Abstract
Background
Spinal tuberculosis is the most common form of skeletal tuberculosis. Various approaches have been described for surgical management of spinal tuberculosis, but many entail wide exposures with attendant morbidity; whether minimally invasive surgical (MIS) approaches are suitable is unknown.
Questions/purposes
We evaluated (1) neurologic results, (2) radiographic results, and (3) complications in patients with thoracic and lumbar spinal tuberculosis treated with two MIS approaches.
Methods
We retrospectively evaluated 22 patients with thoracic and lumbar tuberculosis managed surgically from October 2008 to February 2011 using MIS methods; one patient was lost to followup, leaving 21 patients with a minimum followup of 15 months (mean, 30 months; range, 15–59 months) for analysis. MIS approaches were used for patients with disease below D6 and minimum pedicle diameters of 4.5 mm to permit percutaneous screw placement. The MIS approach was divided into two groups depending on the extent of destruction of the vertebral body: a posterior-only group (n = 9), where posterior transpedicular decompression sufficed, and the hybrid group (n = 12), requiring anterior débridement and ventral-column reconstruction by conventional or mini-open thoracotomy. All but two patients with more than two contiguous bodies involvement underwent MIS posterior fixation by percutaneous transpedicular screws. Plain radiographs were evaluated for deformity correction and correction maintenance. Neurologic recovery and complications were ascertained by chart review.
Results
All patients with neurologic deficits recovered completely with no motor deficits at followup; 13% improved by three grades, 53% by two grades, and 33% by one grade. Mean correction was 2.5° (thoracic) and 8° (lumbar) in the posterior-only group and 4.2° in the hybrid group. Some correction loss occurred with healing (2° and 1.6° in the posterior-only and hybrid groups, respectively), but in none of those who had fixation did this progress to more than preoperative status. Two of 22 patients (9%) had complications. One had a malposition of L5 screw causing painful radiculopathy without motor deficit and required repositioning. The other had an intraoperative dural tear repaired by onlay fascial patch and cerebrospinal fluid diversion. There were no approach-related complications, neurologic deterioration, or implant fatigue at last followup.
Conclusions
We found evidence of neurologic recovery, avoidance of deformity progression, and few complications with these MIS approaches. Comparative trials are called for between open and MIS approaches for patients with spinal tuberculosis.
Level of Evidence
Level IV, therapeutic study. See Instructions for Authors for a complete description of levels of evidence.
Introduction
Tuberculosis of the spine, a paucibacillary disease, is one of the common forms of skeletal tuberculosis [18]. The management goals are to eradicate infection, prevent or treat neurologic deficits, and correct and avoid spinal deformity progression. Current surgical methods include anterior débridement, decompression, fusion followed by ventral or posterior instrumentation, and posterior decompression with or without dorsal fixation [9, 10, 13, 14, 20, 22, 23, 25, 27]. Conventional surgical approaches to the ventral aspect include thoracotomy with extrapleural or transpleural access, lateral extracavitary and costotransversectomy approaches for the thoracic spine, and retroperitoneal approaches for the lumbar spine [7, 20, 22, 25, 27]. These are extensive approaches with associated morbidity [15].
Minimally invasive surgical (MIS) approaches are increasingly being used in managing various spinal disorders, such as degenerative spine, trauma, and tumors [2, 3, 16, 21, 26, 28]. Video-assisted thoracoscopic anterior surgery (VATS) has been described as an MIS option for managing tuberculosis of the dorsal spine [10, 11, 15]. However, more research into whether these approaches are suitable for the management of spinal tuberculosis is needed.
We therefore evaluated (1) neurologic results, (2) radiographic results, and (3) complications in patients with tuberculosis of the thoracic and lumbar spine treated using either a posterior-only MIS approach or a hybrid MIS approach (conventional or mini-open anterior approach plus MIS posterior fixation).
Patients and Methods
Study Patients
In this retrospective study, we evaluated 22 patients with tuberculosis of the dorsal (below D6) and lumbar spine who were managed using MIS techniques between October 2008 and December 2011. During that same period, a total of 59 patients underwent various surgical procedures for tuberculosis involving the entire spine.
The protocol followed for managing these patients was the “middle path regimen” described by Tuli [30]. All patients with peridiscal involvement, preserved vertebral body height, and no neurologic deficits were managed by ambulatory antitubercular therapy (ATT) with an external orthosis and close clinical and radiographic followup. Surgery was considered in those with persistent severe disabling pain, radicular pain, neurologic deficits, and clinical and radiographic progression of disease despite ATT.
During the period in question, our general indications for the MIS approach were for disease involving the thoracic spine below D6 and the lumbar spine and minimum pedicle diameters of 4.5 mm. Level of disease was the most important factor in deciding between conventional and MIS approaches, in addition to limitations in available instrumentation and surgeon experience. Of the 59 patients, 37 underwent conventional procedures for the following reasons: level of disease (n = 26), nature of disease (n = 4), and hardware limitation (n = 7) (Fig. 1). Four patients had disease at the craniovertebral junction, 10 at the cervical spine, and four at the cervicothoracic spine. In the thoracic spine, eight patients with disease above the D6 level had conventional exposures due to difficulty and our inexperience in using mini-open retractors for exposure and insertion of percutaneous transpedicular screws at these levels. Four patients with en plaque type of tuberculosis with epidural granulation causing cord compression over multiple levels or predominant granulation situated dorsally without much destruction of the vertebral bodies underwent conventional hemilaminectomy (one patient)/laminectomy (three patients). Seven patients had hardware limitations: three with pedicle diameters of less than 4.5 mm precluding usage of percutaneous screws and four in the early part of our series when the Longitude® System (Medtronic, Inc, Minneapolis, MN, USA) was not available to avoid using the lordotic precontoured rods of the Sextant® System (Medtronic, Inc) in the thoracic spine.
