  T4-5  N  HTN, HLD  14.2  5.3  T2-6 (10)  4 (3.1)  281  E  E  N  12  62/M  1  8  T10-11  N  HTN, HLD, PE on xarelto  8.5  9.8  T9-12 (5)  3 (1.2)  234  C  D  N  13  58/F  2  14  T3-4  Hemothorax, R radius fx  HTN, HLD, COPD, DM  0  9.8  T1-7 (13)  6 (0.0)  488  C  E  N  14  73/F  5  21  T11-12  N  Afib on coumadin, HTN, HLD, OSA, CAD, CKD  34.5  1.8  T9-L2 (12)  5 (–16.2)  16  C  D  Y  15  76/F  6  11  T12  N  HTN, HLD, DM  19.7  4.1  T10-L3 (5)  5 (1.9)  215  E  E  N  16  72/M  2  6  T6-7  R occipital SDH  None  10.5  1.3  T4-9 (11)  5 (–1.2)  36  E  E  N  17  59/F  1  13  T3-4  N  HTN, HLD, hypothyroid, DM  14.3  10.5  T1-7 (14)  5 (0.0)  486  D  E  N  18  65/M  0  5  T11  N  HTN, DM  NA  6.50  T10-L1 (5)  3 (–9.6)  5  E  E  N  19  42/F  6  14  T7-8  N  HTN, DM, urinary incontinence  21.5  0  T4-11 (4)  7 (–4.6)  157  E  E  N  20  61/M  1  25  T12-L1  R/L femur and tibia/fibula fx  DM, Hep C  43.3  5  T11-L3 (6)  4 (5.3)  952  D  E  N  21  66/M  5  10  T9-T10  N  None  14.2  6.5  T5-L2 (12)  9 (–1.5)  53  E  E  N  22  89/F  2  12  T9-T10  N  HF, HTN, DM  24.9  2.8  T6-12 (12)  6 (–3.7)  10  E  E  N  23  69/F  5  13  T3-T4  N  CAD on daily ASA/plavix  22.4  4.6  T1-6 (8)  5 (–1.7)  8  D  D  N  24  82/M  0  4  T11  N  HTN, HLD, DM  8.6  2.4  T9-L2 (21)  5 (6.4)  241  E  E  N  25  84/M  0  4  T10  R tibia/fibula fx  Afib, HTN, HLD, CAD, DM  34.6  0  T8-12 (14)  4 (–4.3)  96  E  E  N  26  73/M  1  18  T11  Pelvic fx  None  0  1.2  T9-L1 (12)  4 (0.9)  45  D  D  N  27  69/M  4  8  T9  Pneumothorax  DM  10.8  0.8  T8-11 (15)  3 (–6.6)  53  E  E  N  28  72/F  3  8  T12  N  Afib, DVT on coumadin, HF, HTN, HLD  19.8  7  T8-L4 (20)  8 (–18)  51  D  D  N  Patient number  Age /sex  Time to OR (d)  Length of stay  Fracture level  Other trauma present  Medical morbidities  Fracture angulation (°)  Fracture displacement (mm)  Procedure levels and (total no. of ankylosed levels)  No. of levels stabilized (change in sagittal alignment degrees)  Follow-up (d)  ASIA grade presentation  ASIA grade last follow-up  Mortality (Y/N)  1  72/M  2  13  T12-L1  N  DM  23.5  4.4  T8-L5 (9)  6 (–2.6)  13  A  A  N  2  64/M  0  5  T11-12  N  CAD, CHF, Afib on coumadin, DM  19.4  0  T10-L1 (11)  3 (–12)  158  E  E  N  3  64/M  0  1  T6-7  Pneumothorax  Afib on coumadin, Pacemaker  17.3  25.1  T3-11 (14)  8 (–0.6)  24  D  D  N  4  79/M  0  28  T8-9  Rib fx  HTN, ASA 325  19  1.6  T7-10 (13)  3 (–6.0)  58  E  E  N  5  81/M  7  7  T7  Rib fx, pulmonary contusion  HTN, Afib, CAD, bladder cancer  16.6  19.2  T5-10 (9)  5 (–2.8)  49  A  A  Y  6  65/F  2  13  T10-L1  N  HTN, obesity  23.3  7.1  T7-L4 (14)  9 (0)  178  E  E  N  7  68/M  2  6  T6  Pneumothorax  ASA 325 mechanical aortic valve  8.6  4.5  T4-8 (11)  4 (–5.6)  107  E  E  N  8  83/M  0  6  T4  N  Afib on coumadin, Pacer, CAD  6.9  0.8  T2-6 (13)  4 (–0.3)  19  A  A  Y  9  59/F  4  9  T9  Falcine SDH, Pelvic fx  Sickle cell anemia, DM, HTN  16.8  3.8  T7-11 (10)  4 (1.2)  104  E  E  N  10  67/M  4  12  T11-12  Pelvic fx, R tibia/fibula fx  Prostate cancer  24.7  8.5  T10-L1 (12)  3 (4.5)  8  C  C  N  11  62/M  0  4  T4-5  N  HTN, HLD  14.2  5.3  T2-6 (10)  4 (3.1)  281  E  E  N  12  62/M  1  8  T10-11  N  HTN, HLD, PE on xarelto  8.5  9.8  T9-12 (5)  3 (1.2)  234  C  D  N  13  58/F  2  14  T3-4  Hemothorax, R radius fx  HTN, HLD, COPD, DM  0  9.8  T1-7 (13)  6 (0.0)  488  C  E  N  14  73/F  5  21  T11-12  N  Afib on coumadin, HTN, HLD, OSA, CAD, CKD  34.5  1.8  T9-L2 (12)  5 (–16.2)  16  C  D  Y  15  76/F  6  11  T12  N  HTN, HLD, DM  19.7  4.1  T10-L3 (5)  5 (1.9)  215  E  E  N  16  72/M  2  6  T6-7  R occipital SDH  None  10.5  1.3  T4-9 (11)  5 (–1.2)  36  E  E  N  17  59/F  1  13  T3-4  N  HTN, HLD, hypothyroid, DM  14.3  10.5  T1-7 (14)  5 (0.0)  486  D  E  N  18  65/M  0  5  T11  N  HTN, DM  NA  6.50  T10-L1 (5)  3 (–9.6)  5  E  E  N  19  42/F  6  14  T7-8  N  HTN, DM, urinary incontinence  21.5  0  T4-11 (4)  7 (–4.6)  157  E  E  N  20  61/M  1  25  T12-L1  R/L femur and tibia/fibula fx  DM, Hep C  43.3  5  T11-L3 (6)  4 (5.3)  952  D  E  N  21  66/M  5  10  T9-T10  N  None  14.2  6.5  T5-L2 (12)  9 (–1.5)  53  E  E  N  22  89/F  2  12  T9-T10  N  HF, HTN, DM  24.9  2.8  T6-12 (12)  6 (–3.7)  10  E  E  N  23  69/F  5  13  T3-T4  N  CAD on daily ASA/plavix  22.4  4.6  T1-6 (8)  5 (–1.7)  8  D  D  N  24  82/M  0  4  T11  N  HTN, HLD, DM  8.6  2.4  T9-L2 (21)  5 (6.4)  241  E  E  N  25  84/M  0  4  T10  R tibia/fibula fx  Afib, HTN, HLD, CAD, DM  34.6  0  T8-12 (14)  4 (–4.3)  96  E  E  N  26  73/M  1  18  T11  Pelvic fx  None  0  1.2  T9-L1 (12)  4 (0.9)  45  D  D  N  27  69/M  4  8  T9  Pneumothorax  DM  10.8  0.8  T8-11 (15)  3 (–6.6)  53  E  E  N  28  72/F  3  8  T12  N  Afib, DVT on coumadin, HF, HTN, HLD  19.8  7  T8-L4 (20)  8 (–18)  51  D  D  N  N = no, R = right, L = left, fx = fracture, SDH = subdural hematoma, Afib = atrial fibrillation, ASA = aspirin, CAD = coronary artery disease, CHF = congestive heart failure, COPD = chronic obstructive pulmonary disease, DM = diabetes, DVT = deep venous thrombosis, Hep C = hepatitis C, HF = heart failure (unspecified), HLD = hyperlipidemia, HTN = hypertension, PE = pulmonary embolism, OSA = obstructive sleep apnea. View Large Descriptive Data of Patient Cohort The average duration of last clinical follow-up was 5 mo (range: <1 to 16.2 mo). All underwent posterior pedicle screw–rod stabilization, at an average 2.3 d from time of injury (range: 0-7). Mean number of instrumented levels was 5 (range 3-9 levels). The mean number of ankylosed segments was 11.25 ± 4.1 (SD) across the thoracic and lumbar spine. All underwent postoperative X-rays for evaluation of fracture sagittal alignment and instrumentation. Fourteen of the 28 patients additionally suffered nonspine-related trauma including upper and/or lower extremity fractures, pelvic fractures, or traumatic brain injury (see Table). Complications There were no reported surgery-related complications of postoperative infection, hematoma, instrumentation failure or screw misplacement, reoperation, myocardial infarction, acute respiratory failure, urinary tract infection, deep vein thrombosis, or pulmonary embolus. Three patients died during the follow-up period due to nonneurological and nonsurgical complications related to the initial presenting trauma. One patient was an 81-yr-old male who presented after motor vehicle collision (MVC) with multisystem injuries including a positive focused assessment with sonography in trauma scan, T7 ASIA A SCI, and traumatic intracranial subarachnoid hemorrhage with intraventricular extension. He underwent spine surgical treatment 5 d after initial presentation. Ultimately, the patient was unable to be weaned off pressors or dialysis due to acute kidney injury and cardiac trauma, and family withdrew care on postoperative day 11. The second death was an 83-yr-old male presenting after MVC who, despite report of ambulating after the accident, presented to the hospital with a T2 ASIA A SCI. He ultimately expired over 2 yr after surgery due to unrelated cardiac causes. The third death was a 73-yr-old female presenting after ground level fall with a T11-T12 fracture. She had an extensive medical history including coronary artery disease, sleep apnea, and atrial