Determinants of surgical outcome in spinal tuberculosis: A retrospective analysis of 84 patients

Abstract

Background:

Spinal tuberculosis (STB) is a major cause of spinal deformity, instability, and neurological deficits. While antitubercular therapy (ATT) is the primary treatment, surgery is indicated in advanced disease or complications. This study evaluates surgical management and outcomes in cervical, thoracic, and thoracolumbar STB at a single Indian center.

Methodology:

Retrospective review of 84 patients (cervical 17, thoracic 37, thoracolumbar 14, and lumbar 16) who underwent surgery from 2015 to 2023. Data included demographics, clinical presentation, neurological status (American Spinal Injury Association [ASIA] scale), pain (Visual Analogue Scale [VAS]), surgical approach, fusion technique, complications, and postoperative outcomes. Surgical strategy was individualized based on lesion location, vertebral involvement, and deformity.

Results:

Mean age ranged from 36.5 (cervical) to 49.3 (lumbar) years; males predominated. Pain (90%), fever (80%), and fatigue (66%) were common; neurological deficits occurred in 51%. Anterior approaches were used in cervical (76%) and thoracic (59%) cases, posterior in thoracolumbar (86%) and lumbar (100%), with combined anterior–posterior in 4%. Operative time and blood loss were lowest for cervical (96 ± 15 min; 70 ± 23 mL) and highest for thoracolumbar cases (156 ± 40 min; 264 ± 81 mL). Postoperatively, most patients improved neurologically (ASIA D/E), with significant pain reduction (VAS 7.1–7.6 → 2.2–2.9) and functional recovery (ODI 49–63 → 19–27). Complications were infrequent and manageable.

Conclusion:

Tailored surgical intervention for STB, guided by anatomical site and disease severity, achieves excellent neurological, functional, and radiological outcomes. Early, appropriately planned surgery remains essential for patients with progressive deficits, instability, or inadequate response to ATT.

INTRODUCTION

Tuberculosis (TB), caused by Mycobacterium tuberculosis (MTB), continues to pose a major global health challenge despite advancements in medical care. In 2023 alone, approximately 8.2 million new cases and 1.2 million deaths were reported worldwide, with the burden falling disproportionately on developing nations.[1] Among extrapulmonary manifestations, spinal TB (STB) represents the most common form of osteoarticular TB, accounting for nearly 50%–60% of cases.[2]

STB most frequently involves the thoracic and lumbar regions, though cervical segments are also affected.[3] The disease is characterized by destruction of intervertebral discs and adjacent vertebral bodies, leading to spinal instability, angular deformities such as gibbus formation, and neurological deficits ranging from radiculopathy to complete paraplegia, quadriplegia.[4,5] Although antitubercular chemotherapy remains the cornerstone of treatment, medical therapy alone is often insufficient in patients with advanced disease or severe complications. Surgical intervention is required in cases of progressive deformity, paravertebral abscess formation, persistent pain, instability, or neurological deterioration.[6]

Multiple surgical strategies have been described, including anterior-only, posterior-only, and combined anterior–posterior approaches.[7] The choice of surgical technique depends on disease location, extent of vertebral involvement, and presence of deformity or neurological compression. However, consensus on the optimal surgical approach, particularly for multi-segmental or cervical lesions, remains limited.[8] Existing evidence highlights favorable outcomes with timely surgery, including deformity correction, neurological recovery, and improved quality of life.[9,10]

In this context, we analyzed the surgical treatment and clinical outcomes of patients with cervical, thoracic, and thoracolumbar STB in a single Indian institution, aiming to contribute further evidence to guide management in diverse clinical settings.

METHODOLOGY

This retrospective study was conducted in the Department of Neurosurgery, TNMC and BYL Nair Hospital, Mumbai, and included 84 patients treated surgically for STB between January 2015 and December 2023. The primary objective was to evaluate the surgical management strategies and clinical outcomes of patients diagnosed with cervical, thoracic, or thoracolumbar STB in our institution.