Fig. 1.
A flowchart illustrates the process of choosing conventional methods versus MIS methods and the posterior-only MIS approach versus the hybrid MIS approach.
Of the 22 patients included in this study, one patient was lost to followup after 3 months, leaving 21 patients with a minimum followup of 15 months (mean, 30 months; range, 15–59 months) for analysis. Mean age was 44 years (range, 18–75 years).
Medical Management
The presenting symptoms were back pain, radicular pain, and myelopathy (Tables 1, 2). We evaluated neurologic function using American Spinal Injury Association (ASIA) grading [4] and myelopathy using Nurick’s grading [19]. ASIA grades include Grade A representing complete sensory or motor function impairment below the affected spinal cord segment, Grade B representing complete motor impairment, Grade C representing incomplete (severe) motor impairment, Grade D representing incomplete (less severe) motor impairment, and Grade E representing normal sensory and motor function. Nurick’s grades include Grade 0 representing intact, mild radiculopathy without myelopathy, Grade 1 representing mild myelopathy with no difficulty in walking, Grade 2 representing mild to moderate myelopathy with slight difficulty in walking that does not prevent full-time employment, Grade 3 representing moderate myelopathy with difficulty in walking that prevents full-time employment or ability to do household work but not so severe as to require someone’s help to walk, Grade 4 representing moderate to severe myelopathy with ability to walk with someone’s help or with a frame, and Grade 5 representing severe myelopathy leaving individuals chairbound or bedridden.
Table 1.
Clinical details of patients in the posterior-only group (n = 10, one lost to followup)
| Patient | Age (years) | Sex | Clinical picture | Level | Procedure | Followup (months) | Neurologic status |
|---|---|---|---|---|---|---|---|
| 1 | 47 | Female | Back pain, paraparesis, Grade C* | D11-L1 | MIS transpedicular decompression only | 53 | No deficits, Grade E |
| 2 | 22 | Male | Back pain, paraparesis, Grade C, progression on ATT for 3 months | Multiple thoracic (Fig. 3) | MIS transpedicular decompression only | 3 | Residual Grade 2† spasticity; lost to followup |
| 3 | 53 | Female | Back pain, paraparesis, Grade C | D11-D12 | MIS transpedicular decompression and PCS fixation | 55 | No deficits, Grade E |
| 4 | 55 | Female | Back pain, paraparesis, Grade C | D11-D12 | MIS transpedicular decompression and PCS fixation | 47 | No deficits, Grade E |
| 5 | 62 | Female | Back pain, paraparesis, Grade D | D9-D10 | MIS transpedicular decompression (endoscope assisted) and PCS fixation | 25 | Improved; no motor deficits, Grade E; residual Grade 2 spasticity |
| 6 | 75 | Male | Back pain, paraplegia, Grade C | D8-D9 (Figs. 4, 5) | MIS transpedicular decompression (endoscope assisted) and PCS fixation | 29 | Improved; no motor deficits, Grade E; residual Grade 2 spasticity |
| 7 | 74 | Male | Back pain, bilateral L5 radiculopathy | L4-L5 | MIS hemilaminotomy, decompression and PCS fixation | 59 | No radicular pain |
| 8 | 42 | Female | Back pain, bilateral L5 radicular pain, progression on ATT | L3-L4 | MIS hemilaminotomy, decompression, and PCS fixation | 52 | No pain |
| 9 | 29 | Female | Back pain, bilateral L5 radiculopathy | L4-L5 (Fig. 6) | MIS hemilaminotomy, decompression, and PCS fixation | 44 | No pain |
| 10 | 52 | Female | Back pain, radiculopathy | L1-L2 | MIS transpedicular decompression, and PCS fixation | 15 | Improved; no deficits |
* American Spinal Injury Association grade for neurologic deficit (Grade A–E); †Nurick’s grade for spasticity (Grade 0–5); ATT = antitubercular therapy; MIS = minimally invasive surgery; PCS = percutaneous transpedicular screw.
Table 2.