fibrillation. She underwent posterior fusion 5 d after presentation. She was discharged to acute inpatient rehabilitation on postoperative day 16 in stable condition after which documentation ceased and the patient was marked in the medical record as deceased. Radiographic and Neurological Outcome Individual presenting fracture angulation and displacement are presented in Table. Seven patients (25%) presented with poor ASIA grade (ASIA A-C) compared to 21 (75%) with good ASIA grade (ASIA D, E; Figure 2). Poor ASIA grade patients presented with fracture levels ranging from T1-T12, and mean FA of 16.4° (range 0°-34.5°) and FD of 7.76 mm (range 0.8-9.2). Good ASIA grade patients presented with fracture levels ranging from T3-12, and mean FA of 18.2° (range 0°-43.3°), and FD of 4.77 mm (range 0-25.1). There was no significant difference in FA or FD between poor and good ASIA grade groups (P = .70 and .20, respectively). Change in sagittal alignment (negative number represents increased kyphosis postoperative vs preoperative) is represented for each individual patient in Table. The average change in postoperative sagittal alignment for all patients was –2.6° ± 5.9° (SD). FIGURE 2. View largeDownload slide Graphical representation of admission AIS and last follow-up AIS for 28 patients. The purple line for ASIA E represents 15 patients. AIS, American Spinal Injury Association and International Medical Society of Paraplegia Impairment Scale . FIGURE 2. View largeDownload slide Graphical representation of admission AIS and last follow-up AIS for 28 patients. The purple line for ASIA E represents 15 patients. AIS, American Spinal Injury Association and International Medical Society of Paraplegia Impairment Scale . At last follow-up, 24 (85.7%) were ASIA D or E, compared to 75% at initial presentation. All ASIA A patients at presentation (n = 3) remained ASIA A at last follow-up. No patient worsened neurologically during the follow-up period. Ten patients presented with motor incomplete SCI (ASIA C, D), of which 5 (50%) improved at least 1 ASIA grade. Of these, 1 ASIA C and 2 ASIA D patients recovered normal neurological function (ASIA E). The improvement rate for ASIA C and D patients was 3.95 events/person-year of follow-up (95% CI 0.95-16.42). The improvement rate was not significantly related to either admission FA or FD, P = .17 and .81, respectively. DISCUSSION Progressive multisegmental autofusion of the ankylotic thoracic spine predisposes patients to potentially devastating hyperextension injuries. The altered biomechanics in diffuse ankylosis do not allow for compensation of normal physiologic loading by distributing flexion-extension and translational forces across multiple motion segments. Particularly, ankylosis across a kyphotic region of the spine predisposes to potentially catastrophic anterior tension band failure when subjected to extension loads.13,14,15 Instability at the affected segment in thoracic ankylosis is magnified by the long functional lever arms created by the multisegment fused vertebrae above and below the fracture. In our series, we observed an average of >11 ankylosed motion segments across the thoracic and lumbar spine. This instability is subject to ongoing dynamic forces that put the spinal cord at risk until adequately stabilized. Therefore, prompt recognition of these injuries, appropriate immobilization in anatomic alignment, and ultimately surgical stabilization is critical for preservation of neurological function. We reviewed our series of 28 patients with fractures of the ankylotic thoracic spine treated by a single neurosurgery department to demonstrate clinical and radiographic characteristics of these unique injuries. To our knowledge, this represents the largest surgical case series to date of patients with fractures in the setting of thoracic ankylosis.13,16 Nearly half (46.4%) of the patients in our series presented with neurological deficits, which is comparable to previous series and attests to the instability and spinal cord risk of these injuries.17 Surgical treatment with posterior pedicle screw–rod stabilization was not only protective of neurological function (none deteriorated during follow-up), but may facilitate recovery of function. Half of the motor incomplete ASIA C and D patients on presentation improved 1 or more ASIA grades during the follow-up period. Based on this high recovery rate, we suspect that fractures of the ankylotic thoracic spine result in hypermobile, dynamic spinal cord compression, which in some cases have shown potential for recovery of neurological function with surgical stabilization. Furthermore, this benefit with surgical treatment can be performed safely with relatively low morbidity and risk of complications. Paradoxical Relation of Fracture Geometry to Clinical Outcome Despite the high rate of neurological recovery in this surgical series, a significant proportion of patients did not improve. Fifty-seven per cent (4 of 7) that were ASIA A-C at presentation remained poor SCI grade at last follow-up. Of note, 3 of 4 patients presented as ASIA A. Therefore, we sought to determine if admission fracture characteristics predicted neurological function at presentation, and likelihood of recovery. All fractures were consistent with a type B3, hyperextension injury with disruption through the anterior longitudinal ligament and disc space, and a characteristic “fish mouth” widening anteriorly. In our series, degree of fracture angulation did not correlate with either ASIA grade at presentation or likelihood of recovery. Extent of fracture displacement also did not correlate with neurological presentation or recovery, although there was a nonsignificant mean 2.99 mm greater retrolisthesis in the poor ASIA grade patients compared to good ASIA grade group. This unexpected lack of correlation between radiographic fracture characteristics with ASIA grade is counter to other thoracolumbar fractures in which extent of fracture displacement relates to severity of neurologic impairment.11,18,19 A possible explanation in this unique fracture type is that hyperextension injury in ankylosis causes hypermobile instability at a normally kyphotic region of the spine. Therefore, significant dynamic changes in the fracture pattern may occur depending on postural changes during positioning, including the potential for relatively benign appearing, minimal displacement when positioned upright. Head-of-bed elevated positioning may promote reduction of the fracture into normal anatomic kyphotic alignment, whereas supine positioning (ie, thoracic extension) may cause greater hyperextension angulation and/or displacement with possible worsening spinal cord compression. This repetitive dynamic dislocation-reduction pattern with variable patient positioning may explain the lack of relationship between fracture alignment on imaging study and neurological presentation. This observation speaks to the importance of prompt recognition of fractures of the ankylotic thoracic spine, as the severity of instability and neurological injury may not be readily apparent based on fracture alignment during a single imaging study. Furthermore, special consideration should be made when positioning