Patient selection

Patient records were reviewed from the institutional medical records.

Inclusion criteria were:

1. Confirmed diagnosis of STB based on microbiological, histopathological, or radiological evidence

2. Patients who underwent surgical intervention with or without preceding anti-tubercular therapy (ATT)

3. Availability of complete clinical and radiological follow-up data.

Exclusion criteria were:

• Patients with isolated extra STB without vertebral involvement

• Incomplete records or inadequate follow-up data

• Patients were treated conservatively without surgical intervention.

Diagnostic criteria

The diagnosis of STB was established through a combination of:

• Microbiological confirmation: Detection of MTB or acid-fast bacilli from intraoperative or computed tomography (CT)-guided biopsy samples

• Radiological findings: CT and/or magnetic resonance imaging (MRI) demonstrating features typical of spinal TB (vertebral body destruction, paravertebral abscess, disc involvement)

• Clinical correlation: Patients presenting with back pain, neurological deficits, or systemic features of TB

• Therapeutic response: Patients with compatible clinical and radiological findings showing improvement with ATT.

Data collection

Patient data were collected from hospital records and included demographic characteristics such as age, sex, and comorbidities. Clinical information was documented with emphasis on presenting symptoms, neurological deficits, and the duration of symptoms prior to surgery. Radiological evaluation detailed the affected spinal levels and the extent of disease. Neurological status was graded according to the American Spinal Injury Association (ASIA) impairment scale, while pain was assessed using the Visual Analog Scale (VAS). Surgical details, including the operative approach, type of decompression, instrumentation, and reconstruction methods, were recorded. Postoperative outcomes were analyzed in terms of neurological recovery, improvement in pain, complications, and recurrence.

Grouping

Patients were categorized into four groups according to the anatomical site of involvement: Cervical, thoracic, thoracolumbar, and lumbar spine.

Preoperative preparation

All patients were prescribed strict bed rest during the preoperative period and initiated on ATT as per national guidelines.

Drug-susceptible spinal tuberculosis

• Intensive Phase (2 months): isoniazid (5 mg/kg/day, max 300 mg), rifampicin (10 mg/kg/day, max 600 mg), pyrazinamide (25 mg/kg/day, max 2000 mg), and ethambutol (15 mg/kg/day, max 1600 mg)

• Continuation Phase (10–16 months): isoniazid, rifampicin, and ethambutol were continued, with the total duration of ATT extended to 18–24 months depending on clinical and radiological response

• Preoperative ATT was continued for at least 2–3 weeks, or until systemic symptoms such as fever, malaise, or raised inflammatory markers improved.

Multidrug-resistant spinal tuberculosis

• Patients with confirmed multidrug-resistant disease were treated using all-oral, longer regimens under programmatic management of drug-resistant TB.

• The regimen typically included:

  • A later-generation fluoroquinolone (e.g. levofloxacin or moxifloxacin)

  • Linezolid

  • Clofazimine

  • Cycloserine or terizidone

  • Ethambutol and pyrazinamide, if susceptible

  • Bedaquiline for the initial 6 months.

• The treatment duration was generally 18–24 months, adjusted according to drug susceptibility testing and patient response.

Supportive optimization

Supportive optimization was undertaken in all patients prior to surgery. Hematological correction focused on maintaining hemoglobin levels above 10 g/dL. Nutritional status was improved with a high-protein diet and measures to address hypoproteinaemia. Biochemical monitoring of liver and renal function was carried out to ensure adequate tolerance of ATT. In addition, comorbidities such as diabetes, hypertension, or chronic infections were appropriately managed, and any potential surgical contraindications were corrected before intervention.

Surgical management

Surgical intervention was undertaken in patients presenting with progressive neurological deficits due to mechanical compression, spinal deformity or instability, intractable pain, extensive necrotic bone, or lack of adequate response to ATT. The choice of surgical approach was individualized according to the anatomical location of the disease and the extent of vertebral involvement.