Clinical details of patients in the hybrid group (n = 12)
| Patient | Age (years) | Sex | Clinical picture | Level | Procedure | Followup (months) | Neurologic status |
|---|---|---|---|---|---|---|---|
| 1 | 55 | Female | Back pain, paraparesis, Grade D* | D10-D11 | DLT, RPl, corpectomy, fusion with rib graft, and PCS fixation | 48 | Improved; no deficits, Grade E |
| 2 | 34 | Male | Back pain, paraparesis, Grade D | D10-D11 | DLT, RPl, corpectomy, fusion with iliac crest graft, and PCS fixation | 40 | Improved; no deficits, Grade E |
| 3 | 42 | Male | Back pain | D10 | DLT, RPl, corpectomy, fusion with cage, and PCS fixation | 26 | Improved; no deficits, Grade E |
| 4 | 40 | Female | Back pain, paraparesis, Grade C | D10-D11 | DLT, RPl, corpectomy, fusion with iliac crest graft, and PCS fixation | 22 | Improved; no deficits, Grade E |
| 5 | 45 | Male | Back pain, paraparesis, Grade B | D11-D12 | DLT, RPl, corpectomy, fusion with cage, and PCS fixation | 20 | Improved; no deficits; Grade E, persistent back pain |
| 6 | 30 | Female | Back pain, paraparesis Grade D | D9-D10 | DLT, RPl approach, decompression, fusion with iliac crest graft, and PCS fixation | 18 | Improved; no deficits, Grade E |
| 7 | 25 | Male | Back pain, deformity (gibbus) | L2-L3 | Retroperitoneal approach, corpectomy, fusion with cage, and PCS fixation | 18 | Improved; no deficits; occasional back pain |
| 8 | 41 | Female | Back pain, paraparesis, Grade C | D10-D11 | MIS thoracotomy using tubular retractors, fusion with iliac crest graft, and PCS fixation | 18 | Improved; no deficits, Grade E |
| 9 | 18 | Female | Back pain, paraparesis, Grade C | D9-D10 (Fig. 8) | MIS thoracotomy using tubular retractors, fusion with iliac crest graft, and PCS fixation | 17 | Improved; no deficits, Grade E |
| 10 | 26 | Female | Back pain, paraparesis, Grade D | D8-D10 | MIS thoracotomy using tubular retractors, decompression and uninstrumented fusion with iliac crest graft | 17 | Improved; no deficits, Grade E |
| 11 | 31 | Female | Back pain, paraparesis, Grade C | D7-D8 | Mini-thoracotomy, decompression, fusion with iliac crest graft, and PCS fixation | 16 | Improved; Grade E, residual Grade 2 spasticity; walking with support |
| 12 | 65 | Male | Back pain, paraparesis, Grade B | D11-D12 | Mini-thoracotomy, decompression, fusion with iliac crest graft, and PCS fixation | 15 | Improved; no deficits, Grade E |
* American Spinal Injury Association grade for neurologic deficit (Grade A–E); †Nurick’s grade for spasticity (Grade 0–5); DLT = dorsolateral thoracotomy; RPl = retropleural approach; PCS = percutaneous transpedicular screw; MIS = minimally invasive surgery.
Initial imaging included radiographs, MRI with contrast, and CT of the spine to identify the level of the disease, extent of vertebral body destruction and collapse, degree of deformity (Cobb’s angle), location of the abscess and granulation tissue, and extent and location of thecal sac compression. Followup MRI scans were taken at 3 months of treatment to assess adequacy of response to medical treatment and to monitor disease progression and at 12 months to assess healing of tuberculosis and to decide about continuation of ATT. The scans were done early if the patient did not improve or deteriorated clinically. Healing was defined as complete resolution of vertebral body edema and resolution of prevertebral, paravertebral, and epidural granulation. ATT consisted of four drugs (isoniazid 5 mg/kg, rifampicin 10 mg/kg, ethambutol 15 mg/kg, and pyrazinamide 25 mg/kg) for 3 months followed by two drugs (isoniazid and rifampicin) for 9 to 12 months depending on the resolution of the disease based on MRI findings. Second-line ATT was initiated in those with multidrug-resistant strains based on culture reports or radiographic progression on first-line agents. Routine histopathologic and microbiologic analysis was performed on the pus and necrotic granulation material. This included acid-fast bacillus staining, bacterial culture to rule out secondary infection of cold abscess, and Bactec™ culture (Becton Dickinson and Co, Franklin Lakes, NJ, USA) for Mycobacteriumtuberculosis.
Four patients were on ATT for 1 to 3 months before surgery. They underwent surgical intervention with onset of neurologic deficits. The mean duration of ATT was 13.1 months. All patients had resolution of the disease on MRI after which ATT was stopped.
Surgical Approaches
The patients were divided into two groups based on MIS approach: (1) the posterior-only group (n = 10; one lost to followup) (Table 1) and (2) the hybrid group (n = 12) (Table 2). The main criterion for choosing the type of surgery was the extent of destruction of the vertebral body. Patients with relatively preserved vertebral body heights (< 25% collapse) underwent the first procedure, an entirely posterior approach. All of these patients had significant epidural collection causing cord compression and neurologic deficits or severe radiculopathy but no significant vertebral body collapse to require reconstruction of the ventral vertebral column. The primary aim was to decompress the spinal cord to hasten neurologic recovery. Internal fixation was performed to provide immediate stability, allow for early mobility, and avoid progression of deformity during healing. Patients with more than 25% collapse required ventral column reconstruction additionally and underwent the second procedure.
Posterior-only MIS Technique
The posterior-only MIS technique involved MIS transpedicular débridement (for the thoracic level) and MIS hemilaminotomy and débridement (for the lumbar spine) followed by internal fixation using percutaneous transpedicular screws (Fig. 2). The patient was placed prone on bolsters with care taken to keep the abdomen free. The level to be decompressed was localized with a C-arm. There were differences with respect to the procedure when performed at the thoracic and lumbar regions.
Fig. 2A–F.
MIS transpedicular decompression and percutaneous fixation (posterior-only MIS method) are illustrated. (A) The X-Tube is positioned. (B) The position of the X-Tube is confirmed with the C-arm. (C) An intraoperative microscopic view after decompression of the thecal sac is shown. (D) An endoscopic view allows better visualization of the ventrolateral aspect of the dura using the 30° forward-angled 4-mm telescope. Asterisk = thecal sac. (E) Endoscope-assisted decompression of the ventral aspect of the thecal sac is performed using an angled curette. Arrow = necrotic bone. (F) A C-arm image shows the extent of decompression with angled curettes.
