Surgical technique and outcomes for tumors simultaneously involving the thoracic vertebrae and posterior chest wall

Abstract

Background

Tumors that simultaneously involve the thoracic vertebrae and posterior chest wall are rare and pose clinical challenges for musculoskeletal oncologists. The optimal treatment modality for these tumors remains poorly defined. This study aimed to explore the clinical outcomes of patients treated at our center and determine effective multidisciplinary therapeutic approaches.

Methods

Between May 2019 and December 2023, 26 patients with tumors involving both the thoracic vertebrae and posterior chest wall who received multidisciplinary therapy centered on en bloc resection and reconstruction surgeries at our center were identified from the hospital database. Patient demographics, tumor characteristics, treatment history, surgical outcomes, and complications were collected from electronic medical records.

Results

En bloc resection was successfully performed in all patients, and none underwent intralesional resection. Wide or marginal margins were achieved in all cases. The median follow-up duration was 14 months (range 6 − 42 months). Local recurrence was recorded in only one patient. Five patients experienced wound complications that were successfully treated with debridement and antibiotics. Respiratory complications were observed in five patients, who recovered after respiratory function exercises combined with antibiotics. No neurological complications, implant failure, hemothorax, or thrombosis were observed during follow-up.

Conclusions

Thorough preoperative evaluation and accurate pathological diagnosis are important for tumors involving both the thoracic vertebrae and posterior chest wall. Additionally, comprehensive treatments centered on en bloc resection and reconstruction surgeries are feasible and effective.

Background

Adequate surgical resection represents a crucial precondition for achieving the best local control in patients with primary malignant, solitary metastatic, and locally aggressive benign bone tumors [1, 2]. En bloc resection, which aims to ensure a tumor-free margin, is the recommended procedure for these tumors [3].

When tumors simultaneously involve the thoracic vertebrae and posterior chest wall, en bloc resection is challenging for musculoskeletal oncologists owing to the complex anatomical structures involved, such as the spinal cord, nerve roots, dura, pleura, lungs, major vessels, and bony structures [4, 5]. Additionally, after tumor resection, the reconstruction of thoracic spinal stability and chest wall integrity with sufficient coverage is technically difficult.

Only a few case reports have described en bloc spondylectomy of thoracic spinal tumors and extended surgical resection of the chest wall locally invaded by lung cancers [6–9]. The optimal treatment modality for primary malignant, solitary metastatic, or locally aggressive benign tumors involving both the thoracic vertebrae and posterior chest wall remains unclear. Therefore, we reviewed corresponding patients treated at our institution and aimed to explore the clinical outcomes of multidisciplinary therapeutic approaches centered on en bloc resection and reconstruction surgery. Operative techniques for different patients are also discussed.

Methods

Patients

Between May 2019 and December 2023, 26 consecutive patients (12 males and 14 females; average age: 44.3 [range: 22–69] years) with primary malignant, solitary metastatic, or locally aggressive benign tumors involving both the thoracic vertebrae and posterior chest wall who underwent en bloc resection and reconstruction surgeries in our center were retrospectively identified from the hospital database (Table 1). Patients with multiple metastatic tumors, unresectable tumors, or tumors originating from major blood vessels were excluded. Patient demographics, surgical data, treatment history, and clinical outcomes were collected from electronic medical records. The main complaints were back pain in 15 patients, occasional intercostal neuralgia in 1 patient, and shortness of breath in 1 patient, while the remaining patients were asymptomatic and diagnosed during physical examinations. Although nine patients had grade 2 or 3 epidermal spinal cord compression, no apparent sensory or motor deficits of the lower limbs were found preoperatively. The most common diagnoses were chondrosarcoma (n = 6), giant cell tumors (n = 4), and osteosarcoma (n = 4). The surgeon provided a detailed written informed consent for the operation to the patient preoperatively, and informed consent was obtained from all study participants. Patient data were collected independently from the participants and blindly analyzed. This study was approved by the Institutional Review Board of our center.

Table 1.