that may place the injured spine in hyperextension (eg, flat, supine for CT or MRI). When obtaining supine imaging, care should be made to optimally maintain anatomic alignment of the patient's normal thoracic kyphosis with clinical assessment for any neurological changes. Limitations There are several limitations to this study. The retrospective design relied on medical chart and imaging review for data collection, with inherent potential for limited follow-up. As a tertiary referral center, it is common for acute trauma patients to be transferred hospital–hospital for neurosurgical evaluation and treatment, and then return to the initial admitting hospital for subsequent care. Therefore, radiographic and clinical follow-up is often limited to early assessment during the acute perioperative period. Despite this, we observed that neurological outcome was preserved after surgery, with a high portion (50%) of motor incomplete SCI patients demonstrating recovery of 1 or more ASIA grade. Second, the diagnosis of thoracic ankylosis was made using CT imaging of the thoracic spine for evidence of continuous, multilevel anterior longitudinal ligament and disc space calcification. While this imaging finding was typical of the characteristic “bamboo spine” of AS, the definitive diagnosis of AS was not determined by HLA-B27 gene testing. Regardless, the relevant anatomy and biomechanics of the ankylotic thoracic spine in this case series was likely similar to that observed in AS. Last, this case series represents the preferred management strategy of the senior spine neurosurgeons, which is to treat these fractures during the presenting hospitalization with posterior pedicle screw–rod fixation.11 Since this was a retrospective surgical case series, reasonable conclusions about nonoperative external immobilization for these fractures cannot be made, although conventional orthoses that tend to place the spine in extension may be inadequate. Therefore, conclusions regarding the benefit of nonoperative vs operative, or various operative techniques for this particular fracture type, and their potential impact on neurological recovery and clinical outcome cannot be made. CONCLUSION Traumatic injuries of the ankylotic thoracic spine result in unstable fracture angulation and displacement with potential for significant SCI. Severity of fracture dislocation does not appear to correlate with neurological function or likelihood of recovery. Early surgical posterior stabilization is observed to preserve neurological function and improve recovery in a portion of individuals presenting with incomplete SCI, with a relatively low risk of perioperative complications. Disclosure The authors have no personal, financial, or institutional interest in any of the drugs, materials, or devices described in this article. REFERENCES 1. Braun J, Sieper J. Ankylosing spondylitis. Lancet . 2007; 369( 9570): 1379- 1390. Google Scholar CrossRef Search ADS PubMed  2. Mcgonagle D, Gibbon W, Emery P. Classification of inflammatory arthritis by enthesitis. Lancet . 1998; 352( 9134): 1137- 1340. Google Scholar CrossRef Search ADS PubMed  3. Vosse D, Feldtkeller E, Erlendsson J et al.   Clinical vertebral fractures in patients with ankylosing spondylitis. J Rheumatol . 2004; 31( 10): 1981- 1985. Google Scholar PubMed  4. Ulu MA, Batmaz İ, Dilek B et al.   Prevalence of osteoporosis and vertebral fractures and related factors in patients with ankylosing spondylitis. Chin Med J . 2014; 127( 15): 2740- 2747. Google Scholar PubMed  5. Mitra D, Elvins DM, Speden DJ et al.   The prevalence of vertebral fractures in mild ankylosing spondylitis and their relationship to bone mineral density. Rheumatology (Oxford) . 2000; 39( 1): 85- 89. Google Scholar CrossRef Search ADS PubMed  6. Lukasiewicz AM, Bohl DD, Varthi AG et al.   Spinal fracture in patients with ankylosing spondylitis: cohort definition, distribution of injuries, and hospital outcomes. Spine . 2016; 41( 3): 191- 196. Google Scholar CrossRef Search ADS PubMed  7. Ma J, Wang C, Zhou X et al.   Surgical therapy of cervical spine fracture in patients with ankylosing spondylitis. Medicine (Baltimore) . 2015; 94( 44): e1663. Google Scholar CrossRef Search ADS PubMed  8. Lazennec JY, D’astorg H, Rousseau MA. Cervical spine surgery in ankylosing spondylitis: review and current concept. Orthop Traumatol Surg Res . 2015; 101( 4): 507- 513. Google Scholar CrossRef Search ADS PubMed  9. Geusens P, Vosse D, Van der heijde D et al.   High prevalence of thoracic vertebral deformities and discal wedging in ankylosing spondylitis patients with hyperkyphosis. J Rheumatol . 2001; 28( 8): 1856- 1861. Google Scholar PubMed  10. Vaccaro AR, Oner C, Kepler CK et al.   AOSpine thoracolumbar spine injury classification system: fracture description, neurological status, and key modifiers. Spine . 2013; 38( 23): 2028- 2037. Google Scholar CrossRef Search ADS PubMed  11. Vaccaro AR, Schroeder GD, Kepler CK et al.   The surgical algorithm for the AOSpine thoracolumbar spine injury classification system. Eur Spine J . 2016; 25( 4): 1087- 1094. Google Scholar CrossRef Search ADS PubMed  12. El tecle NE, Abode-iyamah KO, Hitchon PW et al.   Management of spinal fractures in patients with ankylosing spondylitis. Clin Neurol Neurosurg . 2015; 139: 177- 182. Google Scholar CrossRef Search ADS PubMed  13. Lu ML, Tsai TT, Lai PL et al.   A retrospective study of treating thoracolumbar spine fractures in ankylosing spondylitis. Eur J Orthop Surg Traumatol . 2014; 24 ( suppl 1): S117- S123. Google Scholar CrossRef Search ADS PubMed  14. Magerl F, Aebi M, Gertzbein SD et al.   A comprehensive classification of thoracic and lumbar injuries. Eur Spine J . 1994; 3( 4): 184- 201. Google Scholar CrossRef Search ADS PubMed  15. Gelman MI, Umber JS. Fractures of the thoracolumbar spine in ankylosing spondylitis. AJR Am J Roentgenol . 1978; 130( 3): 485- 491. Google Scholar CrossRef Search ADS PubMed  16. Westerveld LA, Verlaan JJ, Oner FC. Spinal fractures in patients with ankylosing spinal disorders: a systematic review of the literature on treatment, neurological status and complications. Eur Spine J.  2009; 18: 145- 156. Google Scholar CrossRef Search ADS PubMed  17. Sapkas G, Kateros K, Papadakis SA et al.   Surgical outcome after spinal fractures in patients with ankylosing spondylitis. BMC Musculoskeletal Disord  10: 96, doi:10.1186/1471-2474-10-96. CrossRef Search ADS   18. Denis F, Burkus JK. Shear fracture dislocation of the thoracic and lumbar spine associated with forceful hyperextension (lumberjack paraplegia). Spine . 1992; 17: 156- 161. Google Scholar CrossRef Search ADS PubMed  19. Joaquim AF, De Almeida Bastos DC, Jorge Torres HH et al.   Thoracolumbar injury classification and injury severity score system: a literature review of its safety. Global Spine J . 2016; 6( 1): 80- 85. Google Scholar CrossRef Search ADS PubMed  Copyright © 2017 by the Congress of Neurological Surgeons http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Operative Neurosurgery Oxford University Press

Radiographic and Neurological Outcome After Surgical Treatment of Traumatic Fractures of the Ankylotic Thoracic Spine: A Retrospective Case Series

Loading next page...