In the thoracic region, MIS decompression was achieved by transpedicular access. An oblique incision was placed paramedially 2.0 to 2.5 cm from the midline. The lumbodorsal fascia was incised, paraspinal muscles were split by passing sequential dilators, and then an expandable tubular access channel (METRx™ X-Tube; Medtronic, Inc; 25-mm diameter) was docked over the facet (Fig. 2A–B) at that level. The facet and pedicle were drilled sequentially. The medial and inferior margins of the pedicle were kept intact until the end and removed once the vertebral body was reached to avoid injury to the root and thecal sac. C-arm images were taken intermittently to confirm the trajectory. Intracorporeal and epidural decompression of the necrotic tissue and removal of the purulent material, loose granulation tissue, and necrotic bone fragments were performed (Fig. 2C). After maximal microscopic decompression, endoscope assistance could be used to visualize the ventral aspect of the thecal sac directly and further débridement continued using angled instruments (Fig. 2D–E). Adequate decompression could thus be achieved using a combination of the microscope and endoscope. After this, a thorough wash was performed and the wound was closed in layers.
In the lumbar spine, a hemilaminotomy was performed on the more symptomatic side. The ligamentum flavum was excised, the root retracted medially, and the disc space entered. Removal of pus, granulation tissue, and loose bone fragments was performed.
Patients 1 and 2 underwent minimally invasive transpedicular decompression only (no fixation) due to extensive disease (> two contiguous segments) (Fig. 3). Patients 3 to 10 underwent single-stage minimally invasive decompression and fixation. For fixation of the thoracic and thoracolumbar spine, the Longitude® System (Medtronic, Inc) was used (Figs. 4, 5), and for fixation of the lumbar spine, the Sextant® System (Medtronic, Inc) was used (Fig. 6). Implant removal was planned in patients with fixation at the dorsolumbar and lumbar levels in the presence of radiographic or clinical evidence of implant fatigue.
Fig. 3A–B.
Preoperative MR images show the spine of Patient 2 (posterior-only group) who underwent MIS decompression only. (A) A T1-weighted sagittal image with contrast shows extensive multilevel involvement of the spine. (B) A T1-weighted axial image shows epidural granulation and cord compression.
Fig. 4A–F.
Images illustrate the case of Patient 6 (posterior-only group). Preoperative T2-weighted (A) sagittal and (B) axial MR images show epidural and paravertebral abscess and granulation with cord compression and cord edema. Preoperative (C) sagittal and (D) axial CT scans show the extent of vertebral body destruction and necrotic bone. Postoperative (E) sagittal and (F) axial CT scans show the extent of bony removal and canal decompression achieved.
Fig. 5A–C.
Followup images are shown for the patient in Figure 4 (Patient 6). (A) A T2-weighted sagittal MR image shows healed tuberculosis. (B) Sagittal and (C) AP CT scans show no deformity progression.
Fig. 6A–E.
Images illustrate the case of Patient 9 (posterior-only group). Preoperative T1-weighted (A) sagittal and (B) axial MR images with contrast show significant epidural and paravertebral abscess and psoas collection. Photographs show the incisions (C) before and (D) after healing. (E) A lateral followup radiograph shows preserved spinal alignment at the healed stage.
Hybrid MIS Technique
The hybrid MIS technique involved ventral decompression and fusion performed in the lateral position by the transthoracic route (using two options: (1) dorsolateral thoracotomy and the extrapleural approach or (2) mini-thoracotomy [mini-open or using tubular retractor system] and the transpleural approach), drainage of abscess, débridement of granulation tissue and necrotic bone to achieve adequate cord decompression, ventral column reconstruction using iliac crest or rib autograft or titanium cage filled with autograft, and posterior fixation using percutaneous transpedicular screws in the same sitting (Fig. 7). The dorsolateral extrapleural approach was used more often in the lower dorsal spine and in the earlier part of the surgical series. With the availability of expanding tubular retractors with longer blades (Mars™ 3 V System; Globus Medical, Inc, Audubon, PA, USA; and the Direct Lateral Interbody Fusion® System; Medtronic, Inc), these were used more often for performing mini-thoracotomy later in the series (Figs. 7, 8). Decompression included corpectomy or conservative intracorporeal débridement of necrotic bone until healthy bone margins were encountered. This was supplemented with posterior fixation with posterior transpedicular screws inserted percutaneously in the prone position in the same sitting as a single stage.
Fig. 7A–D.
Intraoperative images illustrate mini-thoracotomy with expandable tubular retractors. (A) A three-blade retractor is positioned, with one blade retracting the lung. (B) Paravertebral pus is drained. (C) A defect (arrow) remains after débridement. (D) Fusion is performed with iliac crest autograft (arrow).
Fig. 8A–E.
Images illustrate the case of Patient 9 (hybrid group). (A) A preoperative T2-weighted sagittal MR image shows destruction of vertebral body, kyphosis, and epidural and paravertebral abscess with cord compression. (B) A preoperative sagittal CT scan shows sequestered bone compromising the canal. Postoperative (C) AP and (D) sagittal CT scans show the spine after intracorporeal decompression and iliac crest graft in situ and correction of deformity. (E) A lateral followup radiograph shows bony fusion without deformity progression.