Demographic and characteristics of the patients

Case	Age at surgery (years)	Gender	Pathology	Primary /Metastasis	Frankel grade	ESCC	Tomita type
1	67	Female	Chondrosarcoma	Recurrent	D	3	V
2	49	Male	Chondrosarcoma	Primary	E	1b	VI
3	22	Male	Ewing sarcoma	Recurrent	E	0	VI
4	67	Female	Metastatic/Lung	Metastasis	E	1b	VI
5	30	Male	Osteosarcoma	Primary	E	3	VII
6	49	Female	Schwannoma	Primary	E	2	V
7	39	Male	Chondrosarcoma	Primary	E	2	VI
8	69	Male	Undifferentiated sarcoma	Primary	E	0	VI
9	50	Male	Undifferentiated sarcoma	Primary	E	2	VI
10	65	Female	Malignant mesothelioma	Primary	E	1c	VI
11	61	Female	Chondrosarcoma	Primary	E	0	VI
12	49	Female	Metastatic/Leiomyosarcoma	Metastasis	E	0	V
13	32	Male	Giant cell tumor	Primary	E	1b	VII
14	29	Female	Giant cell tumor	Primary	E	0	VI
15	40	Male	Metastatic/Colon	Metastasis	E	2	VI
16	30	Female	Osteosarcoma	Primary	D	3	VI
17	59	Female	Chondrosarcoma	Primary	E	0	VI
18	47	Male	Schwannoma	Primary	E	1c	V
19	38	Male	Hemangioma	Primary	E	2	V
20	41	Female	Unspecified high-grade sarcoma	Primary	E	0	VI
21	65	Female	Hemangioma	Primary	D	3	V
22	45	Male	Giant cell tumor	Primary	E	1b	VI
23	22	Female	Giant cell tumor	Primary	E	0	VII
24	28	Male	Osteosarcoma	Recurrent	E	0	V
25	23	Female	Chondrosarcoma	Primary	E	0	VI
26	36	Female	Metastatic/Osteosarcoma	Metastasis	E	0	V

Preoperative preparation

Routine blood tests, plain radiographic examinations, computed tomography (CT), magnetic resonance (MR) enhancement scans of the spine and chest, and technetium 99 m bone scintigraphy (or positron emission tomography/CT in some cases) were performed. After appropriate imaging, a needle biopsy was performed to assist in the preoperative diagnosis. Radiological evaluation, pathological diagnosis, and treatment planning for all patients were discussed by the musculoskeletal oncology multidisciplinary team.

European Society for Medical Oncology (ESMO) sarcoma clinical guidelines were followed. Neoadjuvant chemotherapy using standard protocols was administered to six patients with osteosarcoma, Ewing sarcoma, and undifferentiated sarcoma. Preoperative denosumab (120 mg on days 0, 7, 15, and 28) was administered to four patients with giant cell tumors [10]. The remaining patients did not receive neoadjuvant therapy. No preoperative radiation was administered. We carefully evaluated and designed the surgical procedures, and the goal of achieving R0 resection whenever possible.

Surgical procedures

Regarding the surgical plan, a single posterior approach was preferred by most patients (23). Patients were placed in the prone position under general anesthesia. A double-lumen endotracheal tube was inserted to allow selective lung collapse. After a dorsomedial skin incision, the paraspinal muscles were detached from the spinous processes, laminae, and facet joints. To adhere to oncological principles, the vertebrae and the parts of the chest wall with direct tumor invasion were resected en bloc. Based on our previous study, total en bloc spondylectomy (TES) can be performed when the tumor crosses the vertebral midline, and sagittal en bloc resection (SEBR) can be adopted to incise the part of the vertebral body with tumor involvement when the tumor is confined to the vertebral midline, according to our novel surgical classification system [11]. On the side of the affected chest wall, the parts of the chest wall with tumor invasion were resected at least 3 cm from the tumor. After dissecting the normal surrounding tissue, the ribs were exposed and amputated, and the residual soft tissue was mobilized for coverage. An additional transverse incision (T-shaped incision) was made in 13 patients with wide rib or pleural tumor invasion that was difficult to resect through a single dorsomedial incision. The T2-T12 nerve roots were removed when necessary. Parts of the adjacent structures, including the paraspinal muscles, pleura, lungs, and diaphragm, also needed to be resected to acquire safe margins. A combined anterior-posterior approach was used in three patients because it was difficult to separate the tumors from important organs through a single posterior approach. Surgical specimens were sent for C-arm fluoroscopy and histopathological examination immediately after surgery to confirm resection margins according to the Enneking classification. On the spinal side, because there is no strict anatomic barrier to prevent tumor cell infiltration, positive residual tumor margins were classified as intralesional resection, tumors extending within 1 mm of the resection margins without grossly positive wound contamination were classified as marginal resection, and tumors extending > 1 mm of the resection margins were classified as wide resection [12, 13].