Cervical spine

For lesions confined to the anterior column, an anterior approach was preferred, allowing decompression, debridement of infected tissue, and interbody fusion with or without instrumentation. In cases with significant instability or kyphotic deformity, anterior procedures were supplemented with posterior fixation to achieve circumferential stability.

Thoracic and thoracolumbar spine

When posterior elements were involved or deformity correction was required, a posterior (dorsal) approach with laminectomy, decompression, and instrumentation was performed. In cases with extensive vertebral body destruction, a combined dorsoventral approach was employed, incorporating anterior debridement, vertebral body replacement using autograft or cage, and posterior stabilization. Pure anterior debridement and fusion were reserved for patients with localized anterior disease and preserved posterior stability.

Overall, the surgical strategy aimed at thorough debridement of infected tissue, neural decompression, correction or prevention of deformity, and restoration of spinal stability while minimizing morbidity.

RESULTS

A retrospective analysis was conducted on 84 patients who underwent surgical intervention at the Department of Neurosurgery, B.Y.L. Nair Charitable Hospital, Mumbai, over 8 years from January 2015 to December 2023.

A total of 84 patients were included in the study, comprising 17 cervical, 37 thoracic, 14 thoracolumbar, and 16 lumbar cases. The mean age ranged from 36.5 years in cervical to 49.3 years in lumbar disease, with a slight male predominance across all groups. The most common presenting symptoms were back or neck pain (90%), followed by fever (80%) and fatigue (66%), whereas systemic features like weight loss (28%) and night sweats (14%) were less frequent. Neurological involvement was evident in nearly half of the patients, with motor deficits in 43 (51%), sensory deficits in 34 (40%), and bladder or bowel dysfunction in 4 (5%). The mean duration of symptoms varied between 3.4 and 5.1 months. Diagnosis was confirmed most frequently through intraoperative tissue sampling (94%), while CT-guided biopsy (61%) and hematological evaluation (83%) also supported diagnosis in a significant proportion. Radiologically, kyphotic deformity was most frequent in thoracic cases (62%), and paravertebral abscesses were observed in 63% overall. Cord compression was seen in 60% of thoracic and 65% of cervical cases, whereas vertebral collapse was more common in thoracic involvement (76%). Multiple vertebral involvement predominated across all regions, especially in thoracolumbar and lumbar disease [Table 1].

Table 1.

Demographic characteristics, clinical presentation, diagnostic modalities, and radiological profile of patients with spinal tuberculosis stratified by spinal region

Parameter	Cervical	Thoracic	Thoracolumbar	Lumbar
Demography
Number of patients	17	37	14	16
Mean age (year)	36.5	43.7	46.5	49.3
Gender
Male	10	22	8	9
Female	7	15	6	7
Clinical presentation
Fever	15	29	10	13
Fatigue	12	24	9	11
Night sweat	2	5	3	2
Weight loss	5	9	2	2
Back/neck pain	15	34	12	14
Sensory deficit	7	16	5	6
Motor deficit	9	20	7	7
Bladder/bowel dysfunction	1	2	0	1
Duration of symptoms (months)
Mean±SD	3.4±2.1	4.2±3.1	4.8±3.3	5.1±3.6
Range	1–9	2–13	2–15	2–18
Diagnosis
CT-guided biopsy	12	24	9	7
Intraoperative tissue	16	34	14	15
Hematological	15	30	12	13
Image findings
Spinal kyphotic deformity	3	23	7	3
Presence of paravertebral abscess	8	26	9	10
Cord compression	11	24	8	7
Collapse vertebra	9	28	5	8
Vertebral involvement
Single	3	6	0	2
Multiple	14	31	14	14

SD - Standard deviation; CI - Computed tomography

Anterior approach was most commonly used for cervical (76%) and thoracic (59%) lesions, while the posterior approach predominated in thoracolumbar (86%) and lumbar (100%) cases. A combined anterior and posterior approach was required in only three patients (4%), mainly for extensive thoracic or thoracolumbar disease. For fusion, autologous bone grafts were used in 37% of patients, while cages filled with bone graft were utilized in 63%, with higher cage use in all spinal level cases. The mean intraoperative blood loss was lowest for cervical cases (70 ± 23 ml) and highest in the thoracolumbar group (264 ± 81 ml). The mean operative time ranged from 96 ± 15 min in cervical cases to 156 ± 40 min in thoracolumbar cases, with thoracic and lumbar groups showing intermediate values [Table 2].