Outcomes
The patients were evaluated clinically once a month until healing of the disease. A plain radiograph was taken at 6 weeks, 6 months, and 12 months after surgery. MRI scan was done at 3 months to assess the response to ATT and at 12 months to assess healing. Subsequently, the patients were asked to visit the outpatient clinic (by telephone call or letter) and plain radiographs of the spine were taken. At the followup clinical evaluations, the treating surgeon (NG) assessed neurologic recovery using ASIA grading [4] for neurologic function and Nurick’s score [19] for myelopathy. Deformity progression, fusion, and implant fatigue were monitored on the spine radiographs. To assess deformity, a radiologist (RV) measured Cobb’s angle in the sagittal plane on plain radiographs as the angle between the superior endplate of the uppermost affected vertebrae and inferior endplate of the lowermost involved vertebral body. No intraobserver reliability testing was performed. In addition, the treating surgeon (NG) performed a chart review to identify and document reoperations and complications, including procedure-related complications, such as wound infections and pulmonary complications, after surgery.
Results
All patients showed neurologic improvement from one to three grades: 33% improved one grade, 53% two grades, and 13% three grades (Table 3). In the posterior-only group (Table 1), five of six patients with thoracic tuberculosis presented with neurologic deficits preoperatively (four ASIA Grade C, one Grade D). At followup, all improved to Grade E with no residual motor deficits. Three of these five patients had residual spasticity (Nurick’s Grade 2) at followup. In the hybrid group (Table 2), of the 10 patients who presented with neurologic deficits, two had ASIA Grade B deficits, four Grade C, and four Grade D. At followup, all improved to Grade E with no deficits. One patient had residual spasticity (Nurick’s Grade 2). There were no cases of deterioration in neurologic function.
Table 3.
Neurologic outcomes (n = 15 patients)
| Group | American Spinal Injury Association grade (number of patients) | ||
|---|---|---|---|
| Preoperative | Postoperative | Followup | |
| Posterior-only (n = 5) | C (4) | D (4) | E (5) |
| D (1) | E (1) | ||
| Hybrid (n = 10) | B (2) | C (2) | E (10) |
| C (4) | D (6) | ||
| D (4) | E (2) | ||
Preoperative, immediately postoperative, and last followup Cobb’s angles for each patient in both groups are shown (Table 4). In the posterior-only group, the mean ± SD preoperative kyphotic angle was 13.3° ± 4.3° in thoracic spine and 32.8° ± 2.8° in the lumbar spine. Immediately postoperatively, these angles improved to 10.8° ± 2.8° and 24.3° ± 10°, representing mean corrections of 2.5° (18.9% correction) and 8.5° (24.4% correction) in the thoracic and lumbar spines, respectively. At last followup, these angles were 12.8° and 27°, representing mean losses of 2° and 2.7° in the thoracic and lumbar spine, respectively. In none of the eight patients who underwent fixation were the postoperative Cobb’s angles more than preoperative angles. In the two patients who underwent decompression only without fixation, these values were 11.5° preoperatively, 12.5° postoperatively, and 14.5° at last followup, with progression of deformity. In the hybrid group, the mean preoperative Cobb’s angle was 19.8° (range, 5°–50°), which was corrected to 15.6° (range, 7°–44°) immediately postoperatively, with a mean correction of 4.2° (21% correction). At last followup, the angle was 17.3°, with a mean correction loss of 1.6° (10.5%).
Table 4.
Radiographic outcomes
| Patient | Level | Cobb’s angle (°) | Followup (months) | ||
|---|---|---|---|---|---|
| Preoperative | Postoperative | Followup | |||
| Posterior-only group | |||||
| 1 | D11-L1 | 13 | 14 | 18 | 53 |
| 2 | D7-D8, D10-L2 | 10 | 11 | 11 | 3 |
| 3 | D11-D12 | 9 | 8 | 9 | 55 |
| 4 | D11-D12 | 13 | 9 | 12 | 47 |
| 5 | D9-D10 | 14 | 12 | 13 | 25 |
| 6 | D8-D9 | 17 | 14 | 17 | 29 |
| 7 | L4-L5 | 35 | 24 | 26 | 59 |
| 8 | L3-L4 | 30 | 14 | 18 | 52 |
| 9 | L4-L5 | 33 | 29 | 32 | 44 |
| 10 | L1-L2 | 33 | 30 | 32 | 15 |
| Hybrid group | |||||
| 1 | D10-D11 | 20 | 14 | 19 | 48 |
| 2 | D10-D11 | 38 | 28 | 30 | 40 |
| 3 | D10 | 8 | 7 | 8 | 26 |
| 4 | D10-D11 | 16 | 14 | 14 | 22 |
| 5 | D11-D12 | 50 | 44 | 45 | 20 |
| 6 | D9-D10 | 18 | 10 | 13 | 18 |
| 7 | L2-L3 | 25 | 12 | 15 | 18 |
| 8 | D10-D11 | 15 | 12 | 13 | 18 |
| 9 | D9-D10 | 20 | 13 | 15 | 17 |
| 10 | D8-D10 | 5 | 10 | 11 | 17 |
| 11 | D7-D8 | 13 | 10 | 11 | 16 |
| 12 | D11-D12 | 15 | 10 | 11 | 15 |
Complications were observed in two of 22 patients (9%). One patient had a dural tear requiring patch repair and cerebrospinal fluid diversion (hybrid group, Patient 11). Of a total of 76 screws placed percutaneously, one screw breached the medial cortex causing significant radicular pain without deficit (right L5; posterior-only group, Patient 9). This screw had to be repositioned during another surgical procedure; this has been the only reoperation in this series to date. There were no significant approach-related complications in either group. The elderly patients tolerated these procedures well, with no significant pulmonary complications. No evidence of implant fatigue, pullout, or breakage was noticed in any patient at latest followup. Though removal of screws was planned after healing in the posterior-only group, none of these patients have clinical or radiographic evidence of implant failure. The screws have therefore not been removed.
