Minimally invasive surgery for paravertebral or psoas abscess with spinal tuberculosis — a long-term retrospective study of 106 cases

The diagnosis of PA associated with tuberculous thoracolumbar or lumbar spondylitis was preoperatively based on the clinical symptoms, such as mild fever, night-sweats, back pain, and fatigue [8]; laboratory examinations such C-reactive protein (CRP), erythrocyte sedimentation rate (ESR), purified protein derivative, and the interferon gamma release assay [9];, and imaging examinations such as X-ray, computed tomography (CT), and magnetic resonance imaging (MRI) [6]. All patients were administered experimental anti-TB therapy for 2–4 weeks preoperatively [10]. The diagnoses of tuberculosis-induced PA was confirmed by pathological examination, acid-fast staining and/or culture of mycobacterium TB [2].

All operations were performed under local anesthesia. PCI chemotherapy infusion via an epidural tube was directed towards the intervertebral TB lesions and PCD via a double-lumen was performed for drainage of the abscess cavity based on preoperative CT and MRI findings and intraoperative CT guidance. Disposable AS-E/S II anesthesia puncture kits were provided by Shandong Weigao Group Medical Polymer Co., Ltd. (Weihai, P. R. China).

For thoracic vertebral lesions, the epidural tubes were placed above the transverse process and into the intervertebral space, whereas for lumbar lesions they were placed through the Kambin triangle into the intervertebral space. Considering the need for an adequate volume to cover vertebral infectious lesions, double PCI chemotherapy was administered from both sides via 2 epidural tubes to increase drug concentration. For psoas abscesses or PAs, placement of the double-lumen was performed vertically from the dorsal side into abscess cavity (Fig. 2).

The details of the surgical procedures are as follows. For PCI chemotherapy, once CT scanning confirmed placement of the epidural needle in the lesion’s center, the inner core was removed and the epidural tube placed. When the final position of the epidural tube was reconfirmed, the needle was removed and the tube was attached tightly to the skin with medical paste. When needed, the same procedure would be performed on the other side.

For PCD via a double-lumen tube, once puncture had been successfully accomplished, the inner core was replaced by a guidewire. Using the guidewire, a 5-mm OD expansion pipe with a tapered front was placed inside. A CT scan was performed to confirm proper placement of the 5-mm OD working tube. The double-lumen tube was them placed through the working tube in the PA cavity and CT confirmation was repeated, followed by careful removal of the working tube (Fig. 3).

After the catheterization process was completed, the infusion tube was sealed with heparin saline and the drainage tube was connected to a negative-pressure sterile drainage bag. The abscess material obtained from the pretreatment puncture was subjected to TB and standard bacterial culture along with drug susceptibility testing [11, 12].

During local infusion of chemotherapy and abscess drainage, 24-h continuous lavage and drainage was performed using 500 ml of saline + 1 g isoniazid via the PCD tube over a period of 3–6 weeks. Chemotherapy (0.2 g isoniazid) was infused via the PCI tube over a period of 1 month [13]. If necessary, local chemotherapy infusion was extended to 3 months [14]. Extubation was not performed until the fluid drained from the abscess cavity was clear and without obvious necrotic tissue [15, 16]. After removal of the infusion and drainage tubes and patient discharge, oral anti-TB drugs were continued for 1 year, while functional rehabilitation and lumbar back muscle exercises were simultaneously performed under the protection of a thoracolumbar brace [17].

ESR and CRP levels were dynamically monitored preoperatively and post-operatively at regular intervals to evaluate the infection status. The visual analog scale (VAS) was used to evaluate symptom relief and the Oswestry disability index (ODI) was used to assess rehabilitation condition and patient activity. Follow-up MRI was performed to assess the PA status. Treatment success was defined as short- and long-term absence of symptoms and improvement in laboratory and findings (Fig. 4 and Fig. 5). Total recovery was defined as absence of recurrent infection or abscess, bony fusion, and relief from back pain.

