Efficacy of circumferential surgery for multisegmental pyogenic cervical spine infection

Abstract

Objective

To evaluate the efficacy of circumferential surgery, involving combined anterior and posterior approaches, in the management of multisegmental cervical pyogenic spondylitis (CPS).

Methods

A retrospective analysis was conducted on 12 patients with multisegmental CPS who underwent circumferential surgical intervention between January 2015 and November 2023. The study evaluated medical records, spinal alignment correction, postoperative complications, and neurological functional recovery.

Results

In total, 12 patients (8 men and 4 women; mean age: 57.1 years) met the inclusion criteria. Identified risk factors included diabetes mellitus in four patients, a history of substance abuse in two patients, and long-term immunosuppressive therapy following organ transplantation in one patient. Preoperatively, nine patients presented with spinal cord or nerve root compression. All patients underwent circumferential surgery with anterior column reconstruction using autologous bone grafts; titanium alloy plates were additionally used in two cases. Posterior stabilization was achieved using pedicle screws or lateral mass screws. The mean number of corpectomies performed was 3.7 (range, 3–5); the mean number of fused levels was 4.8 (range, 4–7). The mean follow-up duration was 19.4 (range, 6–38) months. Cervical alignment improved significantly, with the mean C2-7 Cobb angle changing from − 5° preoperatively to 3° postoperatively. Neurological function significantly improved in all patients (P = 0.023), and successful bone fusion was achieved in 11 patients (91.7%) at the final follow-up. Pathogens were identified in seven cases, including Staphylococcus aureus Staphylococcus epidermidis (n = 2), Klebsiella pneumoniae (n = 1), and Escherichia coli (n = 1). All patients received intravenous antibiotic therapy for 4–6 weeks postoperatively, followed by oral antibiotic treatment for an additional 6–8 weeks. No intraoperative complications were reported. Postoperatively, one patient developed pneumonia with respiratory failure, and another experienced posterior cervical wound infection.

Conclusions

Circumferential surgical intervention combined with appropriate antibiotic therapy appears to be an effective treatment strategy for achieving satisfactory outcomes in patients with multisegmental CPS.

Introduction

Pyogenic spondylitis (PS) is an infectious disorder involving the vertebrae, intervertebral discs, and adjacent soft tissues and includes conditions such as pyogenic spondylodiscitis (PSD), spinal epidural abscess, and pyogenic vertebral osteomyelitis [1–3]. The lumbar spine is the most commonly affected region, followed by the thoracic and cervical spines. Cervical involvement is relatively uncommon, accounting for only 3–6% of all PS cases [4–5]. However, the incidence of cervical PS (CPS) has increased in recent years, likely owing to population aging, the growing prevalence of immunocompromised states, and the increased use of invasive procedures associated with bacteremia [6–7].

In the early stages of CPS, treatment primarily consists of culture-guided antibiotic therapy combined with external immobilization. Nonetheless, early diagnosis and treatment of CPS remains challenging due to its insidious onset and nonspecific clinical manifestations [2, 4, 5, 7]. Delayed diagnosis may lead to progressive multilevel cervical spine deterioration, kyphotic deformity, and severe neurological impairment. Patients with multilevel cervical involvement are particularly susceptible to kyphosis and paraplegia. Consequently, surgical intervention is frequently required to address neurological deficits, restore spinal stability, and prevent vertebral collapse or kyphosis progression [8–9]. Alexander A et al. demonstrated favorable therapeutic outcomes in the treatment of multisegmental CPS using anterior cervical debridement combined with anterior fixation. However, postoperative graft or cage displacement occurred in two patients. Furthermore, the safety and appropriateness of instrumentation placement in an infected surgical field remain controversial. In the present study, 12 patients with multisegmental CPS underwent radical debridement through an anterior cervical approach, followed by anterior reconstruction using autologous bone grafts combined with posterior fixation using cervical pedicle screw or lateral mass screws (LMS). This strategy achieved satisfactory therapeutic outcomes.

