Vertebral column resection for post tuberculosis severe kyphotic deformity: Results of 5 year follow-up

Abstract

Background

Spinal TB is endemic in our study region and many patients present with severe kyphotic deformities and neurological deficit. We corrected such deformities with all posterior single stage surgeries. This study was undertaken to evaluate the results, efficacy and safety of this technique.

Methods

Deformity correction of 16 patients was done during January 2012 to December 2014. All patients underwent posterior only approach for vertebral column resection at peri-apical region, posterior instrumentation with pedicular screws and anterior reconstruction using mesh cage. Postoperative X-ray films were evaluated. All patients were followed up at six weeks, 12 weeks, 18 weeks, six months and yearly thereafter. At follow-up patients were evaluated neurologically and radio-graphically.

Results

Mean age of the patients was 19.43 years. (Range 3–37) An average 1.62 vertebrae were excised and 5.93 vertebral levels were instrumented. Mean blood loss was 1013 ml and the mean duration of surgery was 6.78 h. The decrease in mean kyphotic deformity from preoperative 90.08⁰ to postoperative 38.06⁰ was statistically significant. (P < 0.000) Mean percentage correction was 57.59%. No pseudoarthrosis was found on X-rays. The decrease in Oswestry's Disability Index was from 55.43 to 10.06 was statistically significant. (P < 0.000) Two patients had neurological complications and one patient had wound complication.

Conclusion

The safety and efficacy of Posterior VCR technique for post tuberculosis severe kyphotic deformity is favorable with no severe late stage complications. Excision of ribs, careful handling of cord and gradual correction of deformity with good hemostasis is important.

1. Introduction

The incidence of tuberculosis (TB) is rising in both developed and developing world. More than two million people suffer from active spinal tuberculosis. Although it can be completely treated by chemotherapy, an average increase of 15 ⁰ deformity is seen in conservative treatment. Five percent patients end up having deformity of 60⁰.1, 2, 3 Spinal tuberculosis can be either complicated or uncomplicated. Complicated spinal TB presents with deformity, instability and neurological deficit.⁴ Progressive destruction of vertebral body leading to deformation of spine and involvement of anterior column leading to progressive destruction results in kyphotic deformity. A kyphotic deformity of 60⁰ or more is disabling and increases the chances of neurological deficit.⁵ Severe kyphosis in growing children can cause cosmetic and psychological disturbances and result in costopelvic impingement, cardiorespiratory problems and late-onset neurological deficit. Surgical correction of kyphotic deformity is required in cases having persistent pain, neurological deficit, cosmetic and functional disability.⁶ Kyphotic deformities are corrected by posterior only approach but can also be combined with anterior approach in cases like neurofibromatosis where anterior release at multiple disc levels helps in overall correction. Anterior only approach is difficult in deformity more than 60⁰. Moreover, the associated fibrosis and adhesions further complicate the procedure. Pure anterior opening osteotomy can lead to severe neurological deficit. Combined anterior and posterior procedures are major surgeries and have to be performed as single or double stage procedure. Pure posterior only procedures such as closing wedge osteotomy can lead to kinking of cord causing neurological deficit.^7,8 Spinal TB is endemic in our study region and many patients present with severe kyphotic deformities in our tertiary care center. Many such patients also present with neurological deficit. Such deformities were corrected with vertebral column resection at peri-apical region, posterior instrumentation and fusion with anterior reconstruction using titanium mesh cage. This study was conducted with an aim to evaluate the results of vertebral column resection for correction of severe post TB kyphosis and also evaluate the efficacy and safety of this technique.