Discussion
Tuberculosis of the spine is a common site of extraskeletal tuberculosis. Depending on the level and extent of the disease, it can cause back pain, deformity of the spine, and neurologic deficits. As the disease primarily involves vertebral bodies, conventional exposures to the ventral aspect of the thoracic and lumbar spine are extensive, with their attendant morbidities, and may be poorly tolerated by these patients [15]. MIS techniques are being applied to treat such spinal disorders as trauma, tumors, and deformity [2, 3, 16, 21, 28] with good results. We therefore evaluated (1) neurologic results, (2) radiographic results, and (3) complications associated with either a posterior-only MIS approach or a hybrid MIS approach in patients with tuberculosis of the thoracic and lumbar spine.
For those requiring ventral column reconstruction, VATS has been used as a MIS technique for biopsy and decompression alone [10, 12, 15] or in combination with percutaneous posterior transpedicular fixation [11, 24]. However, there are some limitations to this technique [11, 15]. It requires additional equipment, dedicated long instruments, a steep learning curve, a two-dimensional view, hemodynamic alterations due to collapse of the lung, pleural adhesions limiting this approach, and significant pulmonary complications [15]. Mini-thoracotomy (mini-open or using tubular retractor ports) on the other hand overcomes all these limitation in addition to being familiar to spine surgeons because the anatomic landmarks are similar to those used in conventional approaches. Combined with percutaneous transpedicular fixation, this hybrid MIS method helps to achieve all the aims of conventional approaches using MIS principles.
In a subgroup of patients with focal peridiscal involvement and relatively preserved vertebral body height but with significant neurologic deficits due to epidural compression, conventional approaches may be considered an overtreatment. Keeping these patients on ATT medications without offering surgery may result in poor neurologic outcomes [7], delayed or incomplete neurologic recovery, and progression of deformity, causing significant pain and delayed neurologic deterioration [8, 29, 31]. Other MIS techniques such as nucleotome-based percutaneous aspiration [5, 32], endoscopic suction and drainage [6], and open transpedicular decompression [1] have also been described for infective spondylodiscitis. These have been used predominantly for evacuation of the pus and their utility is limited in those with thicker pus or granulation tissue. Lee et al. [13] described conventional posterior transpedicular curettage and decompression augmented with transpedicular screws in adjacent healthy vertebrae for similar patients with preserved vertebral body heights. Kandwal et al. [11] used transforaminal access using MIS methods in lumbar vertebral bodies. The posterior-only MIS approach as described in this series is feasible in patients with both thoracic and lumbar tuberculosis, with good neurologic recovery and nonprogression of deformity.
Good neurologic recovery is achieved using either conventional methods irrespective of ventral or dorsal approaches for decompression [22, 23] or using MIS methods such as VATS [10, 11, 15]. Similar results were observed in our study. Decompression using MIS methods is sufficient to achieve good neurologic outcomes.
Spinal deformity is a dreaded complication occurring during the active healing phase and after healing in those with severe deformity [9]. In our posterior-only group, there was only 2° loss of correction over 30 months of followup. This was much less than an increase of 6° to 15° in kyphosis seen in those treated nonoperatively [1, 17]. The deformity progression in the first two patients of posterior-only group were comparable to those with surgical decompression only [7, 9, 17, 20]. Due to the relatively preserved vertebral body height, spontaneous fusion is expected to occur in these patients after healing and internal fixation helped to avoid deformity progression in this subgroup of patients till healing occurred. Ringel et al. [24] treated 15 patients with infective spondylodiscitis of the thoracic and lumbar spine with transpedicular fixation and antibiotics only. These patients were not fit to undergo a second-stage ventral procedure. Lee et al. [13] performed drainage and curettage and fixation without interbody fusion in 10 patients with infective lumbar spondylodiscitis. All patients achieved fusion. Posterior transpedicular fixation is better in correcting and maintaining deformity [14, 22]. Percutaneous methods achieve this aim with minimal dissection and tissue disruption. In two studies by Lee et al. [14] and Pu et al. [22] using conventional methods of posterior transpedicular decompression and fixation, mean preoperative kyphosis ranged from 18° to 24°. This was corrected by 4° to 8°, with a 6° loss of correction over a mean followup of 22.5 months [22]. These modest corrections were similar to our posterior-only group. Patients with deformities between 16° and 40° treated by ventral débridement and posterior instrumentation using either conventional or VATS methods achieved 8° to 21° correction in deformity, with losses of 4° to 5° [9, 11, 22]. In our hybrid group, the mean preoperative kyphosis was 19.8°, with a 4° correction achieved postoperatively and a 1.6° loss of correction at last followup. Though there has been some loss of correction in both of our groups, this was mild and may not affect fusion. Upadhaya et al. [31] and Jain et al. [9] showed that the deformity does not progress once healing of tuberculosis occurs. With minimum followup of 15 months and healed tuberculosis in all, further progression of deformity is not expected in our group of patients.
Complications were observed in only 9% of patients. This is significantly less than the 24% to 31% rates reported using conventional approaches [15, 22, 25]. Even MIS methods such as VATS have significant risk of pulmonary complications [10, 11, 15].