Table 2.

Surgical approaches, fusion techniques, intraoperative parameters, and operative outcomes in cervical, thoracic, thoracolumbar, and lumbar cohorts

	Cervical	Thoracic	Thoracolumbar	Lumbar
Surgical approach
Anterior approach	13	22	1	0
Posterior approach	4	13	12	16
Combined approach	0	2	1	0
Fusion technique
Autologous bone graft	5	15	5	6
Cage (PEEK/titanium mesh cage filled with bone graft)	12	22	9	10
Intraoperative blood loss (mean±SD) (mL)	70±23	225±72	264±81	246±76
Operative time (mean±SD) (min)	96±15	130±32	156±40	122±26

SD - Standard deviation; PEEK - Polyetheretherketone

The most common complications were dural tears and surgical site infections, followed by neurological deterioration. Less frequent complications included wound dehiscence, hardware-related issues, respiratory problems, disease recurrence, nerve root injury, spinal cord injury, and thromboembolic events. Most were managed successfully without long-term sequelae [Table 3].

Table 3.

Spectrum of intraoperative and postoperative complications associated with surgical management of spinal tuberculosis across spinal levels

Complications	Cervical	Thoracic	Thoracolumbar	Lumbar
Intra-operative complications
Dural tear	1	2	1	1
Nerve root injury	0	1	0	1
Spinal cord injury	0	0	1	0
Postoperative complications
Surgical site infection	0	3	1	1
Wound dehiscence	0	1	1	0
Hardware-related (loosening/breakage)	0	1	1	0
Neurological deterioration	0	2	1	1
Respiratory complications	1	1	0	0
Thrombo-embolic events (DVT/PE)	0	0	1	0
Recurrence of disease	0	2	1	1

PE - Pulmonary embolism; DVT - Deep vein thrombosis

Postoperative outcome

Most patients showed neurological improvement, with the majority reaching ASIA D or E postoperatively. Functional outcomes improved significantly, with mean ODI decreasing from 49 to 63 preoperatively to 19–27 postoperatively and VAS from 7.1–7.6 to 2.2–2.9 across all regions. Radiological fusion was achieved in most patients, with grade 1 fusion predominating, and grade 4 (including hardware loosening) was rare. Mean intensive care unit (ICU) stay was 1.4–2.4 days, hospital stay 4.9–7.4 days, and follow-up ranged from 16.3 to 18.2 months, demonstrating sustained recovery [Table 4].

Table 4.

Neurological recovery (American Spinal Injury Association grades), functional outcomes (Oswestry disability index, Visual Analog Scale), radiological fusion status, and postoperative course including intensive care unit stay, hospital stay, and follow-up duration