 
/lp/ou_press/radiographic-and-neurological-outcome-after-surgical-treatment-of-PtxnDpU0S0
Publisher
Oxford University Press
Copyright
Copyright © 2017 by the Congress of Neurological Surgeons
ISSN
2332-4252
eISSN
2332-4260
D.O.I.
10.1093/ons/opx099
Publisher site
See Article on Publisher Site

Abstract

Abstract BACKGROUND Spontaneous thoracic ankylosis is a progressive degenerative process that predisposes patients to potentially highly unstable traumatic injuries. Acute hyperextension fractures result in dynamic instability putting the spinal cord at risk. OBJECTIVE To describe preoperative radiographic characteristics of fractures of the ankylotic thoracic spine and relate findings to early postoperative radiographic and clinical outcomes. METHODS A single center, retrospective review was performed of 28 surgically treated patients with fractures of the ankylotic thoracic spine. Radiographic assessment included preoperative fracture angulation (FA) and fracture displacement (FD), and postoperative change in sagittal alignment. Early clinical outcomes included preoperative and postoperative American Spinal Injury Association (ASIA) grade and perioperative complications. RESULTS Seven patients (25%) presented with poor neurological grade (ASIA A-C) compared to 21 (75%) with good grade (ASIA D, E). At presentation, poor grade patients had a mean FA of 16.4° (range 0°-34.5°), and FD of 7.76 mm (range 0.8-9.2). Good grade patients had a mean FA of 18.2° (range 0°-43.3°), and FD of 4.77 mm (range 0-25.1). There was no statistically significant difference in FA or FD between groups (P = .70 and .20 respectively). All underwent posterior pedicle screw fixation for stabilization. Fifty per cent of patients presenting with ASIA C or D spinal cord injury improved 1 or more ASIA grades. There were no perioperative complications. Early postoperative sagittal alignment was maintained with a mean change of –2.6°. CONCLUSION Presenting fracture alignment does not significantly correlate with pre- or postoperative neurological status. Early posterior stabilization preserved neurological function, with neurological recovery occurring in a portion of individuals. Ankylosis, Extension-distraction injury, Fracture angulation, Fracture displacement, Hyperextension-spondylolysis, Posterior stabilization, Spinal cord injury ABBREVIATIONS ABBREVIATIONS ASIA American Spinal Injury Association CI confidence interval CT computed tomography FA fracture angulation FD fracture displacement FL fracture level IRB Institutional Review Board MVC motor vehicle collision SCI spinal cord injury SD standard deviation Spontaneous ankylosis of the spine is a chronic, degenerative disorder characterized by progressive autofusion across multiple motion segments of the vertebral column. Ankylosing spondylitis (AS) is one etiology of spinal ankylosis due to progressive enthesitis at the intervertebral disc space leading to syndesmophyte formation and eventual fusion;1-3 however, degenerative ankylosis can occur in the absence of AS secondary to calcification of the anterior or posterior longitudinal ligaments or disc space. Regardless of underlying pathogenesis, multilevel vertebral ankylosis, particularly at kyphotic segments, puts affected individuals at risk for severe hyperextension injury with spinal trauma.4 Fractures of the ankylotic cervical spine have been previously well described, mostly in patients with underlying AS.5 Increased risk of spinal cord injury (SCI) after AS-related cervical fractures is attributed to severe instability that occurs with hyperextension through the ankylosed “bamboo” spine. Immediate immobilization and/or stabilization of AS-related cervical spine fractures is critical to prevent further neurological injury from unstable fracture dislocation and dynamic mechanical compression of the spinal cord.6-8 While much has been published about cervical injuries, little exists in the literature describing treatment of traumatic fractures of the ankylotic thoracic spine and neurological outcomes. The thoracic spine is normally kyphotic, and therefore like AS-related cervical fracture, is at risk for worsening injury with hyperextension.9 In addition, because of the regional thoracic kyphosis, fracture in the setting of multilevel ankylosis may pose challenges in maintaining anatomic alignment and protection of the spinal cord when transferring an acute trauma patient. Paradoxically, positioning a patient supine (eg, flat on a trauma board) may worsen hyperextension injury through the fracture segment of an ankylotic thoracic spine, underscoring the importance of fracture stabilization. A recent thoracolumbar spine classification system characterized the potential instability of this hyperextension fracture in ankylotic disorders,10 with the recommendation that these fractures generally be treated surgically.11 To date, the existing literature reporting this fracture type in the setting of thoracic ankylosis or AS involves only small case series (<10 patients).12,13 Therefore, we report the largest known single center, case series of surgically treated fractures of the ankylotic thoracic spine with early radiographic and neurological outcomes. METHODS Study Design A single-center retrospective chart review was performed of surgically treated patients with traumatic fracture of the ankylotic thoracic spine. All patients underwent treatment by the neurosurgery department between 2010 and 2015. Institutional Review Board (IRB) approval was obtained prior to study initiation (IRB#: I201500874). In accordance with IRB policy, patient consent did not need to be obtained for this study. Patients were identified from operative case logs of the senior spine neurosurgeons. Medical records and radiological imaging studies were reviewed to select those with ankylosis of the thoracic spine, as defined by computed tomography (CT) evidence of continuous bridging calcification across ≥3 intervertebral discs. Imaging of the sacroiliac joints and HLA-B27 gene testing was not routinely performed to determine if AS was the underlying cause of ankylosis. Admission CT imaging was reviewed by a single observer to describe presenting fracture characteristics including fracture level (FL), AO classification fracture type,11,14 fracture angulation (FA), and anteroposterior fracture displacement (FD). All patients presented as a trauma alert and underwent a contrasted CT body scan, which included thin-slice, multiplanar imaging of the