Materials and methods

Study population

This study was approved by the Ethics Committee of our hospital. Patients diagnosed with CPS between January 2015 and November 2023 in the spinal department of our hospital were retrospectively analyzed. The inclusion criteria were patients aged > 18 years with prior surgical intervention for multilevel CPS, defined as lesions involving at least two motor units, by using circumferential surgical techniques (combined anterior and posterior approaches). The exclusion criteria were patients necessitating reoperation due to secondary infections following cervical spine surgery and those presenting with concurrent infectious diseases that affected other organ systems. The diagnosis was corroborated through intraoperative observations, culture results, or postoperative pathological examinations. The surgical indications identified in our study encompassed multilevel CPS leading to neurological deficits, severe neck pain, pronounced cervical kyphosis, and abscess formation that affected adjacent organs and was refractory to pharmacological interventions. Thereafter, a comprehensive review of the patients’ medical records was conducted, encompassing demographic information, medical comorbidities, inflammatory markers, surgical interventions, microbiological findings, neurological status, perioperative complications, and follow-up outcomes.

Surgical technique

Following intubation under general anesthesia, the patient was positioned prone with the cervical spine in a neutral position using continuous skull traction. A standard posterior midline incision was created, extending one or two vertebrae above and below the affected segments. The paraspinous muscles were dissected to expose the laminae and lateral mass. Depending on the extent of vertebral damage, cervical pedicle screws or LMS were inserted into one or two normal vertebrae above and below the affected area. A pre-bent rod was then installed to correct kyphosis and stabilize the cervical spine. In cases of abscess formation behind the spinal canal or cervical spinal stenosis, partial laminectomy and decompression were performed. After confirming satisfactory internal fixation and kyphosis correction via C-arm fluoroscopy, a drainage tube was placed, and the incision was closed in layers.

After removing the skull traction, the patients were positioned supinely with a slight cervical extension. A transverse skin incision was made on the right side of the anterior cervical region, followed by a layered dissection of the subcutaneous tissue and platysma. The prevertebral space was accessed through the interspace between the medial border of the vascular sheath and the lateral border of the visceral sheath. To remove the purulent material, either a total discectomy or corpectomy was conducted after exposing the disc space. The excised specimen was then submitted for bacterial culture and histopathological examination. Following extensive irrigation with saline solution, the structural autogenous bone was implanted into the vertebral gap so as to reconstruct the anterior column defect without or with stabilization through anterior cervical plating. The bone graft material was sourced from the iliac crest or the sternal manubrium [10]. After confirming satisfactory internal fixation and correction of kyphotic deformity using C-arm fluoroscopy, a drainage tube was inserted, and the incision was meticulously closed layer-wise.

Postoperative management and follow-up

The drainage tube was removed once the drainage volume reduced to < 50 mL in 24 h. Postoperative nutritional support was administered to address anemia and hypoproteinemia. All patients received intravenous antibiotic therapy for 4–6 weeks, as per bacteriological findings, followed by administration of oral antibiotics for an additional 6–8 weeks after operation [7–9, 11]. For patients lacking bacteriological results, broad-spectrum antimicrobials were selected based on the clinical expertise of physicians in the Department of Internal Medicine specializing in infectious diseases. After the surgery, a rigid cervical brace was applied and maintained for 12 weeks.

During the first 3 months after the operation, routine blood tests, erythrocyte sedimentation rate (ESR), and C-reactive protein (CRP) levels were monitored monthly to evaluate the infection status. Radiographic assessments, including plain films or computed tomography (CT) scans, were conducted at 3, 6, and 12 months post-operation to investigate the extent of bone fusion. The American Spinal Injury Association (ASIA) score was applied to assess neurological function both preoperatively and postoperatively. The pain severity was determined using the Visual Analog Scale (VAS). The kyphosis angle was evaluated preoperatively and postoperatively by using the standard method for measuring the C2–C7 Cobb angle [12]. The Bridwell grading system was applied to determine the interbody bone graft fusion rate in this study. Imaging data from the 12 patients were first anonymized and assigned numerical identifiers, and then two researchers independently performed the grading. Inter-observer discrepancies, if any, were analyzed to assess the consistency of the Bridwell grading between the researchers.

