Clinical and epidemiological analysis of 893 patients with spinal tuberculosis: an 11-Year investigation of a general hospital in East China

Abstract

Background

The burden of spinal tuberculosis (STB) in China remains substantial, with the country ranking third in the number of tuberculosis cases globally in 2022, among the 30 countries with a high tuberculosis burden. In East China, few large-scale studies have been conducted on STB.

Methods

This retrospective study analyzed 893 confirmed STB cases (2010–2020). Demographic, clinical, and diagnostic data were statistically characterized using χ²/t-tests for categorical/continuous variables (significance at P < 0.05).

Results

The annual number of confirmed STB cases of spinal TB showed a sustained upward trend. Among 893 STB patients (male: female = 1.4:1; median age 56 years), rural populations exhibited higher prevalence (P = 0.264 for delayed hospitalization vs. urban). Farmers/laborers predominated (84.7%), with hypertension (32.4%), diabetes (18.5%) and osteoporosis (12.9%) as major comorbidities. Concurrent pulmonary TB occurred in 435 cases (48.7%) and other extrapulmonary TB in 71 (8.0%). Diagnostic evaluations revealed TSPOT (82.7%), histopathology (75.9%) and Xpert-MTB/RIF (71.8%) as most sensitive methods. Compared with histopathological gold standard, Xpert demonstrated 81.5% sensitivity, 58.8% specificity, 87.0% PPV and 48.3% NPV (kappa = 0.374). Combining histopathology with Xpert achieved 86.6% diagnostic accuracy, significantly surpassing individual methods (P < 0.001). Lesion distribution showed lumbar (44.3%) and thoracic (42.3%) predominance, mostly involving ≥ 2 contiguous vertebrae (91.3% continuous vs. 8.7% skip lesions). Chemotherapy remained primary treatment, with 6.5% drug resistance rate showing annual escalation (mainly monoresistance). Surgical intervention achieved favorable outcomes in 648 cases (72.6%).

Conclusion

Over the course of the study period, the overall diagnosis rate of spinal tuberculosis exhibited an upward trend. Despite East China’s relatively advanced socioeconomic and healthcare systems, spinal tuberculosis remains a substantial public health challenge, primarily due to the region’s complex population composition and high population mobility. The prevention and management of spinal tuberculosis continue to present considerable challenges. Early diagnosis, combined with an appropriate treatment course, ensures that both chemotherapy and surgical interventions yield satisfactory outcomes. Increasing the allocation and investment of medical resources for tuberculosis, enhancing health management for migrant populations, and raising public health awareness are essential.

Introduction

Spinal tuberculosis (STB) is a prevalent form of extrapulmonary tuberculosis (EPTB). According to the “Global Tuberculosis Report 2023” [1]released by the World Health Organization (WHO), an estimated 10.6 million tuberculosis cases occurred globally in 2022. In China, 748,000 new tuberculosis cases were reported in 2022, placing the country third among the 30 high-burden nations, accounting for 7.1% of global cases—lower than India (27%) and Indonesia (10%). Notably, the tuberculosis incidence in China decreased by 4.1% compared to 2021. The overall trend of tuberculosis incidence has shown a gradual decline, with an average annual reduction of 3.2% [2]. Although the tuberculosis burden in China has decreased, prevention and treatment efforts have been largely successful. This success is attributed to China’s focus on early diagnosis and prompt treatment of tuberculosis, aiming to detect the disease early and initiate treatment as soon as possible. However, regional epidemiological disparities and uneven healthcare distribution present significant challenges to tuberculosis prevention and control in China. According to the 5th National Tuberculosis Epidemiological Survey [3], tuberculosis of the bones and joints accounts for 1–5% of all tuberculosis cases. Shi Shiyuan et al. [4]reported that spinal tuberculosis comprises 69.11% of bone and joint tuberculosis cases. Spinal tuberculosis typically results from the hematogenous spread of Mtb to the spine, leading to destructive lesions in vertebral bodies or intervertebral spaces. This condition may cause spinal instability, kyphosis, and in severe cases, spinal cord compression, resulting in neurological deficits or paraplegia. In recent years, China has implemented a dual approach to tuberculosis detection, combining passive case finding with active screening in high-risk groups such as close contacts, the elderly, and migrant populations. Although active screening is not yet conducted nationwide, it is being increasingly promoted in targeted regions to improve early detection and reduce the burden of undiagnosed cases [5]. Consequently, spinal tuberculosis presents a considerable clinical challenge in orthopedic practice. Several studies [6, 7] have reported on the incidence characteristics of spinal tuberculosis in Southwest China, but reports on the epidemiological characteristics of spinal tuberculosis in Eastern China remain limited.