Neurological outcome
ASIA grade	Cervical (number of patients)		Thoracic (number of patients)		Thoracolumbar (number of patients)		Lumbar (number of patients)
	Preoperative	Postoperative	Preoperative	Postoperative	Preoperative	Postoperative	Preoperative	Postoperative
A	1	0	2	1	1	0	0	0
B	2	1	4	1	1	1	2	0
C	3	2	6	3	2	1	1	2
D	3	4	8	11	3	4	3	2
E	8	10	17	21	7	8	10	12
Functional outcome (mean±SD)
	Cervical		Thoracic		Thoracolumbar		Lumbar
	Preoperative	Postoperative	Preoperative	Postoperative	Preoperative	Postoperative	Preoperative	Postoperative
ODI	57.1±7.2	23.2±5.1	62.5±7.8	27.1±5.9	54.6±6.1	21.4±3.5	49.4±4.6	19.6±3.7
VAS	7.1±1.1	2.6±0.6	7.6±0.9	2.9±0.7	7.3±1.1	2.5±0.8	7.1±0.9	2.2±0.7
Radiological outcome
	Cervical			Thoracic	Thoracolumbar		Lumbar
Bridwell’s fusion grades (number of patients)
Grade 1	14		30		12		13
Grade 2	3		5		1		2
Grade 3	0		1		0		1
Grade 4	0		1		1		0
Postoperative stay (days), mean±SD
ICU stay	1.4±0.7		2.1±1.2		2.4±1.3		1.7±1.0
Hospital stay	4.9±1.5		7.1±1.8		7.4±1.9		6.3±1.6
Mean follow-up duration (months)	17.0±3.5		18.2±3.6		16.6±2.8		16.3±3.0

ASIA - American Spinal Injury Association; ODI - Oswestry disability index; VAS - Visual Analog Scale; SD - Standard deviation; ICU - Intensive Care Unit

Illustrative case

• Case 1: An 18-year-old female patient came with complaints of neck pain, fever, and cachexia for 4 months. MRI revealed destruction of the C4 and C5 vertebrae with spondylolisthesis. Anterior corpectomy with fixation using an expandable cage and screw was done [Figure 1]

• Case 2: A 51-year-old male patient came with complaints of neck pain, fever, and lower limb weakness for 5 months. MRI revealed destruction of the C5 and C6 vertebrae with collapse, spondylolisthesis, and severe kyphotic deformity. Anterior corpectomy with fixation using an expandable cage and screw was done [Figure 2]

• Case 3: A 43-year-old male patient came with complain with upper back pain, fever, night sweats, both lower limb weakness. MRI revealed a paravertebral abscess with involvement of D4–D6 vertebral bodies. Transthoracic D5, D6 corpectomy with fusion and fixation using an expandable cage was performed [Figures 3 and 4].

Figure 1.

(a) Magnetic resonance imaging T2-weighted image showing destruction of C4, C5 vertebrae with spondylolisthesis. (b) showing postoperative x-ray cervical spine lateral view following C4, C5 corpectomy with expandable cage and screw in place with correction of kyphotic deformity

Figure 2.

(a) Magnetic resonance imaging T2-weighted image showing complete destruction of C5, C6 vertebrae with spondylolisthesis and severe kyphotic deformity. (b) showing post-operative x-ray cervical spine lateral view following C5, C6 corpectomy with expandable cage and screw in place with correction of kyphotic deformity

Figure 3.

(a) Showing Left lateral decubitus positioning of the patient for surgery. (b) Showing Incision through the 5^th Intercostal space. (c) Showing an intraoperative picture after the collapse of the lung showing the lesion. (d) Showing intraoperative picture after evacuation of D5, D6 paravertebral abscess along corpectomy with fixation and fusion using an expandable cage

Figure 4.

(a) Magnetic resonance imaging postcontrast T1-weighted image showing Paravertebral abscess with involvement of D4-D6 vertebral bodies. (b) MRI postcontrast T1-weighted image showing fusion and fixation of D5, D6 vertebrae using expandable cage

DISCUSSION

STB, also known as Pott’s disease, is a chronic infectious condition caused by MTB that primarily affects the vertebral column, often leading to deformities, instability, and neurological deficits.[11] Surgical intervention is indicated in cases of progressive neurological deterioration, severe deformity, intractable pain, or failure of conservative management.[12]

In this study of 84 patients, the mean age ranged from 36.5 years in cervical cases to 49.3 years in lumbar disease, with a slight male predominance. Back or neck pain (90%), fever (80%), and fatigue (66%) were the most common presenting symptoms, while systemic features like weight loss (28%) and night sweats (14%) were less frequent. Neurological deficits were present in 51% of patients, with motor deficits being the most common. These findings are consistent with prior reports demonstrating thoracic predominance, similar symptom patterns, and a male preponderance in spinal TB.[13,14]