To restore spinal continuity and stability after tumor resection, pedicle screw and rod fixation were routinely performed at levels above and below the resected vertebra. The choice of the anterior reconstruction method for the resected vertebral body was based on our classification system [11]. To reconstruct the mechanical structure and realize bone ingrowth, anterior reconstruction was performed in 21 patients using a three-dimensional (3D) printed or standardized artificial vertebral body (AVB) or titanium mesh cage filled with autograft. The “sandwich” method (methyl methacrylate sandwiched between two layers of Gynemesh) was used to reconstruct the chest wall according to the severity of the defects (defects > 5 cm in diameter or including > four ribs). The myocutaneous flap was tightly sutured to ensure optimal for better closing of the thoracic cavity closure and implant coverage. When extensive resection of the paraspinal or chest wall muscle was required, soft tissue reconstruction with various flaps was strongly recommended. Closed thoracic drainage tubes were placed when the pleural cavity was incomplete, and drainage tubes were placed at the incision site in all patients. The representative cases are shown in Figs. 1 and 2.

Fig. 1.

(Case 2) A 49-year-old male presented with increasing back pain during the previous 3 months. (A-D) Neoplasm involving T9-11 and chest wall, including the 9th–11th ribs, was revealed by coronal and axial MRI, CT. Biopsy confirmed the diagnosis of chondrosarcoma. (E) The surgical plan of osteotomy. Through a T-shaped incision, en bloc resection of the tumor, including TES in T10 as well as SEBR in T9 and T11, was performed with reconstruction of the vertebral column and chest wall. (F) Intraoperative image showing allograft bone, AVB, and screw-rod internal fixation system used for reconstruction after the vertebral body resection. (G) The “sandwich” method was used to reconstruct the defective chest wall. (H) Completion of en bloc resection, with the surgical specimen shown. (I) X-ray of the specimen confirmed en bloc resection of the tumor, partial vertebral body, and back chest wall. (J, K) Postoperative X-ray and CT showing satisfactory position of the AVB, allograft bone, screw-rod internal fixation system, and “sandwich” plate. Abbreviations: MRI, magnetic resonance imaging; CT, computed tomography; TES, total en bloc spondylectomy; SEBR, sagittal en bloc resection; AVB, artificial vertebral body

Fig. 2.

(Case 14) A 29-year-old female was found to have a neoplasm on the rib during physical examination. (A-F) Plain X-ray radiograph, MRI, and CT revealed a neoplasm involving T1-3 and the second rib. Biopsy confirmed the diagnosis of giant cell tumor. (G-H) After denosumab treatment, CT showed tumor shrinkage and formation of sclerotic margins. (I) The surgical plan of osteotomy. Through a posterior dorsomedial incision, en bloc resection of the tumor, including SEBR in T1-3, resection of the chest wall with tumor involvement, and wedge resection of the lung tissue with tumor involvement were carried out with reconstruction of the vertebral column and chest wall. (J) Completion of en bloc resection, with surgical specimen shown. X-ray radiographs confirmed en bloc resection of the tumor, partial vertebral body, rib, and wedge resected lung tissue. (K, L) Postoperative X-ray showed satisfactory position of the 3D-printed AVB and screw-rod internal fixation system. Abbreviations: CT, computed tomography; MRI, magnetic resonance imaging; SEBR, sagittal en bloc resection; AVB, artificial vertebral body

Postoperative treatment and follow-up

Postoperatively, the thoracic and wound drainage tubes were removed when the drainage fluid volume was less than 30 mL within 1 day. Patients were allowed to perform off-bed activities 1 week after the operation and were instructed to wear a thoracolumbar corset for 3 months. Safe and adequate rehabilitation activities were encouraged, and respiratory function exercises were recommended to promote pulmonary expansion and prevent respiratory complications.

Adjuvant treatment was administered according to the results of the surgery and pathological examination. Adjuvant chemotherapy using standard protocols was recommended in accordance with neoadjuvant chemotherapy. For patients with high-risk malignancies (tumor diameter > 5 cm, marginal resection margins, and recurrent tumors), postoperative radiation was administered to reduce local recurrence.