Statistical analysis

Data analysis was conducted utilizing SPSS for Windows (version 19.0; IBM, New York, NY, USA). The study primarily focused on paired comparisons of preoperative and postoperative data from the cohort of 12 patients. The normality of differences in continuous variables was assessed using the Shapiro–Wilk test. For the comparison of VAS scores, ESR, and Cobb angles, paired sample t-tests were employed. The Wilcoxon signed-rank test was applied to analyze the CRP levels and to evaluate changes in the ASIA grading so as to assess the neurological function status. Inter-observer variability was evaluated using the intraclass correlation coefficient (ICC), which was calculated based on a two-way random-effects model with absolute agreement, averaged across measurements. p < 0.05 indicated statistical significance.

Results

In total, 12 patients with multisegmental CPS who underwent circumferential surgical intervention between January 2015 and November 2023 were identified from our institutional surgical database. The cohort included eight men and four women, with a mean age of 57.1 (range, 36–74) years. Four patients presented with fever and sepsis at admission. According to the ASIA classification, three patients were classified as grade B, one as grade C, five as grade D, and three as grade E. Identified risk factors included diabetes mellitus in four patients, a history of substance abuse in two patients, and long-term immunosuppressive therapy following organ transplantation in one patient. Involvement of three cervical vertebral levels was observed in seven patients, four levels in four patients, and five levels in one patient (Table 1).

Table 1.

Demographics and comorbidities of patients

Case	Gender	Age	Risk factor	Level	Fever	CRP(mg/L)	ESR(mm/h)	Neurologic deficites	VAS	Cobb
1	M	64	DM	C2-C5	YES	45	120	D	8	−10
2	M	36	DA	C2-4	YES	34	61	D	6	−1
3	F	73	None	C4-6	NO	27	56	B	4	−9.8
4	M	64	None	C3-7	NO	66	102	D	7	0
5	F	71	DM	C4-6	YES	17	70	D	6	−4
6	F	70	IP	C4-6	NO	60	90	E	5	−20
7	M	40	DA	C6-T1	NO	30	30	E	4	−5
8	F	35	DM	C3-5	NO	40	54	E	6	−19
9	M	63	None	C5-7	NO	46	77	C	5	−5
10	M	56	None	C3-6	NO	50	109	D	6	4
11	M	43	DM	C2-5	NO	56	88	B	5	−10
12	M	71	None	C3-6	YES	191	120	B	6	−2

DM Diabetes mellitus, DA Drug abusers

All patients underwent single-stage circumferential surgery using a posterior-anterior approach. Posterior fixation with PS was performed in seven patients, LMS fixation in two patients, and combined PS and LMS fixation in three patients. Anterior column reconstruction was supplemented with titanium plating in 2 cases, whereas reconstruction with bone grafting alone was performed in 10 cases. Autologous grafts were harvested from the iliac crest in seven patients and from the sternal manubrium in five patients. The mean operative time was 318 (range, 240–420) min, and the average intraoperative blood loss was 595 mL (range, 350–1200 mL). Microbiological cultures were obtained from all patients, with pathogens identified in seven cases. The isolated organisms included Staphylococcus aureus in three cases, Staphylococcus epidermidis in two cases, Escherichia coli in one case, and Klebsiella pneumoniae in one case. Histopathological findings were generally nonspecific (Table 2).

Table 2.