Diagnosis relied primarily on intraoperative tissue sampling (94%), supplemented by CT-guided biopsy (61%) and hematological investigations (83%). Radiologically, kyphotic deformities were most common in thoracic cases (62%), vertebral collapse in 76% of thoracic cases, paravertebral abscesses in 63%, and cord compression in 60%–65% overall. Multiple vertebral involvement was the predominant pattern. This aligns with studies conducted by Gan et al. and Alam et al., highlighting the importance of combined clinical, microbiological, and radiological assessment for accurate diagnosis.[12,15]

Surgical planning was individualized according to lesion location and extent. Anterior approaches were preferred for cervical13/17 cases (76%) and thoracic 22/37 cases (59%), whereas posterior approaches predominated for thoracolumbar 12/14(86%) and lumbar 16/16(100%) cases. Combined anterior-posterior approaches were required in only 3/84 patients (4%). Autologous bone grafts were used in 26/84 cases (37%), while cages filled with bone graft were used in 58/84 cases (63%). Intraoperative blood loss and operative time were lowest for cervical cases (70 ± 23 mL; 96 ± 15 min) and highest for thoracolumbar cases (264 ± 81 mL; 156 ± 40 min). These findings are consistent with a study conducted by Romaniyanto et al., emphasizing individualized surgical planning based on lesion location and severity.[16]

The most common complications were dural tears and surgical site infections, followed by neurological deterioration. Less frequent complications included wound dehiscence, hardware-related issues, respiratory complications, thromboembolic events, and recurrence of disease. These complication rates are comparable with previously reported series and demonstrate that most adverse events are manageable without long-term sequelae.[17]

Postoperative outcomes were favorable. Most patients in our study showed neurological improvement, with the majority reaching ASIA D or E postoperatively. Functional outcomes improved significantly, with mean ODI decreasing from 49 to 63 preoperatively to 19–27 postoperatively and VAS from 7.1–7.6 to 2.2–2.9 across all regions. Radiological fusion was achieved in most patients, with grade 1 fusion predominating, and grade 4 (including hardware loosening) was rare. Mean ICU stay was 1.4–2.4 days, hospital stay 4.9–7.4 days, and follow-up ranged from 16.3 to 18.2 months, demonstrating sustained recovery. These results are consistent with previous studies conducted by Kilinc et al. and Huang et al., reporting favorable neurological, functional, and radiological outcomes after surgical intervention.[13,18]

Overall, this study reinforces the efficacy and safety of tailored surgical strategies in the management of STB, highlighting that individualized approaches based on anatomical location and disease severity result in excellent neurological recovery, functional improvement, and high fusion rates.

This study has several limitations. Its retrospective design and single-center setting may introduce selection bias and limit the generalizability of the results. In addition, the absence of a conservatively managed control group precludes direct comparison between surgical and non-surgical outcomes. Future prospective multicenter studies with larger cohorts are warranted to validate these findings and to develop standardized surgical guidelines for STB. Further research exploring minimally invasive approaches and advanced biomaterials for spinal fusion may also enhance patient outcomes.

CONCLUSION

Tailored surgical intervention for STB offers reliable and reproducible outcomes in terms of neurological recovery, pain relief, functional improvement, and spinal stability. The selection of anterior, posterior, or combined approaches based on the anatomical location, extent of vertebral involvement, and severity of deformity is critical to optimizing patient outcomes. Early and appropriately planned surgery facilitates effective neural decompression, correction of kyphotic deformity, and high rates of solid spinal fusion, while minimizing perioperative complications. This study reinforces that individualized surgical strategies remain a cornerstone in the management of STB, particularly in patients with progressive neurological deficits, structural instability, or inadequate response to medical therapy, underscoring the importance of a multidisciplinary approach for achieving optimal long-term results.

Conflicts of interest

There are no conflicts of interest.

Funding Statement

Nil.