Materials and methods

Zhejiang Province is a province located on the eastern coast of China. In 2020, the urban population in Zhejiang Province accounted for 72.17%. The per capita disposable income in Zhejiang Province was 52,397 yuan CNY, and the per capita disposable income in Hangzhou was 61,879 yuan CNY. According to reports [3], the prevalence of active, smear-positive, and bacterium-positive pulmonary tuberculosis in Eastern China are lower than those in Central and Western China but still reflect a significant tuberculosis burden. This study was conducted at a general hospital in Hangzhou, Zhejiang Province, China. The hospital where the study was conducted serves as one of the clinical research bases of the National Center for Infectious Diseases in China, as well as the Zhejiang Provincial Tuberculosis Diagnosis and Treatment Center. This study aims to retrospectively analyze the characteristics of 893 patients treated for STB at the Orthopedics Department of a general hospital in Zhejiang Province, China, from 2010 to 2020.The objective is to summarize and assess the epidemiological characteristics of spinal tuberculosis in this region, with the goal of contributing to the early and accurate diagnosis as well as the effective clinical management of spinal tuberculosis. This study was approved by the Medical Ethics Committee of Hangzhou Red Cross Hospital. We certify that the study was performed in accordance with the 1964 declaration of HELSINKI and later amendments. Written informed consent was obtained from all the participants prior to the enrollment of this study.

Inclusion criteria

Patients diagnosed with spinal tuberculosis based on comprehensive diagnostic criteria. The composite reference standard (CRS) for spinal tuberculosis is based on the guidelines [8] for diagnosing bone and joint tuberculosis as developed by the National Institute for Health and Care Excellence (NICE) in 2011. The diagnostic criteria for spinal tuberculosis in this study include the following: (1) symptoms and signs consistent with the abscesses and sequestra associated with local tuberculosis lesions observed through imaging (including X-ray, CT, and MRI); (2) pathological examination of lesion specimens revealing caseous necrosis and Langerhans granulomas; (3) microbiological examination of lesion specimens (via traditional Roche culture) confirming Mtb infection; (4) regular anti-tuberculosis treatment for more than 6 months, resulting in the resolution of systemic and local symptoms. A diagnosis of spinal tuberculosis was confirmed if any two of the above criteria were met.

Exclusion criteria

(1) Patients with unclear infections or confirmed non-tuberculous mycobacterial infections identified through genetic testing, pathogen culture, histopathology, or other diagnostic methods were excluded. (2) Additionally, patients receiving effective anti-infection treatment but with an unclear clinical diagnosis were also excluded.

Statistical analysis

Statistical analysis was performed using SPSS 25.0 software. Categorical variables were presented as frequency and composition ratio (N,%), while continuous variables were presented as (). For variables not conforming to a normal distribution, values were expressed as the median along with interquartile range (IQR). The chi-square test was employed for categorical variables, while the t-test or nonparametric tests were used for continuous variables. The kappa test was applied to assess consistency. A P-value of < 0.05 was considered statistically significant.

Result

Demographics and epidemiology

From 2010 to 2020, a total of 893 patients were diagnosed with spinal tuberculosis in our hospital. The number of confirmed cases increased annually, from 10 cases in 2010 to 141 cases in 2020. Although the annual growth rate has slowed in recent years, the average annual increase remained approximately 30.29% (Fig. 1). Of these, 521 (58.34%) were male and 372 (41.66%) were female, resulting in a male-to-female ratio of 1.40:1. A total of 511 patients were from rural areas and 382 from urban areas. The majority of patients were from Zhejiang, Anhui, Jiangsu, and Jiangxi provinces (Fig. 2), all located in the East China region. The occupations of the patients were predominantly farmers, workers, retirees, and freelancers, with a smaller proportion of teachers and students. The median time from symptom onset to hospitalization was 4 (2–12) months. The median time from symptom onset to hospitalization for rural patients was 4 (2–12) months, while that for urban patients was 3.5 (2–10) months. There was no significant difference between the two groups (Z=−1.117, P = 0.264). Common comorbidities included hypertension (170, 19.04%), diabetes (75, 8.40%), osteoporosis (64, 7.17%), and hypoproteinemia (35, 3.92%). No AIDS patients were found in the study (Table 1).

Fig. 1.

Composition of rural and urban patients with STB over the years

Fig. 2.

Local distribution of 893 patients with STB. The outer circles represent the proportion of patients from each province. The inner circles represent the geographical relationship of each province to Zhejiang Province

Table 1.