full thoracic and lumbar spine. FA was determined by measuring the degree of angulation of the fracture distraction (Figure 1A). FD was determined by measuring the horizontal distance (mm) between vertical lines drawn along the dorsal vertebral body of the rostral and caudal vertebra (Figure 1B). All radiological measurements were made using computer imaging software (Visage Imaging v7.1.6, Richmond, Australia). Change in sagittal alignment was determined by comparing the preoperative and postoperative Cobb angle formed by the inferior endplate of the adjacent vertebra cranial to the fracture level and the superior endplate of the adjacent caudal vertebra. FIGURE 1. View largeDownload slide Measurements of A, fracture angulation (FA) and B, fracture displacement (FD). The figure demonstrates measurements of FA (left) and FD (right) for this cohort of patients. FIGURE 1. View largeDownload slide Measurements of A, fracture angulation (FA) and B, fracture displacement (FD). The figure demonstrates measurements of FA (left) and FD (right) for this cohort of patients. Clinical data were obtained from individual patient medical charts. American Spinal Injury Association (ASIA) grade was determined by documentation of motor and sensory score at admission, postoperative day 1, and last follow-up. Surgical procedure performed including type of stabilization, and number of levels was determined from operative reports. Complications reviewed included postoperative neurological deterioration, reoperation, surgical site infection, postoperative hematoma, instrumentation failure, and mortality. Statistical Methods Association between spinal fracture radiographic characteristics and clinical outcomes was examined using descriptive statistics. Patients were grouped for analysis by either “poor ASIA grade” (ASIA A-C) or “good ASIA grade” (ASIA D, E). The relationship between average FA and FD to poor vs good ASIA grade at presentation and last follow-up were analyzed using Wilcoxon rank-sum tests to generate P values. This approach was used to account for variable length of follow-up in determining a relationship between fracture alignment and clinical outcome. All P values were considered significant if ≤.05. Interval-censored exponential survival regression was used to estimate the average rate of neurological improvement in ASIA grade from initial presentation to last follow-up. This improvement rate was then tested for association with FA and FD, with a confidence interval (CI) set at 95%. RESULTS Twenty-eight patients with traumatic fracture of the ankylotic thoracic spine were identified. Individual patient clinical characteristics are listed in Table. The mean age was 69.1 ± 9.95 yr (standard deviation [SD]) and 18 (64%) were male. All fractures were classified as type B3 hyperextension injuries with disruption of the anterior longitudinal ligament and through the disc space or vertebral endplate. TABLE. General Patient Characteristics of the 28 Individuals in the Study Cohort Patient number  Age /sex  Time to OR (d)  Length of stay  Fracture level  Other trauma present  Medical morbidities  Fracture angulation (°)  Fracture displacement (mm)  Procedure levels and (total no. of ankylosed levels)  No. of levels stabilized (change in sagittal alignment degrees)  Follow-up (d)  ASIA grade presentation  ASIA grade last follow-up  Mortality (Y/N)  1  72/M  2  13  T12-L1  N  DM  23.5  4.4  T8-L5 (9)  6 (–2.6)  13  A  A  N  2  64/M  0  5  T11-12  N  CAD, CHF, Afib on coumadin, DM  19.4  0  T10-L1 (11)  3 (–12)  158  E  E  N  3  64/M  0  1  T6-7  Pneumothorax  Afib on coumadin, Pacemaker  17.3  25.1  T3-11 (14)  8 (–0.6)  24  D  D  N  4  79/M  0  28  T8-9  Rib fx  HTN, ASA 325  19  1.6  T7-10 (13)  3 (–6.0)  58  E  E  N  5  81/M  7  7  T7  Rib fx, pulmonary contusion  HTN, Afib, CAD, bladder cancer  16.6  19.2  T5-10 (9)  5 (–2.8)  49  A  A  Y  6  65/F  2  13  T10-L1  N  HTN, obesity  23.3  7.1  T7-L4 (14)  9 (0)  178  E  E  N  7  68/M  2  6  T6  Pneumothorax  ASA 325 mechanical aortic valve  8.6  4.5  T4-8 (11)  4 (–5.6)  107  E  E  N  8  83/M  0  6  T4  N  Afib on coumadin, Pacer, CAD  6.9  0.8  T2-6 (13)  4 (–0.3)  19  A  A  Y  9  59/F  4  9  T9  Falcine SDH, Pelvic fx  Sickle cell anemia, DM, HTN  16.8  3.8  T7-11 (10)  4 (1.2)  104  E  E  N  10  67/M  4  12  T11-12  Pelvic fx, R tibia/fibula fx  Prostate cancer  24.7  8.5  T10-L1 (12)  3 (4.5)  8  C  C  N  11  62/M  0  4  T4-5  N  HTN, HLD  14.2  5.3  T2-6 (10)  4 (3.1)  281  E  E  N  12  62/M  1  8  T10-11  N  HTN, HLD, PE on xarelto  8.5  9.8  T9-12 (5)  3 (1.2)  234  C  D  N  13  58/F  2  14  T3-4  Hemothorax, R radius fx  HTN, HLD, COPD, DM  0  9.8  T1-7 (13)  6 (0.0)  488  C  E  N  14  73/F  5  21  T11-12  N  Afib on coumadin, HTN, HLD, OSA, CAD, CKD  34.5  1.8  T9-L2 (12)  5 (–16.2)  16  C  D  Y  15  76/F  6  11  T12  N  HTN, HLD, DM  19.7  4.1  T10-L3 (5)  5 (1.9)  215  E  E  N  16  72/M  2  6  T6-7  R occipital SDH  None  10.5  1.3  T4-9 (11)  5 (–1.2)  36  E  E  N  17  59/F  1  13  T3-4  N  HTN, HLD, hypothyroid, DM  14.3  10.5  T1-7 (14)  5 (0.0)  486  D  E  N  18  65/M  0  5  T11  N  HTN, DM  NA  6.50  T10-L1 (5)  3 (–9.6)  5  E  E  N  19  42/F  6  14  T7-8  N  HTN, DM, urinary incontinence  21.5  0  T4-11 (4)  7 (–4.6)  157  E  E  N  20  61/M  1  25  T12-L1  R/L femur and tibia/fibula fx  DM, Hep C  43.3  5  T11-L3 (6)  4 (5.3)  952  D  E  N  21  66/M  5  10  T9-T10  N  None  14.2  6.5  T5-L2 (12)  9 (–1.5)  53  E  E  N  22  89/F  2  12  T9-T10  N  HF, HTN, DM  24.9  2.8  T6-12 (12)  6 (–3.7)  10  E  E  N  23  69/F  5  13  T3-T4  N  CAD on daily ASA/plavix  22.4  4.6  T1-6 (8)  5 (–1.7)  8  D  D  N  24  82/M  0  4  T11  N  HTN, HLD, DM  8.6  2.4  T9-L2 (21)  5 (6.4)  241  E  E  N  25  84/M  0  4  T10  R tibia/fibula fx  Afib, HTN, HLD, CAD, DM  34.6  0  T8-12 (14)  4 (–4.3)  96  E  E  N  26  73/M  1  18  T11  Pelvic fx  None  0  1.2  T9-L1 (12)  4 (0.9)  45  D  D  N  27  69/M  4  8  T9  Pneumothorax  DM  10.8  0.8  T8-11 (15)  3 (–6.6)  53  E  E  N  28  72/F  3  8  T12  N  Afib, DVT on coumadin, HF, HTN, HLD  19.8  7  T8-L4 (20)  8 (–18)  51  D  D  N  Patient number  Age /sex  Time to OR (d)  Length of stay  Fracture level  Other trauma present  Medical morbidities  Fracture angulation (°)  Fracture displacement (mm)  Procedure levels and (total no. of ankylosed levels)  No. of levels stabilized (change in sagittal alignment degrees)  Follow-up (d)  ASIA grade presentation  ASIA grade last follow-up  Mortality (Y/N)  1  72/M  2  13  T12-L1  N  