Implant removal remains controversial in those who undergo fixation without fusion, as it requires second surgery and general anesthesia. Studies done by De Iure [3] and Court and Vincent [2] on patients with spinal fractures treated by percutaneous transpedicular fixation methods show that the need for implant removal may be low, as few patients have clinical or radiographic evidence of failure. Infective pathologies may behave differently, as fusions between involved bodies occur in them. Lee et al. [13] did not remove screws in any of their 10 patients with infective spondylodiscitis at a mean followup of 29 months (maximum, 61 months), although they noticed screw loosening in three patients. None of our patients with a mean followup of 30 months (maximum, 59 months) had any evidence of implant fatigue. Screws have not been removed in any of our patients.
This study had a number of limitations. First, there is no comparison group between the conventional and MIS procedures. Thus, it is difficult to quantify the difference in efficacy between MIS procedures and conventional ones. Second is the short-term followup with no formal assessment of fusion. We used no standardized outcome scores for pain or function so we cannot be sure whether our MIS approaches are any easier on patients than traditional approaches. Our deformity corrections were modest, with no intraobserver reliability tests performed. Finally, the MIS approaches cannot be applied universally and selection criteria should be laid down depending on the level of the disease and the available expertise and instrumentation.
In conclusion, this study demonstrated that MIS techniques can be used appropriately to treat patients with tuberculosis of the thoracic and lumbar spine, with good neurologic recovery, avoidance of deformity progression, and few complications. Properly designed studies are needed to compare the MIS approaches with open approaches with respect to different end points.
Footnotes
Each author certifies that he or she, or a member of his or her immediate family, has no funding or commercial associations (eg, consultancies, stock ownership, equity interest, patent/licensing arrangements, etc) that might pose a conflict of interest in connection with the submitted article.
All ICMJE Conflict of Interest Forms for authors and Clinical Orthopaedics and Related Research editors and board members are on file with the publication and can be viewed on request.
Clinical Orthopaedics and Related Research neither advocates nor endorses the use of any treatment, drug, or device. Readers are encouraged to always seek additional information, including FDA approval status, of any drug or device before clinical use.
Each author certifies that his or her institution approved or waived approval for the human protocol for this investigation and that all investigations were conducted in conformity with ethical principles of research.
References
- 1.Chacko AG, Moorthy RK, Chandy MJ. The trans-pedicular approach in the management of thoracic spine tuberculosis: a short term follow up study. Spine (Phila Pa 1976). 2004;29:E363–E567. [DOI] [PubMed]
- 2.Court C, Vincent C. Percutaneous fixation of thoracolumbar fractures: current concepts. Orthop Traumatol Surg Res. 2012;98:900–909. doi: 10.1016/j.otsr.2012.09.014. [DOI] [PubMed] [Google Scholar]
- 3.De Iure F, Cappuccio M, Paderni S, Bosco G, Amendola L. Minimally invasive percutaneous fixation of thoracic and lumbar spine fractures. Min Invasive Surg. 2012;2012:141032. doi: 10.1155/2012/141032. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.El Masry WS, Tsubo M, Katoh S, El Miliqui YH, Khan A. Validation of the American Spinal Injury Association (ASIA) motor score and the National Acute Spinal Cord Injury Study (NASCIS) motor score. Spine (Phila Pa 1976). 1996;21:614–619. [DOI] [PubMed]
- 5.Hanaoka N, Kawasaki Y, Sakai T, Nakamura T, Nanamori K, Nakamura E, Uchida K, Yamada H. Percutaneous drainage and continuous irrigation in patients with severe pyogenic spondylitis, abscess formation and marked bone destruction. J Neurosurg Spine. 2006;4:374–379. doi: 10.3171/spi.2006.4.5.374. [DOI] [PubMed] [Google Scholar]
- 6.Ito M, Sudo H, Abumi K, Kotani Y, Takahata M, Fujita M, Minami A. Minimally invasive surgical treatment for tuberculous spondylodiscitis. Minim Invasive Neurosurg. 2009;52:250–253. doi: 10.1055/s-0029-1220685. [DOI] [PubMed] [Google Scholar]
- 7.Jain AK. Treatment of tuberculosis of the spine with neurological complications. Clin Orthop Relat Res. 2002;398:75–84. doi: 10.1097/00003086-200205000-00011. [DOI] [PubMed] [Google Scholar]
- 8.Jain AK. Tuberculosis of the spine: a fresh look at an old disease. J Bone Joint Surg Br. 2010;92:905–913. doi: 10.1302/0301-620X.92B7.24668. [DOI] [PubMed] [Google Scholar]
- 9.Jain AK, Dhammi IK, Jain S, Mishra P. Kyphosis in spinal tuberculosis-prevention and correction. Indian J Orthop. 2010;44:127–136. doi: 10.4103/0019-5413.61893. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Jayaswal A, Upendra B, Ahmed A, Chowdhury B, Kumar A. Video assisted thoracoscopic anterior surgery for tuberculous spondylitis. Clin Orthop Relat Res. 2007;460:100–107. doi: 10.1097/BLO.0b013e318065b6e4. [DOI] [PubMed] [Google Scholar]