Characteristics of 893 spinal TB patients

Characteristics	Value
Sex, M/F	521/372
Age group, N (%)
2-10y	2(0.22)
11-20y	32(3.58)
21-30y	143(16.01)
31-40y	80(8.96)
41-50y	102(11.42)
51-60y	174(19.48)
61-70y	184(20.60)
71-80y	143(16.01)
81-90y	33(3.70)
Source, Urban/Rural	511/382
Occupation, N (%)
Farmer	505(56.55)
Worker	92(10.30)
Retiree	136(15.23)
Freelancer	105(11.76)
Student	21(2.35)
Other	34(3.81)
Duration of symptoms, M(Q1-Q3)	4（2-12）months
Urban	3.5（2-10）months
Rural	4（2-12）months
STB types, N (%)
Exclusive STB	387(43.34)
STB & PTB	435(48.71)
STB & EPTB (without PTB)	71(7.95)
Comorbiditiy, N (%)
Hypertension	170(19.04)
Diabetes	75(8.40)
Osteoporosis	64(7.17)
Hypoproteinemia	35(3.92)
Anemia	34(3.81)
Pneumonia	30(3,36)
Viral hepatitis	27(3.02)
Chronic kidney disease	18(2.02)
Rheumatic immune disease	15(1.68)
Coronary heart diseas	14(1.57)
COPD	9(1.01)
Pneumosilicosis	8(0.90)
Syphilis	5(0.56)
Other	10(1.23)

The median age of the 893 patients included in this study was 56 (36–68) years, with male patients having a median age of 56 (36–67) years and female patients having a median age of 56 (36-69.75) years. No significant difference in age distribution was found between the two groups (Z=−0.429, P = 0.668). All age groups were affected, with the largest number of patients (184, 20.60%) in the 61–70 age group, and the smallest number (2 cases) in the 1–10 age group (Table 1). From 2010 to 2020, the proportion of elderly patients (aged 60 years and above) gradually increased, while the proportion of younger patients (aged 40 years and below) decreased significantly (Fig. 3). The average length of hospital stay was 41.64 days, with no significant fluctuations in the average length of stay after 2013 (Fig. 4). The average hospitalization cost was 72,003.41 yuan, exhibiting an overall upward trend (Fig. 5).

Fig. 3.

Changes in the proportion of adolescents and elderly patients with STB over the years

Fig. 4.

The average length of hospital stay in patients with STB over the years

Fig. 5.

The change of the average hospitalization cost of patients with STB over the years

Of the 893 patients with spinal tuberculosis, 506 had multifocal tuberculosis, with 435 (48.71%) having pulmonary tuberculosis (PTB), and 71 (7.95%) having other types of extrapulmonary tuberculosis (EPTB) (Table 1). Ten patients had a history of contact with pulmonary tuberculosis patients, though two of them were not diagnosed with the disease. A total of 315 other types of EPTB lesions were identified in this study (some patients had multiple lesions), primarily involving bone and joint tuberculosis (193, 61.27%), lymph node tuberculosis (25, 7.94%), tuberculous pleurisy (24, 7.62%), renal tuberculosis (18, 5.71%), and intestinal tuberculosis (10, 3.17%). Among bone and joint tuberculosis, Mtb primarily affected the sacroiliac joints (60, 20.32%), iliac bones (49, 15.56%), ribs (24, 7.62%), and hip joints (19, 6.03%) (Fig. 6).

Fig. 6.

Distribution of other EPTB lesions in 893 patients with STB

Laboratory findings

The statistical results showed that the TSPOT sensitivity of patients in this study was 82.70%, and the serum tuberculosis antibody (TB-Ab) sensitivity was 71.62%. Further diagnosis was made by histopathological examination (HPE), GeneXpert-MTB/RIF assay (hereinafter referred to as “Xpert”), BACTEC MGIT 960 liquid rapid culture, tuberculosis DNA detection, tuberculosis RNA detection. The sensitivity of the results were 75.93%, 71.84%, 46.48%, 59.15% and 49.51%, respectively (Fig. 7).

Fig. 7.

Yield of different diagnostic procedures

In this study, histopathological examination was performed on 72.56% of the patients. Pathological diagnoses frequently described “granulomatous inflammation,” “caseous necrosis,” and “coagulative necrosis,” along with observations of “epithelioid cells” and “multinucleated giant cells.” Among the 492 patients with a confirmed pathological diagnosis of spinal tuberculosis, 54 tested positive for acid-fast bacilli staining, resulting in a positivity rate of 10.97%. We used the results of pathological examination as “the gold standard for diagnosis”, applying the following diagnostic criteria [9]: Class I, a clear diagnosis (e.g., tuberculosis, lymphoma); Class II, suggestive of a particular diagnosis (e.g., chronic granulomatous inflammation, indicating high suspicion of tuberculosis); Class III, descriptive without bias; and Class IV, non-diagnostic. Class I and II represent pathologically confirmed tuberculosis, while Class III and IV represent pathologically unconfirmed tuberculosis.