Patients were followed up every 3 months within the first 2 years after surgery, every 6 months until 5 years after surgery, and yearly thereafter. Routine physical examinations and pulmonary function tests were performed at each follow-up visit. Radiography and CT of the spine and chest were also performed. Complications and local control status were evaluated at each follow-up.

Results

Surgical outcomes

En bloc resection of all vertebrae, chest walls, and neighboring organs suspected of having tumor invasion was successfully performed in all patients (Table 2). To ensure a wide surgical margin, pulmonary wedge resection was performed in five patients by thoracic surgeons. The paravertebral muscle was partially resected in four patients, and the diaphragm was partially resected in two patients. Surgical margins were carefully evaluated by painting the specimen with ink and serially sectioning the lesion. R0 resections were performed in all cases. No patient underwent intralesional resection, and all patients underwent wide resection of the chest wall. Wide margins were achieved in 17 patients, and marginal margins were achieved in 9 patients because of intraspinal tumor invasion on the spinal side. The median duration of surgery was 360 min (range: 150–900 min), and the median blood loss was 1500 mL (range: 400–11200 mL). Seven patients were treated in the intensive care unit for a mean postoperative duration of 1.7 days (range: 1–3 days). An additional closed thoracic drainage tube was placed unilaterally in 16 patients and bilaterally in 8 patients. The drainage tubes were removed in all patients within 2 weeks postoperatively.

Table 2.

Surgical data and outcomes of the patients

Case	Resected vertebra	Resected ribs	Combined resection	Incision	Anterior reconstruction	Posterior fixation	Chest wall reconstruction	Surgery duration (min)	Blood loss (mL)	Thoracic drainage	Thoracic drainage duration (d)	Tumor margins
1	T11 TES	Left 11th	Pleura	DOR	3D AVB	T9-L1	Muscle flaps	310	1000	Bilateral	7, 7	Marginal
2	T10 TES, T9,11 SEBR	Left 9-11th	Pleura	T-shape	AVB	T6-T12	“Sandwich” +Muscle flaps	360	2700	Unilateral	7	Wide
3	T6-9 TES	Right 7th	Pleura	DOR	3D AVB	T3-12	Muscle flaps	490	3000	Unilateral	11	Wide
4	T8-9 TES	Right 8-9th	Pleura + paravertebral muscle	T-shape	Titanium mesh cage	T6-12	Muscle flaps	400	1900	Unilateral	5	Wide
5	T4-8 TES	Left 5-7th	Pleura	T-shape	3D AVB	T1-12	“Sandwich” +Muscle flaps	720	4000	Unilateral	13	Marginal
6	T8 SEBR	Left 7-8th	Pleura	T-shape	Titanium mesh cage	T6-10	“Sandwich” +Muscle flaps	300	1500	Unilateral	6	Marginal
7	T11-12 SEBR	Right 11-12th	Pleura + diaphragm	T-shape	3D AVB	T9-L2	Muscle flaps	360	1300	Unilateral	9	Marginal
8	T10-12 SEBR	Left 10-12th	Pleura + paravertebral muscle + diaphragm	T-shape	None	T8-12	“Sandwich” +Muscle flaps	340	600	Unilateral	12	Wide
9	T8-9 TES	Right 8-9th	Pleura	DOR	Titanium mesh cage	T6-11	Muscle flaps	350	1200	Bilateral	9,12	Marginal
10	T7-9 TES	Left 7-9th	Pleura + lung	T-shape	AVB	T4-12	“Sandwich” +Muscle flaps	400	5000	Bilateral	6, 9	Wide
11	T12 TES, T11, L1 SEBR	Left 11-12th	None	DOR	3D AVB	T9-L3	Muscle flaps	570	11,200	Unilateral	12	Wide
12	T9 TES	Left 9th	Pleura	DOR	AVB	T7-11	Muscle flaps	380	1500	None	0	Wide
13	T7,8 TES, T6,9 SEBR	Right 6-10th	Pleura + paravertebral muscle + lung	T-shape	3D AVB	T3-12	“Sandwich” +Muscle flaps	490	2600	Unilateral	6	Wide
14	T1-3 SEBR	Right 2nd	Pleura + lung	DOR	3D AVB	T1-4	Muscle flaps	480	2200	Unilateral	9	Wide
15	T1-3 TES	Left 1st-3rd	Pleura	DOR	AVB	C5-T6	Muscle flaps	540	2500	Bilateral	7,5	Marginal
16	T3,4 TES, T5 SEBR	Right 3-5th	Pleura	DOR	3D AVB	C7-T7	Muscle flaps	360	1000	Bilateral	8,9	Marginal
17	T3-5 SEBR	Right 3-5th	Pleura	T-shape	3D AVB	T2-6	Muscle flaps	285	1000	Unilateral	8	Wide
18	T8 SEBR	Left 8th	Pleura	DOR	None	T6-10	Muscle flaps	285	400	Unilateral	7	Wide
19	T4 TES	Right 4-5th	Pleura + paravertebral muscle	DOR	AVB	T2-6	Muscle flaps	360	2500	Bilateral	7,8	Marginal
20	T7-9 TES	Right 7-9th	Pleura	DOR	AVB	T5-11	Muscle flaps	450	800	Unilateral	7	Wide
21	T3 TES	Right 3-4th	Pleura	DOR	AVB	T1-5	Muscle flaps	360	1300	Bilateral	13,7	Marginal
22	T5-6 TES	Left 5-7th	Pleura + lung	T-shape	AVB	T2-9	“Sandwich” +Muscle flaps	900	3400	Bilateral	9,7	Wide
23	T8-11 SEBR	Right 8-11th	Pleura	T-shape	3D AVB	T7-L1	“Sandwich” +Muscle flaps	420	3300	Unilateral	10	Wide
24	T4 SEBR	Left 4th	Pleura + lung	T-shape	None	T3-5	Muscle flaps	230	800	Unilateral	7	Wide
25	T7-8 SEBR	Right 7-8th	Pleura	T-shape	None	T6-9	Muscle flaps	270	1000	Unilateral	8	Wide
26	T5 SEBR	Right 5th	None	DOR	None	T3-6	Muscle flaps	150	400	None	0	Wide