- 11.Kandwal P, Garg B, Upendra BN, Chowdhury B, Jayaswal A. Outcome of minimally invasive surgery in the management of tuberculous spondylitis. Indian J Orthop. 2012;46:159–164. doi: 10.4103/0019-5413.93680. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Kapoor SK, Kapoor S, Agarwal M, Aggarwal P, Jain BK., Jr Thoracoscopic decompression in Pott’s spine and its long term follow up. Int Orthop. 2012;36:331–337. doi: 10.1007/s00264-011-1453-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Lee BH, Lee HM, Kim TH, Kim HS, Moon ES, Park JO, Chong HS, Moon SH. Transpedicular curettage and drainage of infective lumbar spondylodiscitis: technique and clinical results. Clin Orthop Surg. 2012;4:200–208. doi: 10.4055/cios.2012.4.3.200. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Lee SH, Sung JK, Park YM. Single-stage transpedicular decompression and posterior instrumentation in treatment of thoracic and thoracolumbar spinal tuberculosis: a retrospective case series. J Spinal Disord Tech. 2006;19:595–602. doi: 10.1097/01.bsd.0000211241.06588.7b. [DOI] [PubMed] [Google Scholar]
- 15.Lu G, Wang B, Li J, Liu W, Cheng I. Anterior debridement and reconstruction via thoracoscopy assisted mini-open approach for the treatment of thoracic spinal tuberculosis: minimum 5 year follow up. Eur Spine J. 2012;21:463–469. doi: 10.1007/s00586-011-2038-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.McLain RF. Spinal cord decompression: an endoscopically assisted approach for metastatic tumors. Spinal Cord. 2001;39:482–487. doi: 10.1038/sj.sc.3101194. [DOI] [PubMed] [Google Scholar]
- 17.Medical Research Council A 15 year assessment of controlled trials of the management of tuberculosis of the spine in Korea and Hong Kong. J Bone Joint Surg Br. 1998;80:456–462. doi: 10.1302/0301-620X.80B3.8544. [DOI] [PubMed] [Google Scholar]
- 18.Moon MS. Tuberculosis of the spine: controversies and a new challenge. Spine (Phila Pa 1976). 1997;22:1791–1797. [DOI] [PubMed]
- 19.Nurick S. The pathogenesis of the spinal cord disorder associated with cervical spondylosis. Brain. 1972;95:87–100. doi: 10.1093/brain/95.1.87. [DOI] [PubMed] [Google Scholar]
- 20.Okada Y, Miyamoto H, Uno K, Sumi M. Clinical and radiological outcome of surgery for pyogenic and tuberculous spondylitis: comparisons of surgical technique and disease type. J Neurosurg Spine. 2009;11:620–627. doi: 10.3171/2009.5.SPINE08331. [DOI] [PubMed] [Google Scholar]
- 21.Palmisani M, Gasbarrini A, Brodano GB, De Iure F, Cappuccio M, Boriani L, Amendola L, Boriani S. Minimally invasive percutaneous fixation in the treatment of thoracic and lumbar spine fractures. Eur Spine J. 2009;18(suppl 1):S71–S74. doi: 10.1007/s00586-009-0989-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Pu X, Zhou Q, He Q, Dai F, Xu J, Zhang Z, Branko K. A posterior versus anterior surgical approach in combination with debridement, interbody autografting and instrumentation for thoracic and lumbar tuberculosis. Int Orthop. 2012;36:307–313. doi: 10.1007/s00264-011-1329-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Rasouli MR, Mirkoohi M, Vaccaro AR, Yarandi KK, Movaghar VR. Spinal tuberculosis: diagnosis and management. Asian Spine J. 2012;6:294–308. doi: 10.4184/asj.2012.6.4.294. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Ringel F, Stoffel M, Stuer C, Meyer B. Minimally invasive transmuscular pedicle screw fixation of the thoracic and lumbar spine. Neurosurgery. 2006;59(4 suppl):361–367. doi: 10.1227/01.NEU.0000223505.07815.74. [DOI] [PubMed] [Google Scholar]
- 25.Sharkawi MM, Said GK. Instrumented circumferential fusion for tuberculosis of the dorso-lumbar spine: a single or double stage procedure? Int Orthop. 2012;36:315–324. doi: 10.1007/s00264-011-1401-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Smith JS, Ogden AT, Fessler RG. Minimally invasive posterior thoracic fusion. Neurosurg Focus. 2008;25:E1–E9. doi: 10.3171/FOC/2008/25/8/E9. [DOI] [PubMed] [Google Scholar]
- 27.Soundararaj GD. Simultaneous anterior decompression and posterior instrumentation of the tuberculous spine using anterolateral extrapleural approach. J Bone Joint Surg Br. 2009;91:702–703. doi: 10.1302/0301-620X.91B5.22532. [DOI] [PubMed] [Google Scholar]
- 28.St Clair SF, McLain RF. Posterolateral spinal cord decompression in patients with metastasis: an endoscope assisted approach. Surg Technol Int. 2006;15:257–263. [PubMed] [Google Scholar]
- 29.Tuli SM. Severe kyphotic deformity in tuberculosis of the spine. Int Orthop. 1995;19:327–331. doi: 10.1007/BF00181121. [DOI] [PubMed] [Google Scholar]
- 30.Tuli SM. Tuberculosis of the spine: a historical review. Clin Orthop Relat Res. 2007;460:29–38. doi: 10.1097/BLO.0b013e318065b75e. [DOI] [PubMed] [Google Scholar]
- 31.Upadhyay SS, Saji MJ, Sell P, Sell B, Yau AC. Longitudinal changes in spinal deformity after anterior spinal surgery for tuberculosis of the spine in adults; a comparative analysis between radical and debridement surgery. Spine (Phila Pa 1976). 1994;19:542–549. [DOI] [PubMed]
- 32.Yu MY, Siu C, Wing PC, Schweigel IF, Jetha N. Percutaneous suction aspiration for osteomyelitis: report of two cases. Spine (Phila Pa 1976). 1991;16:198–202. [PubMed]