We analyzed the results of both pathological examination and Xpert: Among the 426 patients who underwent both tests, 329 cases (77.2%, 95% CI: 0.732 ~ 0.812) were diagnosed with tuberculosis in pathological categories I and II, while 97 cases (22.8%, 95% CI: 0.189 ~ 0.271) were undiagnosed in categories III and IV. Xpert detected tuberculosis in 308 cases (72.3%, 95% CI: 0.678 ~ 0.765) and was negative in 118 cases (27.7%, 95% CI: 0.234 ~ 0.320). Combining pathological examination and Xpert, 369 cases (86.6%, 95% CI: 0.830 ~ 0.897) were diagnosed with tuberculosis, while 57 cases (13.4%, 95% CI: 0.101 ~ 0.166) remained undiagnosed (Table 2).

Table 2.

The results of Xpert and histopathological diagnosis of STB compared to CRS

	Number (%)	95% CI
Xpert
Positive	308(72.3)	0.678 ~ 0.765
Negative	118(27.7)	0.234 ~ 0.320
Histopathology
1 st,2nd diagnosis	329(77.2)	0.732 ~ 0.812
3rd,4th diagnosis	97(22.8)	0.189 ~ 0.271
Xpert + Histopathology
Positive	369(86.6)	0.830 ~ 0.897
Negative	57(13.4)	0.101 ~ 0.166

Abbreviation: CI Confidence interval

Using CRS as the gold standard, a paired chi-square test revealed that the positive rate of pathological diagnosis was higher than that of Xpert detection (X²=60.516, P < 0.001). When comparing the positive rates of Xpert alone with Xpert combined with pathological examination, the difference was significant (X²=171.762, P < 0.001). Similarly, the comparison between pathological examination alone and its combination with Xpert showed significant differences (X²=223.194, P < 0.001) (Table 3). With pathological diagnosis as the gold standard, the sensitivity of Xpert was 81.5% (95% CI: 0.769 ~ 0.853), the specificity was 58.8% (95% CI: 0.488 ~ 0.680), the positive predictive value was 87.0% (95% CI: 0.828 ~ 0.903), and the negative predictive value was 48.3% (95% CI: 0.395 ~ 0.572). The kappa value between pathological examination and Xpert was 0.374 (95% CI: 0.274 ~ 0.473), indicating poor consistency between the two methods (Table 4). The ROC curve for Xpert in diagnosing spinal tuberculosis had an AUC of 0.701 (95% CI: 0.638 ~ 0.764) (Fig. 8).

Table 3.

The results of Xpert and histopathological diagnosis of STB compared to the combined application

	Xpert + Histopathology		Total	X²	P
	Diagnosis of STB	Undiagnosed STB
Xpert
Positive	308	0	308	171.762*	P<0.001
Negative	61	57	118
Total	369	57	426
Histopathology
Positive	329	0	329	223.194#	P<0.001
Negative	40	57	97
Total	369	97	426

*The positive rate of Xpert combined with pathological examination was higher than that of XPERT alone, P=0.000<0.01

#The positive rate of Xpert combined with pathological examination was higher than that of pathological examination alone, P=0.000<0.01

Table 4.

The results of xpert’s diagnosis of STB compared to histopathology

Xpert	Histopathology		Total	X²	P	k value 95% CI	Sensitivity 95% CI	Specificity 95% CI	PPV 95% CI	NPV 95% CI	AUC (95% CI)
	Diagnosis of STB	Undiagnosed STB
Positive	268	40	308	60.516*	P < 0.001	0.374# (0.274 ~ 0.473)	81.5% (0.769 ~ 0.853)	58.8% (0.488 ~ 0.680)	87% (0.828 ~ 0.903)	48.3% (0.395 ~ 0.572)	0.701 (0.638 ~ 0.764)
Negative	61	57	118
Total	329	97	426

Abbreviation: CI Confidence interval, PPV Positive predictive value, NPV Negative predictive value

*The positive rate of pathological examination was higher than that of XPERT, P = 0.000 < 0.01

# k value > 0.75, excellent agreement; 0.75 ≥ k ≥ 0.4, fair to good agreement; k < 0.4, poor agreement

Fig. 8.

ROC curve of Xpert in the diagnosis of STB with histopathology as the gold standard

Imaging findings

Among the 893 patients, a total of 2,508 vertebral lesions were identified, with the lumbar vertebrae (L) most frequently affected (1,112 lesions), particularly at the L4 level (260 lesions). Other affected regions included 1,060 thoracic vertebrae (T), 218 sacral vertebrae (S), 117 cervical vertebrae (C), and 1 coccygeal vertebra (Co) (Fig. 9). Continuous segment involvement was prevalent, with only 78 cases presenting skip lesions. The majority of spinal tuberculosis patients exhibited involvement of two or more vertebrae, while single vertebral involvement was uncommon (Table 5).