DM  23.5  4.4  T8-L5 (9)  6 (–2.6)  13  A  A  N  2  64/M  0  5  T11-12  N  CAD, CHF, Afib on coumadin, DM  19.4  0  T10-L1 (11)  3 (–12)  158  E  E  N  3  64/M  0  1  T6-7  Pneumothorax  Afib on coumadin, Pacemaker  17.3  25.1  T3-11 (14)  8 (–0.6)  24  D  D  N  4  79/M  0  28  T8-9  Rib fx  HTN, ASA 325  19  1.6  T7-10 (13)  3 (–6.0)  58  E  E  N  5  81/M  7  7  T7  Rib fx, pulmonary contusion  HTN, Afib, CAD, bladder cancer  16.6  19.2  T5-10 (9)  5 (–2.8)  49  A  A  Y  6  65/F  2  13  T10-L1  N  HTN, obesity  23.3  7.1  T7-L4 (14)  9 (0)  178  E  E  N  7  68/M  2  6  T6  Pneumothorax  ASA 325 mechanical aortic valve  8.6  4.5  T4-8 (11)  4 (–5.6)  107  E  E  N  8  83/M  0  6  T4  N  Afib on coumadin, Pacer, CAD  6.9  0.8  T2-6 (13)  4 (–0.3)  19  A  A  Y  9  59/F  4  9  T9  Falcine SDH, Pelvic fx  Sickle cell anemia, DM, HTN  16.8  3.8  T7-11 (10)  4 (1.2)  104  E  E  N  10  67/M  4  12  T11-12  Pelvic fx, R tibia/fibula fx  Prostate cancer  24.7  8.5  T10-L1 (12)  3 (4.5)  8  C  C  N  11  62/M  0  4  T4-5  N  HTN, HLD  14.2  5.3  T2-6 (10)  4 (3.1)  281  E  E  N  12  62/M  1  8  T10-11  N  HTN, HLD, PE on xarelto  8.5  9.8  T9-12 (5)  3 (1.2)  234  C  D  N  13  58/F  2  14  T3-4  Hemothorax, R radius fx  HTN, HLD, COPD, DM  0  9.8  T1-7 (13)  6 (0.0)  488  C  E  N  14  73/F  5  21  T11-12  N  Afib on coumadin, HTN, HLD, OSA, CAD, CKD  34.5  1.8  T9-L2 (12)  5 (–16.2)  16  C  D  Y  15  76/F  6  11  T12  N  HTN, HLD, DM  19.7  4.1  T10-L3 (5)  5 (1.9)  215  E  E  N  16  72/M  2  6  T6-7  R occipital SDH  None  10.5  1.3  T4-9 (11)  5 (–1.2)  36  E  E  N  17  59/F  1  13  T3-4  N  HTN, HLD, hypothyroid, DM  14.3  10.5  T1-7 (14)  5 (0.0)  486  D  E  N  18  65/M  0  5  T11  N  HTN, DM  NA  6.50  T10-L1 (5)  3 (–9.6)  5  E  E  N  19  42/F  6  14  T7-8  N  HTN, DM, urinary incontinence  21.5  0  T4-11 (4)  7 (–4.6)  157  E  E  N  20  61/M  1  25  T12-L1  R/L femur and tibia/fibula fx  DM, Hep C  43.3  5  T11-L3 (6)  4 (5.3)  952  D  E  N  21  66/M  5  10  T9-T10  N  None  14.2  6.5  T5-L2 (12)  9 (–1.5)  53  E  E  N  22  89/F  2  12  T9-T10  N  HF, HTN, DM  24.9  2.8  T6-12 (12)  6 (–3.7)  10  E  E  N  23  69/F  5  13  T3-T4  N  CAD on daily ASA/plavix  22.4  4.6  T1-6 (8)  5 (–1.7)  8  D  D  N  24  82/M  0  4  T11  N  HTN, HLD, DM  8.6  2.4  T9-L2 (21)  5 (6.4)  241  E  E  N  25  84/M  0  4  T10  R tibia/fibula fx  Afib, HTN, HLD, CAD, DM  34.6  0  T8-12 (14)  4 (–4.3)  96  E  E  N  26  73/M  1  18  T11  Pelvic fx  None  0  1.2  T9-L1 (12)  4 (0.9)  45  D  D  N  27  69/M  4  8  T9  Pneumothorax  DM  10.8  0.8  T8-11 (15)  3 (–6.6)  53  E  E  N  28  72/F  3  8  T12  N  Afib, DVT on coumadin, HF, HTN, HLD  19.8  7  T8-L4 (20)  8 (–18)  51  D  D  N  N = no, R = right, L = left, fx = fracture, SDH = subdural hematoma, Afib = atrial fibrillation, ASA = aspirin, CAD = coronary artery disease, CHF = congestive heart failure, COPD = chronic obstructive pulmonary disease, DM = diabetes, DVT = deep venous thrombosis, Hep C = hepatitis C, HF = heart failure (unspecified), HLD = hyperlipidemia, HTN = hypertension, PE = pulmonary embolism, OSA = obstructive sleep apnea. View Large Descriptive Data of Patient Cohort The average duration of last clinical follow-up was 5 mo (range: <1 to 16.2 mo). All underwent posterior pedicle screw–rod stabilization, at an average 2.3 d from time of injury (range: 0-7). Mean number of instrumented levels was 5 (range 3-9 levels). The mean number of ankylosed segments was 11.25 ± 4.1 (SD) across the thoracic and lumbar spine. All underwent postoperative X-rays for evaluation of fracture sagittal alignment and instrumentation. Fourteen of the 28 patients additionally suffered nonspine-related trauma including upper and/or lower extremity fractures, pelvic fractures, or traumatic brain injury (see Table). Complications There were no reported surgery-related complications of postoperative infection, hematoma, instrumentation failure or screw misplacement, reoperation, myocardial infarction, acute respiratory failure, urinary tract infection, deep vein thrombosis, or pulmonary embolus. Three patients died during the follow-up period due to nonneurological and nonsurgical complications related to the initial presenting trauma. One patient was an 81-yr-old male who presented after motor vehicle collision (MVC) with multisystem injuries including a positive focused assessment with sonography in trauma scan, T7 ASIA A SCI, and traumatic intracranial subarachnoid hemorrhage with intraventricular extension. He underwent spine surgical treatment 5 d after initial presentation. Ultimately, the patient was unable to be weaned off pressors or dialysis due to acute kidney injury and cardiac trauma, and family withdrew care on postoperative day 11. The second death was an 83-yr-old male presenting after MVC who, despite report of ambulating after the accident, presented to the hospital with a T2 ASIA A SCI. He ultimately expired over 2 yr after surgery due to unrelated cardiac causes. The third death was a 73-yr-old female presenting after ground level fall with a T11-T12 fracture. She had an extensive medical history including coronary artery disease, sleep apnea, and atrial fibrillation. She underwent posterior fusion 5 d after presentation. She was discharged to acute inpatient rehabilitation on postoperative day 16 in stable condition after which documentation ceased and the patient was marked in the medical record as deceased. Radiographic and Neurological Outcome Individual presenting fracture angulation and displacement are presented in Table. Seven patients (25%) presented with poor ASIA grade (ASIA A-C) compared to 21 (75%) with good ASIA grade (ASIA D, E; Figure 2). Poor ASIA grade patients presented with fracture levels ranging from T1-T12, and mean FA of 16.4° (range 0°-34.5°) and FD of 7.76 mm (range 0.8-9.2). Good ASIA grade patients presented with fracture levels ranging from T3-12, and mean FA of 18.2° (range 0°-43.3°), and FD of 4.77 mm (range 0-25.1). There was no significant difference in FA or FD between poor and good ASIA grade groups (P = .70 and .20, respectively). Change in sagittal alignment (negative number represents increased kyphosis postoperative vs preoperative) is represented for each individual patient in Table. The average change in postoperative sagittal alignment for all patients was –2.6° ± 5.9° (SD). FIGURE 2. View largeDownload slide Graphical representation of admission AIS and last follow-up AIS for 28 patients. The purple line for ASIA E represents 15 