TES total en bloc spondylectomy, SEBR sagittal en bloc resection, DOR dorsomedial incision

Posterior spinal fusion with instrumentation was performed for spinal reconstruction in all patients. Anterior reconstruction was performed in 21 patients, with 3D-printed AVB in 10 patients, standardized AVB in 8 patients, and titanium mesh cages in 3 patients. All patients underwent chest wall reconstruction. The “sandwich” method was used in eight patients because of relatively large defects, and the remaining patients received muscle flap coverage.

The median follow-up duration was 14 months (range: 6 − 42 months). In three patients with preoperative neurological deficits (Frankel D), nerve function recovered to Frankel E within 4 weeks postoperatively. Additionally, the main complaints, including back pain, intercostal neuralgia, and shortness of breath, resolved. The pulmonary function results of all patients showed that the postoperative vital capacity (VC%) and forced expiratory volume in one second were similar to those before the surgery. Moreover, all patients returned to normal activities, and no patient suffered from respiratory insufficiency, paradoxical respiration, or oxygen therapy dependence at the final follow-up. Radiographic examination revealed good internal fixation and intervertebral fusion at the final follow-up. Local recurrence occurred in only one patient (case 5; 3.8%) 4 months after surgery, and the patient refused further treatment owing to financial constraints.

Complications

Five patients experienced postoperative wound complications (4/5 with T-shaped incisions), including poor wound healing in two cases and wound infection in three cases, which were successfully treated with debridement and antibiotics. Moreover, respiratory complications, including atelectasis and pneumonia, were observed in five patients after surgery. These patients recovered after treatment with respiratory function exercises combined with antibiotics for approximately 1 week (Table 3). No neurological complications, implant failure, hemothorax, or thrombosis were observed during the follow-up.

Table 3.

Complications of the patients

Case	Follow-up (months)	Complications
1	20	-
2	6	Poor wound healing
3	6	Poor wound healing
4	6	Wound infection
5	7	Wound infection, atelectasis
6	42	-
7	22	-
8	23	Atelectasis
9	14	-
10	10	Wound infection
11	12	-
12	13	-
13	13	-
14	13	-
15	13	-
16	14	-
17	16	-
18	24	-
19	23	-
20	24	-
21	25	Atelectasis
22	28	Pneumonia
23	13	-
24	11	-
25	18	-
26	21	-

Discussion

Tumors that simultaneously involve the thoracic vertebrae and posterior chest wall are uncommon, and no strict anatomic barrier prevents tumor invasion at the lesion sites. Therefore, treating these patients is clinically difficult and surgically complex. In our study, the surgical outcomes were generally satisfactory. This research provided valuable insights and novel ideas for comprehensive treatment and surgical techniques for these tumors.