Fig. 9.

Number of patients with vertebrae involved at each spinal level

Table 5.

Imaging characteristics of 893 spinal TB patients

Characteristics	Value, N(%)
Location
Cervical	117(4.67)
Thoracic	1060(42.26)
Lumbar	1112(44.34)
Sacral	218(8.69)
Caudal	1(0.04)
Involved vertebra
1	35(3.92)
2	529(59.24)
≥3	329(36.84)
Distribution of affected segments
C	18(2.02)
C-T	10(1.12)
C-L	1(0.11)
T	265(29.68)
T-L	83(9.29)
T-S	2(0.22)
L	328(36.73)
L-S	73(8.17)
S	34(3.81)
Co	1(0.11)
Skipped lesion	78(8.74)
Total	893

Treatment and outcomes

Conservative treatment was selected by 245 patients with spinal tuberculosis, utilizing a standard anti-tuberculosis drug regimen consisting of isoniazid (0.3 g/day), rifampicin (0.45 g/day), ethambutol (0.75 g/day), and pyrazinamide (0.75 g/day). Following 2–4 weeks of intensive anti-tuberculosis therapy, 648 patients proceeded to surgical intervention, primarily involving posterior internal fixation combined with anterior lesion removal and intervertebral fusion. Among the 893 patients, 550 (61.59%) presented with neurologic deficits, defined as ASIA grade A, B, C, or D. A total of 648 patients underwent surgical treatment, of whom 481 (74.23%) had neurologic deficits. The remaining 167 patients without neurologic symptoms underwent surgery due to spinal instability, large abscesses, or failure of conservative treatment. Among the 481 patients with neurologic deficits, most experienced varying degrees of neurological improvement following surgery. After an average follow-up period of 19.4 ± 2.1 months, all patients achieved stable internal fixation without signs of loosening. Notably, 51.11% of these patients reached ASIA grade E, indicating complete neurological recovery.

The overall drug resistance rate to anti-tuberculosis treatment was 6.49%, showing an upward trend in drug-resistant cases annually. Single-drug resistance was the predominant form, followed by multidrug resistance. The primary drugs exhibiting resistance were isoniazid (4.05%), streptomycin (3.24%), and rifampicin (2.57%). No cases of extensively drug-resistant tuberculosis or mortality were observed.

Representative surgical cases

Case 1

A 41-year-old male was admitted with complaints of low back pain accompanied by numbness and weakness in the left lower limb for 1.5 months. Preoperative imaging including X-ray, CT, and MRI (Fig. 10a–d) revealed bony destruction of the L2 and L3 vertebral bodies, surrounding abscess formation, spinal canal stenosis, and compression of the spinal cord and nerve roots. The patient was classified as ASIA grade D before surgery. He underwent posterior internal fixation, thorough debridement, and interbody bone graft fusion. Postoperative imaging (Fig. 10e–i) confirmed satisfactory decompression and stable instrumentation. The numbness in the left lower limb was significantly relieved, muscle strength returned to normal, and his neurological status improved to ASIA grade E.

Fig. 10.

Preoperative and postoperative imaging of case 1. a–d Preoperative lumbar spine imaging, including: (a) anteroposterior X-ray, (b) lateral X-ray, (c) mid-sagittal CT, and (d) mid-sagittal T2-weighted MRI. Images show bony destruction of the L2 and L3 vertebral bodies, surrounding abscess formation, spinal canal stenosis, and spinal cord compression. e Postoperative anteroposterior lumbar X-ray showing stable internal fixation; the resected end of the 12th rib is visible, which was cut during the establishment of the extraperitoneal approach along the anterior margin of the psoas major. f Postoperative lateral X-ray showing a PEEK interbody fusion cage positioned at the L2/3 level (g, h) Postoperative sagittal CT images: (g) showing the in-place PEEK cage, and (h) showing the autologous bone graft supported by the cut segment of the 12th rib. i Postoperative mid-sagittal T2-weighted MRI demonstrating adequate decompression of the spinal canal and complete removal of the abscess

Case 2

A 74-year-old male was admitted with a 2-month history of persistent low back pain and a 1-week history of paraplegia. On neurological examination, the patient exhibited hypoesthesia below the level of the nipples and complete loss of superficial and deep sensation below the inferior margin of the rib cage. Muscle strength in both lower limbs was graded 0. Preoperative imaging, including X-ray, CT, and MRI (Fig. 11a–d), revealed severe bony destruction of the T7 and T8 vertebral bodies, surrounding abscess formation, spinal canal stenosis, and spinal cord degeneration. The patient was classified as ASIA grade (A) He underwent posterior internal fixation, thorough debridement, and interbody bone graft fusion. Postoperatively, the patient reported mild numbness below the costal margin, with partial motor recovery in both lower limbs (muscle strength grade 1). His neurological status improved to ASIA grade (B) Postoperative imaging (Fig. 11e–i) confirmed stable internal fixation and adequate decompression.