patients. AIS, American Spinal Injury Association and International Medical Society of Paraplegia Impairment Scale . FIGURE 2. View largeDownload slide Graphical representation of admission AIS and last follow-up AIS for 28 patients. The purple line for ASIA E represents 15 patients. AIS, American Spinal Injury Association and International Medical Society of Paraplegia Impairment Scale . At last follow-up, 24 (85.7%) were ASIA D or E, compared to 75% at initial presentation. All ASIA A patients at presentation (n = 3) remained ASIA A at last follow-up. No patient worsened neurologically during the follow-up period. Ten patients presented with motor incomplete SCI (ASIA C, D), of which 5 (50%) improved at least 1 ASIA grade. Of these, 1 ASIA C and 2 ASIA D patients recovered normal neurological function (ASIA E). The improvement rate for ASIA C and D patients was 3.95 events/person-year of follow-up (95% CI 0.95-16.42). The improvement rate was not significantly related to either admission FA or FD, P = .17 and .81, respectively. DISCUSSION Progressive multisegmental autofusion of the ankylotic thoracic spine predisposes patients to potentially devastating hyperextension injuries. The altered biomechanics in diffuse ankylosis do not allow for compensation of normal physiologic loading by distributing flexion-extension and translational forces across multiple motion segments. Particularly, ankylosis across a kyphotic region of the spine predisposes to potentially catastrophic anterior tension band failure when subjected to extension loads.13,14,15 Instability at the affected segment in thoracic ankylosis is magnified by the long functional lever arms created by the multisegment fused vertebrae above and below the fracture. In our series, we observed an average of >11 ankylosed motion segments across the thoracic and lumbar spine. This instability is subject to ongoing dynamic forces that put the spinal cord at risk until adequately stabilized. Therefore, prompt recognition of these injuries, appropriate immobilization in anatomic alignment, and ultimately surgical stabilization is critical for preservation of neurological function. We reviewed our series of 28 patients with fractures of the ankylotic thoracic spine treated by a single neurosurgery department to demonstrate clinical and radiographic characteristics of these unique injuries. To our knowledge, this represents the largest surgical case series to date of patients with fractures in the setting of thoracic ankylosis.13,16 Nearly half (46.4%) of the patients in our series presented with neurological deficits, which is comparable to previous series and attests to the instability and spinal cord risk of these injuries.17 Surgical treatment with posterior pedicle screw–rod stabilization was not only protective of neurological function (none deteriorated during follow-up), but may facilitate recovery of function. Half of the motor incomplete ASIA C and D patients on presentation improved 1 or more ASIA grades during the follow-up period. Based on this high recovery rate, we suspect that fractures of the ankylotic thoracic spine result in hypermobile, dynamic spinal cord compression, which in some cases have shown potential for recovery of neurological function with surgical stabilization. Furthermore, this benefit with surgical treatment can be performed safely with relatively low morbidity and risk of complications. Paradoxical Relation of Fracture Geometry to Clinical Outcome Despite the high rate of neurological recovery in this surgical series, a significant proportion of patients did not improve. Fifty-seven per cent (4 of 7) that were ASIA A-C at presentation remained poor SCI grade at last follow-up. Of note, 3 of 4 patients presented as ASIA A. Therefore, we sought to determine if admission fracture characteristics predicted neurological function at presentation, and likelihood of recovery. All fractures were consistent with a type B3, hyperextension injury with disruption through the anterior longitudinal ligament and disc space, and a characteristic “fish mouth” widening anteriorly. In our series, degree of fracture angulation did not correlate with either ASIA grade at presentation or likelihood of recovery. Extent of fracture displacement also did not correlate with neurological presentation or recovery, although there was a nonsignificant mean 2.99 mm greater retrolisthesis in the poor ASIA grade patients compared to good ASIA grade group. This unexpected lack of correlation between radiographic fracture characteristics with ASIA grade is counter to other thoracolumbar fractures in which extent of fracture displacement relates to severity of neurologic impairment.11,18,19 A possible explanation in this unique fracture type is that hyperextension injury in ankylosis causes hypermobile instability at a normally kyphotic region of the spine. Therefore, significant dynamic changes in the fracture pattern may occur depending on postural changes during positioning, including the potential for relatively benign appearing, minimal displacement when positioned upright. Head-of-bed elevated positioning may promote reduction of the fracture into normal anatomic kyphotic alignment, whereas supine positioning (ie, thoracic extension) may cause greater hyperextension angulation and/or displacement with possible worsening spinal cord compression. This repetitive dynamic dislocation-reduction pattern with variable patient positioning may explain the lack of relationship between fracture alignment on imaging study and neurological presentation. This observation speaks to the importance of prompt recognition of fractures of the ankylotic thoracic spine, as the severity of instability and neurological injury may not be readily apparent based on fracture alignment during a single imaging study. Furthermore, special consideration should be made when positioning that may place the injured spine in hyperextension (eg, flat, supine for CT or MRI). When obtaining supine imaging, care should be made to optimally maintain anatomic alignment of the patient's normal thoracic kyphosis with clinical assessment for any neurological changes. Limitations There are several limitations to this study. The retrospective design relied on medical chart and imaging review for data collection, with inherent potential for limited follow-up. As a tertiary referral center, it is common for acute trauma patients to be transferred hospital–hospital for neurosurgical evaluation