These tumors should be treated based on their anatomical extent, oncological stage, and pathological diagnosis. Thus, after careful imaging evaluation, diagnostic biopsy is recommended for all patients. Needle biopsy was performed in all patients in our study, and the postoperative pathological diagnoses were consistent with the preoperative biopsy data. Appropriate neoadjuvant therapy for specific tumors, according to the correct pathological diagnosis, may contribute to tumor downstaging and facilitate tumor resection. According to the ESMO sarcoma clinical guidelines and literature reviews [14, 15], neoadjuvant chemotherapy using standard protocols was recommended for patients with osteosarcoma, Ewing’s sarcoma, and undifferentiated sarcoma. Preoperative denosumab (120 mg on days 0, 7, 15, and 28) was administered to patients with giant cell tumors to promote bone margin formation and improve resection margin evaluation. Although en bloc resection of these tumors remained challenging for musculoskeletal oncologists owing to the complicated anatomical structures of the spine and adjacent tissues, intralesional resection should not be considered the first-line surgical option. Additionally, portions of adjacent organs with tumor involvement should be resected en bloc when necessary. In our study, five patients underwent lung wedge resection owing to direct tumor invasion of the lung. Wide or marginal resection margins were achieved in all patients, and only one patient experienced postoperative recurrence, highlighting the importance of achieving safe resection margins for optimal local control. Postoperatively, adjuvant chemotherapy using standard protocols was recommended in accordance with neoadjuvant chemotherapy. Neoadjuvant radiotherapy has been associated with improved local control of the sarcoma [16]; however, for tumors invading the thoracic vertebrae and posterior chest wall, it may increase tissue adhesion, cerebrospinal fluid leakage, and poor wound healing. Therefore, no patient in the present study received neoadjuvant radiotherapy. Nevertheless, patients with high-risk malignancies (tumor diameter > 5 cm, marginal resection margins, and recurrent tumors) underwent postoperative radiation to improve local control.

The en bloc resection of tumors that invade the thoracic spinal column is technically challenging. Tomita et al. [3] reported that TES with total spondylectomy for spinal tumors achieved good clinical outcomes. Since then, TES has been widely adopted by musculoskeletal oncologists for the treatment of primary malignant, locally aggressive, benign, and solitary metastatic spinal tumors [17–20]. Yang et al. [21] performed TES combined with chest wall excision for spinal tumors involving the chest wall in 21 patients and obtained feasible and effective results. All patients in their study underwent TES of the invaded thoracic vertebrae; however, TES may result in a high incidence of complications due to the extensive dissection required. Boriani et al. [22] developed SEBR, in which en bloc resection with sagittal partial vertebrectomy was shown to be superior to TES for tumors growing eccentrically on one vertebral side. SEBR was intended to achieve tumor-free margins while reserving the healthy part of the vertebra, and it could limit dissection, reduce complications, and promote patient recovery, as supported by previous reports and our clinical experience [11, 23]. Therefore, in the present study, SEBR was implemented when the tumor was confined to the vertebral midline in more than half of the patients. Osteotomy strategies were adopted based on the classification system described in our previous studies [11, 12], and safe surgical margins of the spinal column were realized in all cases.

A combined anterior-posterior approach to the thoracic spine has been reported for the resection of tumors involving the spinal column and chest wall [19, 24]. This approach enables direct visualization of the tumor margins, which facilitates wide resection and avoids injuring major nerves, great vessels, and neighboring organs. However, compared to a single posterior approach, it is associated with longer surgical duration, increased blood loss, greater surgical trauma, higher costs, and a higher incidence of surgical complications. In the current study, a single posterior approach was preferred for most patients, while a combined anterior-posterior approach was used in only three patients (Cases 5, 15, and 22) because of the large tumor volume and the difficulty in separating the tumors from important organs, including the lungs, trachea, esophagus, or great vessels, through a single posterior approach. Furthermore, a T-shaped incision with erector muscle preservation was performed when the tumor margin of the chest wall exceeded the lateral border of the erector muscle; in this case, adequate visualization of the tumor margin was difficult to acquire via a single dorsomedial incision. Notably, four of the five wound complications occurred in patients with T-shaped incisions, potentially due to the extensive excision, soft tissue defects, and blood supply deficiency. Therefore, postoperative wound care is particularly important for patients undergoing T-shaped incisions.