Fig. 11.

Preoperative and postoperative imaging of case 2. a–d Preoperative thoracic spine imaging, including: (a) anteroposterior X-ray, (b) lateral X-ray, (c) mid-sagittal CT, and (d) mid-sagittal T2-weighted MRI. Images demonstrate bony destruction of the T7 and T8 vertebral bodies, kyphotic deformity, surrounding abscess formation, spinal canal stenosis, spinal cord compression, and myelopathy. e Postoperative anteroposterior thoracic X-ray showing stable internal fixation and a PEEK interbody fusion cage placed at the T7/8 level. The cut ends of the T7 and T8 rib heads, removed during the transcostotransverse approach, are also visible. f Postoperative lateral X-ray showing correction of kyphotic deformity and the PEEK cage in proper position. g, h Postoperative sagittal CT images: (g) showing the PEEK cage and one of the resected rib heads used as autologous bone graft, and (h) showing the other rib head used for structural support. i Postoperative mid-sagittal T2-weighted MRI showing resolution of the kyphotic deformity and spinal canal stenosis, complete abscess clearance, and disappearance of spinal cord edema

Discussion

Analysis revealed that the annual number of confirmed STB cases of spinal TB increased. The overall trend demonstrated an upward trajectory, with an average annual growth rate of approximately 30.29%. And the annual growth rate has slowed in recent years, aligning with the national trend of decreasing overall tuberculosis incidence [2]. The male-to-female ratio among patients in this study was 1.4:1, consistent with findings from other studies [6, 10]. This disparity may relate to occupational exposure and environmental factors, as male migrant workers may have a higher risk of exposure to Mtb and are more susceptible to disease under conditions of overwork and reduced immunity. Mtb affected all age groups, with ages ranging from 2 to 90 years, and no significant difference was found in age distribution between male and female patients. Consistent with findings from Weng et al.[10], most patients were middle-aged or elderly. Adolescent patients aged 1 to 20 years constituted a relatively small proportion. Research indicates that the protective efficacy of the BCG vaccine lasts for approximately 10 to 15 years [11]. As a result, the immunity established at birth may not confer long-term protection into adulthood [12], and consequently, the proportion of adolescent tuberculosis cases is significantly lower than that in adults. Moreover, the diagnosis of TB in children is inherently difficult due to nonspecific symptoms and limited bacteriological confirmation, which may also contribute to the low proportion observed in this group. With the aging of China’s population [2], the incidence of spinal tuberculosis among elderly patients has increased steadily. During the study period, hospital admissions for spinal tuberculosis in patients aged 60 and above have generally increased, while the proportion of cases among individuals under 40 has gradually declined. This trend may reflect the influence of improved socioeconomic conditions and health policies.

In this study, rural residents represented 57.22% of the patient population, with most being farmers (56.55%) and workers (22.06%). Most patients (77.60%) came from Zhejiang Province. Poor living conditions, limited economic resources, insufficient healthcare access, and low tuberculosis awareness in rural areas lead rural patients to delay seeking treatment until severe pain or neurological symptoms emerge, contributing to delayed diagnosis and treatment. Wang et al. [7] reported an average time from symptom onset to hospital diagnosis of 23.0 months for rural patients, compared to 10.7 months for urban patients. However, in this study, rural patients experienced slightly longer delays than urban patients, and this difference was not statistically significant. The differences observed in comparison to the findings of Wang et al. can largely be attributed to variations in the distribution of medical resources across regions. A well-developed healthcare infrastructure allow rural patients earlier access to hospital treatment. The average hospital stay for spinal tuberculosis patients in this study was 41.64 days, aligning with standard treatment timelines in our facility. The average cost of hospitalization was 72,003.41 yuan CNY, exhibiting an upward trend. The rising costs are associated with advancements in diagnostic technology, increased prices for surgical consumables, and the higher proportion of elderly patients with poor health. However, hospitalization costs have decreased since 2018, possibly due to centralized procurement of medical supplies and the impact of the COVID-19 pandemic.

The most common comorbidities observed in this study included hypertension, diabetes, osteoporosis, and hypoproteinemia, which likely reflect the aging population and their underlying health status. Given the retrospective nature of this study, causality or temporal relationships between these conditions and spinal tuberculosis cannot be established. Mtb can spread within the host through blood or lymphatic pathways [13]. In this study, 48.71% of patients had pulmonary tuberculosis, and 7.95% had extrapulmonary tuberculosis, primarily involving the bones and joints, lymph nodes, and pleura. For bone and joint tuberculosis cases, the sacroiliac joints, ilium, and ribs were most frequently affected. The vertebral body destruction and infiltration of paravertebral abscesses directly or indirectly compromise these structures, leading to bone degradation.