SPSS 24.0 software (SPSS Inc., Chicago, IL) was used for the statistical analysis. The data were expressed as means ± standard deviations, and differences with a P-value < 0.05 were considered significant. The sample size was calculated using PASS 19.0.3 statistical software (NCSS Inc., USA). Sample size of cross-sectional study was calculated as below. Zα/2 is the parameter of significance testing(α = 0.05, Z0.05/2 = 1.96). P is the expected effective rate. According to the literatures reviewed [1, 3, 18], the effective rate is about 50% on the basis of preliminary estimation. Permissible error(δ) is in the range of 0–10%. To keep the accuracy of study, the Permissible error will be set as 10%. The design effect (deff) will be 1.5. According to the result of the calculation, the sample size of adequacy is 74. So the study is powered, since the sample size of the study is 106.

The treatment regimens for thoracic and lumbar TB remained controversial. Individualized medical regimen need to be administrated according to the different situations of spinal tuberculosis. The suitable intervention method should be found out for the corresponding indications.

So the intervention regimens for thoracic and lumbar TB reported in the English literature include traditional open surgical treatment [25, 51] and minimally invasive approaches (percutaneous catheter drainage) [46, 48] on the basis of systematic anti-TB chemotherapy [21].

The indication for the MIS method including: (1) paravertebral abscess or posas abscess secondary to spinal tuberculosis involving the T8-L5 vertebrae, (2) without severe destruction of the vertebral body and obvious kyphosis or spinal instability, (3) without serious nerve or spinal cord compression symptoms. When the situations such as (2) or (3) occurs, open surgical operation is recommended instead of the MIS method [27, 28, 44].

The intervention method of the MIS method is percutaneous while the open surgery including anterior radical debridement with graft fusion or posterior debridement with fusion and fixation with huge invasiveness and more blood loss [22, 52].

The differences between our MIS method and the minimally invasive approach (percutaneous catheter drainage) were clarified below. Our MIS method involved PA elimination by percutaneous catheter drainage (PCD) and intervertebral percutaneous catheter infusion (PCI) chemotherapy for treating the TB lesions while percutaneous catheter drainage only focused on the drainage of abscess. Percutaneous catheter infusion (PCI) and irrigation with high concentrations of chemotherapy can eliminate remnant tubercle bacilli in the abscess cavity and reduce the rate of abscess recurrence [38, 41].

Considering the enrollment criteria, this study focused on MIS for treatment of the abscess and tuberculous vertebrae [28]. While CT-guided percutaneous drainage for ST had already been reported [15, 24, 34, 53‐55], its outcomes have not been clearly and definitively elucidated. In addition, no large-scale, long-term studies have been conducted to investigate the efficacy of a minimally invasive approach involving PCD and PCI chemotherapy. The diagnostic and therapeutic value of PCD with PCI chemotherapy, owing to complete abscess drainage and adequate drug delivery to the infected vertebrae, has been previously reported [56‐58]. Since infection of the T8–L5 vertebrae is common, PA and psoas abscess are not rare entities [15, 34, 55]. In fact, PAs, which include subdural and perivertebral abscesses, are one of the most common complications observed in the T8–L5 region [6]. In contrast, the anatomical location of the psoas abscess muscle means that psoas abscesses usually occur in the context of T12–L5 infection [4].

The advantage of PCD is the continuous and complete drainage with postoperative negative-pressure Hemovac suction [57]. Furthermore, sticky abscesses can be washed out by pressurized irrigation with saline and isoniazid. Irrigation with high concentrations of isoniazid can eliminate remnant tubercle bacilli in the abscess cavity and reduce the rate of abscess recurrence [59].

The PCI approach enables precise and daily delivery of isoniazid to the tuberculous foci [60]. A high local drug concentration has been confirmed as the key factor for treatment of infection foci and avoiding recurrent infection [59, 61, 62]. Furthermore, debridement of the infected disc and adjacent vertebral endplates can be completed through the working tube during catheterization [24]. This minimally invasive technique leads to less morbidity than major open surgery, and provides effective relief of the patients’ back pain by reducing the intradiscal pressure and preserving spinal stability [13, 14, 28]. Therefore, our patients were able to start ambulating with brace protection as early as possible after PCD and PCI chemotherapy.