Reconstruction of the spinal column and posterior chest wall with full coverage after surgical resection is another technical challenge. When a large proportion of the posterior column structures, including the zygapophyseal joint, is removed, the posterior stability must be restored using a screw-rod internal fixation system. All patients in this study underwent posterior spinal fusion using a screw-rod system for spinal reconstruction. Anterior reconstruction was performed with either an AVB or a titanium mesh cage when > 30% of the vertebra was resected, according to our classification system [11]. Chest wall integrity and stability are important for ensuring the protection of intrathoracic organs and the recovery of respiratory function. Previous studies suggest that defects > 5 cm in diameter or including > 4 ribs should be reconstructed because of the high risk of lung herniation and paradoxical chest movement, especially for anterolateral chest wall defects [25, 26]. In addition, posterior chest wall rigidity is important, and we recommend considering reconstruction owing to the risk of scapular entrapment. The goals of chest wall reconstructions are to restore chest wall rigidity, protect intrathoracic organs, preserve respiratory mechanics, and minimize deformity [26, 27]. Moreover, adequate soft tissue coverage is important to reduce the risk of infection. When the tumor resection involves extensive paraspinal or chest wall muscles, soft tissue defects can be reconstructed using various flaps, such as pedicled latissimus dorsi flaps, pedicled omental flaps, rectus abdominis myocutaneous flaps, or free musculocutaneous flaps. In our study, all patients received chest wall reconstruction with muscle flap coverage. The “sandwich” method was used to reconstruct the chest wall defects > 5 cm in diameter, and overall respiratory function preservation was satisfactory. For rib resection resulting in a rigid defect of the posterior chest wall, methyl methacrylate can be used to restore rigidity. Layers of Gynemesh were sutured to the rods of the pedicle screw fixation system on the spine side and to residual ribs and surrounding muscles on the chest wall side, providing a degree of mobility and increasing soft tissue attachment. Tension maintenance during Gynemesh suturing is important to prevent spinal cord compression and paradoxical respiration due to shifting of the methyl methacrylate plate. Moreover, careful layer closure, sufficient drainage, and local compression are recommended to eliminate dead space, reduce effusion, and promote soft tissue healing.

This study has certain limitations. First, because tumors involving both the thoracic vertebrae and posterior chest wall are rare, the sample size was small, and the study represents a single-institution experience. However, to our knowledge, this study represents the largest reported to date. Second, the retrospective design and relatively short follow-up limited the assessment of long-term clinical outcomes. Third, because our surgical techniques were not compared with other excision and reconstruction techniques, we cannot conclude that this surgical strategy is superior to other methods. Nevertheless, despite some limitations, the current follow-up results suggest our surgical strategies involving en bloc resection and reconstruction for tumors affecting both the thoracic vertebrae and the posterior chest wall are feasible and effective. Prospective and randomized controlled trials with larger cohorts are needed to reduce statistical bias and verify long-term outcomes.

Conclusions

For tumors involving both the thoracic vertebrae and posterior chest wall, sufficient preoperative evaluation and correct pathological diagnosis are important, and comprehensive treatments centered on en bloc resection and reconstruction surgeries are feasible and effective.

Abbreviations

CT

Computed tomography

MR

Magnetic resonance

ESMO

European Society for Medical Oncology

TES

Total en bloc spondylectomy

SEBR

Sagittal en bloc resection

3D

Three-dimensional

AVB

Artificial vertebral body

VC

Vital capacity

Authors’ contributions

AW analyzed the patient data and wrote the primary manuscript; XZ prepared the tables; and JH prepared the figures. QT, JW, and JL designed the study and revised the manuscript. All authors read and approved the final manuscript.

Funding

None to declare.

Data availability

The datasets used and/or analyzed in the current study are available from the corresponding author upon reasonable request.

Declarations

Ethics approval and consent to participate

The surgeon provided detailed written informed consent for operation to the patients preoperatively, and informed consent was obtained from the study participants. This study was approved by the Institutional Review Board of Sun Yat-sen University Cancer Center (B2025-977-01 ). The study was conducted in accordance with the principles of the Declaration of Helsinki.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.