Surgical Data of patients

Case	Surgical method	Operative time(min)	Blood loss (ml)	Culture result
1	PS: C2-6 + C2/3,C3/4 discectomy+C4, 5 corpectomy + I	360	1000	Staphylococcus aureus
2	PS: C2 + LMS:3–5 + C2/3,C3/4 discectomy + I	260	400	Negative
3	LMS: C3-C7 + C5 corpectomy + I+TP	270	400	Staphylococcus aureus
4	PS: C2-LMS: C3-7 + C3/4,C4/5,C5/6,C6/7 discectomy + I	420	400	Staphylococcus epidermidis
5	PS: C3-C7 + C3/4 discectomy+C5 corpectomy + I+TP	330	300	Negative
6	PS: C4-7 + C5 corpectomy + M	300	400	Staphylococcus epidermidis
7	PS: C5-T2 + C7 corpectomy + M	420	1200	Negative
8	PS: C2-6 + C4 corpectomy + M	240	1000	Klebsiella pneumoniae
9	PS: C5-T1 + C6 corpectomy + M	300	400	Negative
10	LMS:3–7 + C3/4 discectomy+C5 corpectomy + I	260	350	Escherichia coli
11	PS: C2 + LMS: C3-6 + C2/3 discectomy+C4 corpectomy ± I	300	600	Negative
12	PS:3–7 + C3/4,C4/5 ,C5/6 discectomy + M	360	700	Staphylococcus aureus

PS Pedicle screw, LMS Lateral mass screw, TP Titanium plate, I Iliac graft, M Manubrium graft

The mean follow-up duration was 19.4 (range, 6–38) months. Except for one patient who died of cerebral infarction 6 months postoperatively, all remaining patients achieved successful bone fusion at the final follow-up (Fig. 1). Interobserver reliability for Bridwell fusion grading was excellent, with an ICC of 0.81 (95% confidence interval: 0.72–0.88). At the final follow-up, ESR and CRP levels had normalized in all patients. VAS scores for cervical and limb pain significantly improved compared with preoperative values (t = 10.346, P < 0.001). The mean C2-7 Cobb angle improved from − 5° preoperatively to 3° postoperatively. According to the ASIA classification, all patients demonstrated an improvement of 1–2 grades in neurological function (Z = − 2.271, P = 0.023). No intraoperative complications, including vascular injury or dural tear, were encountered. Postoperatively, one patient developed pneumonia with respiratory failure, which improved after medical management. Another patient developed a posterior cervical wound infection that resolved with local wound care and antibiotic therapy (Table 3).

Fig. 1.

A 64-year-old male with C2–C5 cervical pyogenic spondylitis after radio frequency ablation of cervical spine. A The lateral X-ray film before surgery demonstrated vertebral bone destruction from C2 to C5 with prevertebral soft tissue thickening. B, C Preoperative CT and MRI showed that C2–5 vertebral bone destruction with paravertebral abscess and cervical spinal cord compression. D–F Postoperative image show that the patient underwent cervical pedicle screw fixation from C2 to C6 ,combined anterior debridement and structural iliac bone graft at C2-3,C3-4 and C4-5

Table 3.

Postoperative follow-up data of patients

case	Follow-up(m)	Neurologic deficites*	VAS#	ESR	CRP	Cobb	Bony fusion	complications
1	24	D	4	16	4	5	Ⅰ	No
2	36	E	3	10	12	4	Ⅰ	No
3	36	D	1	18	5	5	Ⅰ	No
4	24	D	2	22	10	13	Ⅰ	No
5	12	E	0	19	8	3	Ⅰ	Respiratory failure+pneumonia
6	12	E	3	27	15	0	Ⅰ	No
7	12	E	1	28	12	0	Ⅰ	No
8	38	E	1	23	7	1	Ⅰ	Wound infection
9	6	E	1	17	14	3	Not available	Die(Cerebral infarction
10	12	E	1	22	6	0	Ⅰ	No
11	9	D	3	21	6	5	Ⅱ	No
12	12	D	1	21	13	0	Ⅰ	No

^* The Wilcoxon signed-rank test was employed to evaluate the differences in neurological function improvement pre- and post-surgery. Z=-2.271 P = 0.023

^# A paired sample t-test was conducted to evaluate the differences in VAS scores for neck pain before and after surgical intervention. t = 10.346 P<0.001