Consistent with findings from other studies [6, 14, 15], the thoracic and lumbar regions were the most frequently affected, accounting for 86.6% of spinal involvement. T8 was the most commonly affected vertebra in thoracic tuberculosis, and L4 was most frequently involved in lumbar tuberculosis, while coccygeal tuberculosis was rare. The sites of spinal tuberculosis are closely associated with spinal anatomical and physiological characteristics, particularly blood supply patterns. Adjacent vertebrae were often involved simultaneously, likely due to the bifurcation of vertebral segmental arteries that supply two neighboring vertebrae [13]. Additionally, tuberculosis bacteria can spread along the anterior and posterior longitudinal ligaments via pus formation, leading to involvement of multiple consecutive or nonadjacent vertebrae [16].

Complete blood count, erythrocyte sedimentation rate (ESR), and C-reactive protein (CRP) can support spinal tuberculosis diagnosis; however, they are non-specific markers and are present in other infectious spinal conditions as well [17]. Numerous studies have analyzed and reported changes in blood count, ESR, and CRP in patients with spinal tuberculosis [13, 18, 19]. Typically, ESR and CRP are markedly elevated, while WBC levels are generally normal, with leukocytosis observed in only 30–50% of spinal tuberculosis cases [16]. Hemoglobin levels typically decrease to varying extents, often resulting in mild anemia. Consequently, this study does not statistically analyze the relationship between these non-specific indicators and spinal tuberculosis. In this study, TSPOT demonstrated the highest sensitivity among laboratory tests (82.7%). However, a positive TSPOT result indicates infection with Mtb but does not confirm active tuberculosis, whereas a negative TSPOT result can often reliably rule out Mtb infection. The sensitivity of serum tuberculosis antibody was 71.62%, differing from other studies: in Yao et al. [6], 15.85% of cases (26 cases) were positive. In Wang et al. [7], TB-Ab positivity was 21.2%. A positive tuberculosis antibody result indicates either current or past Mtb infection and may also be seen post-BCG vaccination. The two studies above were conducted in southwestern China, whereas this study was conducted in eastern China. It has been reported [3] that tuberculosis incidence is lower in eastern than southwestern China. Therefore, the difference between this study and the above two studies may be related to the broader BCG vaccination coverage in the East China.

The “gold standard” for diagnosing spinal tuberculosis involves bacteriological and pathological analyses [20]. However, the BACTEC MGIT 960 rapid liquid culture method presents certain limitations, including extended culture time, high costs, stringent laboratory and personnel requirements, and a low positive rate [21]. Consequently, pathological examination often serves as the primary basis for diagnosing spinal tuberculosis [22]. Following infection of the vertebral body by Mtb, distinctive pathological features emerge, including multinucleated giant cells, Langhans giant cells, caseous necrosis, and granulomas, collectively forming tuberculous nodules, the hallmark histopathological lesion. With disease progression, caseous necrosis advances to form cold abscesses [23]. In 66 cases of spinal tuberculosis reported by Shetty et al. [24], 47 cases (71.21%) exhibited characteristic pathological features of tuberculous granulomas. In this study, 72.56% of patients underwent histopathological examination, with 75.93% of cases consistent with tuberculosis, findings comparable to those of Batirel et al. [20]. Percutaneous vertebral biopsy is regarded as a safe, accurate, and relatively cost-effective diagnostic technique for vertebral osteomyelitis. One study [25]found that three-quarters of spinal tuberculosis patients underwent invasive procedures, such as bone biopsy, to confirm the diagnosis. Reports indicate that among biopsy patients, approximately 60% were diagnosed with spinal tuberculosis through microbiological and/or histopathological examination [13]. Thus, we recommend performing percutaneous vertebral puncture biopsy early in suspected spinal tuberculosis cases, prior to initiating anti-tuberculosis therapy, to confirm diagnosis and guide treatment.

Xpert is a diagnostic tool recommended by WHO. It is widely employed for the rapid and accurate diagnosis of clinical tuberculosis. Additionally, it aids in guiding the treatment of drug-resistant tuberculosis. However, one limitation of Xpert is its inability to differentiate between live and dead bacilli. In this study, the diagnostic sensitivity of Xpert for spinal tuberculosis was 71.84%, consistent with findings from other research. Jin et al. [26]reported a sensitivity of 77.78%, while Karthek et al. [27] observed a sensitivity of 65.1%, both supporting the high diagnostic value of Xpert for spinal tuberculosis. This study compared Xpert and pathological examination results to evaluate their roles in diagnosing spinal tuberculosis. Results demonstrated that, using comprehensive diagnosis as the gold standard, the positive rate of pathological detection surpassed that of Xpert. When pathological diagnosis was used as the gold standard, Xpert exhibited a sensitivity of 81.5% (268/329), a specificity of 58.8% (57/97), a positive predictive value of 87.0% (268/308), and a negative predictive value of 48.3% (57/118). The concordance between histopathology and Xpert detection was low, with a kappa coefficient of 0.374. When the two tests were combined, the diagnostic positivity rate reached 86.6%, significantly outperforming the use of either test alone (P< 0.001), demonstrating high diagnostic value for spinal tuberculosis, consistent with findings from other studies [28, 29]. Therefore, when adequate specimen volume is available, we recommend that both pathological examination and Xpert testing be conducted concurrently.