and treatment, and then return to the initial admitting hospital for subsequent care. Therefore, radiographic and clinical follow-up is often limited to early assessment during the acute perioperative period. Despite this, we observed that neurological outcome was preserved after surgery, with a high portion (50%) of motor incomplete SCI patients demonstrating recovery of 1 or more ASIA grade. Second, the diagnosis of thoracic ankylosis was made using CT imaging of the thoracic spine for evidence of continuous, multilevel anterior longitudinal ligament and disc space calcification. While this imaging finding was typical of the characteristic “bamboo spine” of AS, the definitive diagnosis of AS was not determined by HLA-B27 gene testing. Regardless, the relevant anatomy and biomechanics of the ankylotic thoracic spine in this case series was likely similar to that observed in AS. Last, this case series represents the preferred management strategy of the senior spine neurosurgeons, which is to treat these fractures during the presenting hospitalization with posterior pedicle screw–rod fixation.11 Since this was a retrospective surgical case series, reasonable conclusions about nonoperative external immobilization for these fractures cannot be made, although conventional orthoses that tend to place the spine in extension may be inadequate. Therefore, conclusions regarding the benefit of nonoperative vs operative, or various operative techniques for this particular fracture type, and their potential impact on neurological recovery and clinical outcome cannot be made. CONCLUSION Traumatic injuries of the ankylotic thoracic spine result in unstable fracture angulation and displacement with potential for significant SCI. Severity of fracture dislocation does not appear to correlate with neurological function or likelihood of recovery. Early surgical posterior stabilization is observed to preserve neurological function and improve recovery in a portion of individuals presenting with incomplete SCI, with a relatively low risk of perioperative complications. Disclosure The authors have no personal, financial, or institutional interest in any of the drugs, materials, or devices described in this article. REFERENCES 1. Braun J, Sieper J. Ankylosing spondylitis. Lancet . 2007; 369( 9570): 1379- 1390. Google Scholar CrossRef Search ADS PubMed  2. Mcgonagle D, Gibbon W, Emery P. Classification of inflammatory arthritis by enthesitis. Lancet . 1998; 352( 9134): 1137- 1340. Google Scholar CrossRef Search ADS PubMed  3. Vosse D, Feldtkeller E, Erlendsson J et al.   Clinical vertebral fractures in patients with ankylosing spondylitis. J Rheumatol . 2004; 31( 10): 1981- 1985. Google Scholar PubMed  4. Ulu MA, Batmaz İ, Dilek B et al.   Prevalence of osteoporosis and vertebral fractures and related factors in patients with ankylosing spondylitis. Chin Med J . 2014; 127( 15): 2740- 2747. Google Scholar PubMed  5. Mitra D, Elvins DM, Speden DJ et al.   The prevalence of vertebral fractures in mild ankylosing spondylitis and their relationship to bone mineral density. Rheumatology (Oxford) . 2000; 39( 1): 85- 89. Google Scholar CrossRef Search ADS PubMed  6. Lukasiewicz AM, Bohl DD, Varthi AG et al.   Spinal fracture in patients with ankylosing spondylitis: cohort definition, distribution of injuries, and hospital outcomes. Spine . 2016; 41( 3): 191- 196. Google Scholar CrossRef Search ADS PubMed  7. Ma J, Wang C, Zhou X et al.   Surgical therapy of cervical spine fracture in patients with ankylosing spondylitis. Medicine (Baltimore) . 2015; 94( 44): e1663. Google Scholar CrossRef Search ADS PubMed  8. Lazennec JY, D’astorg H, Rousseau MA. Cervical spine surgery in ankylosing spondylitis: review and current concept. Orthop Traumatol Surg Res . 2015; 101( 4): 507- 513. Google Scholar CrossRef Search ADS PubMed  9. Geusens P, Vosse D, Van der heijde D et al.   High prevalence of thoracic vertebral deformities and discal wedging in ankylosing spondylitis patients with hyperkyphosis. J Rheumatol . 2001; 28( 8): 1856- 1861. Google Scholar PubMed  10. Vaccaro AR, Oner C, Kepler CK et al.   AOSpine thoracolumbar spine injury classification system: fracture description, neurological status, and key modifiers. Spine . 2013; 38( 23): 2028- 2037. Google Scholar CrossRef Search ADS PubMed  11. Vaccaro AR, Schroeder GD, Kepler CK et al.   The surgical algorithm for the AOSpine thoracolumbar spine injury classification system. Eur Spine J . 2016; 25( 4): 1087- 1094. Google Scholar CrossRef Search ADS PubMed  12. El tecle NE, Abode-iyamah KO, Hitchon PW et al.   Management of spinal fractures in patients with ankylosing spondylitis. Clin Neurol Neurosurg . 2015; 139: 177- 182. Google Scholar CrossRef Search ADS PubMed  13. Lu ML, Tsai TT, Lai PL et al.   A retrospective study of treating thoracolumbar spine fractures in ankylosing spondylitis. Eur J Orthop Surg Traumatol . 2014; 24 ( suppl 1): S117- S123. Google Scholar CrossRef Search ADS PubMed  14. Magerl F, Aebi M, Gertzbein SD et al.   A comprehensive classification of thoracic and lumbar injuries. Eur Spine J . 1994; 3( 4): 184- 201. Google Scholar CrossRef Search ADS PubMed  15. Gelman MI, Umber JS. Fractures of the thoracolumbar spine in ankylosing spondylitis. AJR Am J Roentgenol . 1978; 130( 3): 485- 491. Google Scholar CrossRef Search ADS PubMed  16. Westerveld LA, Verlaan JJ, Oner FC. Spinal fractures in patients with ankylosing spinal disorders: a systematic review of the literature on treatment, neurological status and complications. Eur Spine J.  2009; 18: 145- 156. Google Scholar CrossRef Search ADS PubMed  17. Sapkas G, Kateros K, Papadakis SA et al.   Surgical outcome after spinal fractures in patients with ankylosing spondylitis. BMC Musculoskeletal Disord  10: 96, doi:10.1186/1471-2474-10-96. CrossRef Search ADS   18. Denis F, Burkus JK. Shear fracture dislocation of the thoracic and lumbar spine associated with forceful hyperextension (lumberjack paraplegia). Spine . 1992; 17: 156- 161. Google Scholar CrossRef Search ADS PubMed  19. Joaquim AF, De Almeida Bastos DC, Jorge Torres HH et al.   Thoracolumbar injury classification and injury severity score system: a literature review of its safety. Global Spine J . 2016; 6( 1): 80- 85. Google Scholar CrossRef Search ADS PubMed  Copyright © 2017 by the Congress of Neurological Surgeons

Journal

Operative NeurosurgeryOxford University Press

Published: Mar 1, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off