Discussion

Nonsurgical management, comprising intravenous antibiotic therapy combined with external immobilization, has traditionally been considered the first-line treatment for CPS [7, 12–15]. In the early stages of the disease, conservative treatment may successfully achieve infection control, pain relief, spinal stability preservation, and neurological impairment prevention in selected patients. However, before the development of neurological deficits, the clinical manifestations of CPS are often subtle and nonspecific, frequently resulting in delayed diagnosis. Patients older than 65 years with spondylodiscitis present with fewer classic infectious symptoms and signs [16]. Delayed diagnosis and treatment may therefore lead to severe complications, including sepsis and death, particularly in elderly and immunocompromised patients. Furthermore, CPS can progress rapidly and result in devastating neurological impairment [8]. Accordingly, early surgical intervention is recommended in patients with CPS presenting with large epidural abscesses, significant neurological deficits, spinal instability secondary to bony destruction, or refractory neck pain.

The primary objectives of surgical treatment for CPS include excision of infected tissue, spinal alignment restoration, and spinal stabilization [8–14]. Currently, no consensus has been established regarding the optimal surgical approach; moreover, the majority of spine surgeons choose procedures based on personal experience and preference. Because CPS commonly involves both the vertebral bodies and intervertebral discs, the anterior approach is widely regarded as advantageous. It provides direct access to the infected vertebral bodies and disc spaces, facilitates thorough debridement, and allows bone graft placement for spinal stabilization. Additionally, this approach enables adequate tissue sampling for reliable microbiological diagnosis [5, 9, 12, 17]. Heyde et al. [18] reported favorable outcomes in 20 patients with cervical spondylodiscitis treated with anterior radical debridement, autologous bone grafting, and instrumentation, with all patients achieving successful bone fusion and no evidence of recurrent infection.

In patients with infection involving multiple vertebral levels, anterior-only surgical approaches may not provide sufficient biomechanical stability. Burkhardt et al. reported 30 cases of CPS treated with anterior surgery and found that 6 patients (20.7%) required revision surgery within the first 2 weeks postoperatively [5]. The reported complications included loosening and subsidence of fixation or fusion devices in four patients, screw malposition in one patient, and wound nonunion in another. The authors attributed these early implant failures primarily to poor bone quality caused by infection-related destruction. Similarly, Alexander et al. described 19 patients with subaxial cervical osteomyelitis who underwent multilevel corpectomy with anterior column reconstruction and anterior plating. The overall postoperative complication rate was 26%, including implant loosening in two cases, hematoma requiring revision surgery in two cases, and wound infection in one case. These findings suggest that multilevel anterior corpectomies with anterior plate fixation alone are associated with an increased risk of graft or instrumentation failure. Consequently, supplemental posterior fixation is recommended to enhance construct stability in cases involving more than two-level corpectomies.

Posterior debridement, stabilization, and fusion have demonstrated favorable outcomes in thoracolumbar PSD [11]. However, reports describing posterior-only surgical treatment for CPS remain limited. Kono et al. [19] reported successful management of two patients with cervical spondylodiscitis secondary to osteoradionecrosis following radiotherapy using posterior fixation and fusion combined with a six-month course of antibiotic therapy. In contrast, Zhang et al. [20] described a 44-year-old man with CPS and spinal epidural abscess-associated quadriplegia who underwent emergency posterior decompression and fixation combined with antibiotic therapy, resulting in significant improvement in neurological function. However, progressive anterior vertebral destruction developed 3 weeks later, ultimately necessitating anterior debridement, decompression, and fusion. Therefore, we believe that posterior-only internal fixation and fusion may be insufficient for the treatment of patients with CPS. This approach may be appropriate only in specific conditions: patients with cervical epidural abscesses without significant vertebral collapse or spinal cord compression, patients who are not suitable candidates for anterior surgery, or cases in which the infection is confined to the upper cervical spine.