In this study, the initial anti-tuberculosis treatment regimen for patients with spinal tuberculosis was the HRZE combination, and the total drug resistance rate reached 6.49%, showing a yearly increasing trend, with single-drug resistance being the predominant type. The presence of caseous necrotic tissue or sclerotic bone surrounding the lesion impeded local blood supply, making it challenging for adequate drug concentrations to accumulate at the lesion site, thereby compromising therapeutic efficacy [30]. Surgical intervention serves as an adjunctive treatment for spinal tuberculosis, primarily involving posterior internal fixation combined with anterior lesion removal and intervertebral fusion. The goals of surgery are to thoroughly excise lesions, achieve adequate neural decompression, correct deformities such as scoliosis or kyphosis, restore spinal stability, and augment the efficacy of anti-tuberculosis pharmacotherapy. In this study, 648 patients underwent surgical treatment, followed by regular, sufficient, and individualized anti-tuberculosis drug therapy, resulting in favorable outcomes.

Limitation

The data in this study are confined to the epidemiological characteristics of spinal tuberculosis patients treated at a single hospital, which limits the generalizability of the findings. Additionally, the retrospective design of the study may introduce selection bias and information bias, further affecting the interpretation of certain variables such as treatment response and outcome evaluation. The findings may also vary depending on geographic regions, lifestyle factors, economic conditions, healthcare resources, and policies. Therefore, multi-regional collaborative statistical analysis is necessary to accurately determine the epidemiological status of spinal tuberculosis across different regions, which would inform the development of targeted prevention and control strategies.

Conclusion

Over the course of the study period, the overall diagnosis rate of spinal tuberculosis exhibited an upward trend, and a gradual decline in the annual growth rate. Early identification and accurate diagnosis of spinal tuberculosis remain challenging. Bacteriological and pathological diagnosis remain the gold standards for diagnosing spinal tuberculosis. Xpert has demonstrated high diagnostic value and, when combined with other diagnostic methods, can significantly enhance the clinical diagnosis rate of spinal tuberculosis. The primary treatment for spinal tuberculosis remains anti-tuberculosis chemotherapy, which may be supplemented by surgical intervention. However, the duration of anti-tuberculosis drug therapy should not be adjusted based on whether the patient has undergone surgical debridement. Despite East China’s relatively advanced socioeconomic and healthcare systems, spinal tuberculosis remains a substantial public health challenge, primarily due to the region’s complex population composition and high population mobility. The prevention and management of spinal tuberculosis continue to present considerable challenges. Increasing the allocation and investment of medical resources for tuberculosis, enhancing health management for migrant populations, and raising public health awareness are essential.

Abbreviations

STB

Spinal Tuberculosis

IQR

Interquartile Range

PTB

Pulmonary TB

EPTB

Extrapulmonary TB

WHO

World Health Organization

CRS

Composite Reference Standard

NICE

National Institute for Health and Care Excellence

TB-Ab

Tuberculosis Antibody

HPE

Histopathological Examination

Xpert

GeneXpert-MTB/RIF assay

ESR

Erythrocyte sedimentation Rate

CRP

C-reactive Protein

ASIA

American Spinal Injury Association

Authors’ contributions

Manjiang Cao: Writing- Original draft preparation, Formal analysis, Visualization. Shuya Jiao: Writing- Original draft preparation, Formal analysis, Visualization. Quan Zhang: Data curation, Writing- Reviewing and Editing. Bo Zhu: Data curation, Writing- Reviewing and Editing. Shiyuan Shi: Conceptualization, Methodology, Project administration, Writing- Reviewing and Editing.

Funding

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Data availability

The datasets used and analyzed during the current study are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

This study was approved by the Medical Ethics Committee of Hangzhou Red Cross Hospital. We certify that the study was performed in accordance with the 1964 declaration of HELSINKI and later amendments. Written informed consent was obtained from all the participants prior to the enrollment of this study.

Consent for publication

Written informed consent was obtained from the patients presented in Case 1 and Case 2 for the publication of their clinical details and identifying images.

Competing interests

The authors declare no competing interests.