The combined posterior-anterior surgical approach provides enhanced biomechanical stability and facilitates restoration of spinal alignment through posterior fixation using pedicle screws or LMSs. Moreover, this circumferential strategy enables radical debridement of infected tissue and reconstruction of anterior column defects using autologous bone grafts [2, 4–5]. In the present study, all 12 patients with CPS underwent combined posterior-anterior surgery, resulting in a mean operative time of 318 (range, 240–420 min) and a mean intraoperative blood loss of 595 mL. All patients had satisfactory clinical outcomes, with an average follow-up period of 19 months. Nevertheless, the combined posterior-anterior surgical approach has certain limitations. Notably, this combined procedure is associated with longer operative time and greater intraoperative blood loss than single-approach procedures, potentially leading to an increased risk of postoperative complications [11, 21–22]. Furthermore, patients with poor general health or significant medical comorbidities may not tolerate extensive combined procedures. Therefore, we believe that circumferential surgery should be reserved for carefully selected patients with appropriate surgical indications. In our series, these indications included multilevel CPS, cases requiring two-level corpectomy or three-level discectomy, and patients with severe osteoporosis.

Previous studies have reported postoperative complication rates ranging from 13% to 33% following surgical treatment of cervical spinal infections, including implant loosening, graft or cage subsidence, hematoma formation, and wound infection [4, 8, 17, 23]. In contrast, the complication rate in our study was 16%, which was lower than that reported in previous studies. This relatively low complication rate may be associated with several perioperative management strategies. First, a one-stage combined posterior-anterior procedure was performed only in patients with stable general conditions, whereas staged surgery was considered for patients with poor overall health to reduce surgical risk. Additionally, after posterior fixation, the extent of anterior vertebral resection was minimized whenever possible, thereby decreasing the graft length and shortening the fusion time [20–21]. Second, autologous bone grafts were preferentially used for anterior column reconstruction due to their osteoconductive, osteoinductive, and osteogenic properties, which facilitate early and successful fusion. Moreover, anterior plate fixation after corpectomy was avoided in all but three patients. Third, correction of anemia and hypoproteinemia was performed to improve patients’ overall condition and resistance to infection. Fourth, inflammatory markers, such as CRP and ESR, were monitored dynamically to assess the infection status, and antibiotic regimens were adjusted based on bacterial culture and sensitivity results. Finally, the posterior surgical site was routinely irrigated for 3–5 days postoperatively to reduce the local burden of pathogens and inflammatory mediators.

This study has several limitations. First, this was a single-center retrospective study with a relatively small sample size, and therefore, the findings require further validation through multicenter studies with larger patient cohorts. Second, because this study was designed as a case series without a control group, further studies with larger cohorts and appropriate control groups are needed to compare the efficacy of various surgical approaches. Lastly, due to the low incidence of multisegmental CPS and the prolonged duration required for case collection, despite efforts to maintain consistency among the included cases, it remains challenging to eliminate potential bias arising from variations in surgical technique and surgeon experience that may have influenced the study outcomes.

Conclusions

For patients with multilevel CPS, combined anterior debridement, autogenous bone grafting, and posterior instrumentation, together with antimicrobial therapy, may represent an effective treatment strategy. However, due to the high risk of complications associated with this approach, combined surgical intervention should be reserved for patients with clear surgical indications.

Authors’ contributions

W.Q and W.S developed the concept, supervised the project and conducted data analysis. X.S and Z.S conducted the majority of the study , analysed data and prepared the manuscript. X.S and M.Z.Y conducted sample collection and performed statistical analysis.

Funding

This work is jointly funded by Sichuan Science and Technology Program (NO.2022-YFS0628) and Natural Science Foundation of Sichuan (NO.2023NSFSC0333). Research Program of the Affiliated Hospital of Southwest Medical University（NO.24066.

Data availability

The data that support the findings in this study are available from the corresponding author upon request.

Declarations

Ethics approval and consent to participate

This study involved human participants and was conducted in full accordance with the Declaration of Helsinki. This study was approved by the Ethics Committee for Clinical Trials of the Affiliated Hospital of Southwest Medical University (KY2026088). Informed consent was obtained from all individual participants included in the study.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.