2. Methodology-

This retrospective observational study was performed on 16 patients with post TB Kyphotic deformity operated during January 2012–2014. The data was procured from chief surgeon (1st Author) who has performed all the surgeries. These patients were followed up at serial intervals, which was routine protocol for our tertiary care center. Out of 16 patients, 12 were female and four were male. All patients were preoperatively evaluated clinically and radiologically (X-ray) (Fig. 1, Fig. 2). Magnitude of kyphotic deformity was calculated on X-rays. Also the level of involvement of vertebral column, number of vertebrae involved was noted. All patients had undergone posterior only approach for vertebral column resection, posterior instrumentation with pedicular screws and anterior reconstruction using mesh cage (Fig. 3). The surgical plan including the site of resection and degree of correction was decided on the basis of X-ray, CT scan and MRI scan. Surgical time, blood loss during the surgery and ASA grade of patient was noted. On the postoperative x-rays, Cobb's angle was calculated and thus percentage correction of deformity could be calculated. All patients underwent follow up after suture removal at six weeks, 12 weeks, 18 weeks and six months and yearly thereafter. Bony fusion was assessed at six months. If fusion was not noticed on X-ray at six months, patients were followed up every one and half month interval until fusion is seen. In patients detected with bony fusion, follow up was continued yearly thereafter. At follow up patients were evaluated by checking neurology which was graded according to Frankel grade and radiological examination by X-rays. Cobb's angle of operated deformity site was measured to check any loss of correction. In doubtful cases, CT scan was done to evaluate the status of fusion.

Fig. 1.

(A) Preoperative X-ray lateral view of whole spine (B) Preoperative X-ray Anteroposterior view of whole spine (C) Preoperative MRI of whole spine.

Fig. 2.

(A) Preoperative and Postoperative Clinical picture of patient showing deformity correction (B) Postoperative mobilization of same patient with total body contact brace.

Fig. 3.

Postoperative X-ray whole spine Anteroposterior and lateral view.

2.1. Operative technique

Preoperative optimization was done with incentive spirometer, good nutrition and hemoglobin build-up. All patients were operated by first author. They were placed in prone position on operating table under general anesthesia. A midline posterior incision was taken over the kyphotic deformity. Pedicle screws were inserted in two or three segments above and below the vertebral levels planned for resection. Appropriate position of screws was confirmed on fluoroscopy. Temporary rod was used to connect screws on one side (Fig. 4). Vertebral level to be resected was identified. Posterior elements were removed. Wide laminectomy was done extending to unaffected levels to avoid dural impingement following correction. This was important as correction of deformity leads to impingement of thecal sac by adjacent intact laminae. The pedicle of the vertebral body was identified and probed followed by decancellation of the apex. Adjacent ribs were excised on either side of the apex. In the dorsal region, thoracic nerve roots were sacrificed to prevent tenting and stretching of spinal cord. End plates were exposed and anterior and lateral parts of vertebral body were excised. The upper and lower endplates with intervening disc were removed. Posterior wall at the apex of deformity was resected in the end by pushing thinned out posterior wall away from the spinal cord, thereby, avoiding stretch and tenting of the spinal cord (Fig. 4). This step is of paramount importance as it protects the spinal cord till antero-lateral portion of apical deformity is completely excised. A temporary rod was applied for stability of spine and deformity was corrected gradually by segmental compression. Simultaneous anterior distraction by placing lamina spreader between two desired endplates and posterior compression over pedicle screws by using segmental instrumentation was carried out (Fig. 5). Correction was done gradually with mean arterial pressure maintained above 85 mm of Hg to prevent ischemia of spinal cord. This maneuver was followed by Stagnara's wake up test which was well explained pre-operatively to all the patients as a part of protocol. After confirming the correction of deformity on fluoroscopy, final titanium rods were put. Anterior reconstruction was done using titanium mesh cage filled with morcellized bone graft. Autogenous bone grafting was done for posterolateral fusion in all patients. Throughout the procedure emphasis was given to adequate hemostasis by using various hemostatic agents like 10 mg/kg Tranexamic acid infusion, thrombin soaked gelatin sponge and SURGIFLO™ Hemostatic Matrix Kit (Thrombin). Patients were mobilized postoperatively with TLSO bracing till there was documentation of fusion on X-ray.

Fig. 4.

Operative technique showing remnant of posterior thin cortex left behind at apical region and technique of temporary rod bent in a loop shape at the level of deformity for decompression to be carried out from both sides without changing position of the temporary rod.

Fig. 5.

Operative technique of gradual compression done posteriorly and distraction done anteriorly (Lamina spreader).

3. Results

The following results were obtained from data of 16 patients with post TB kyphotic deformity who were operated between January 2012 and December 2014 (Table 1). Average age of the patients were 19.43 years (range 3–37 years, s.d. = 11.30). An average of 1.62 vertebrae were excised in all cases (range 1–3 vertebrae, s.d. = 0.62). A mean of 5.93 vertebral levels were instrumented (range 4–8 levels, s.d. = 1.23). The average intraoperative blood loss was 1013 ml (range 160 ml–1800ml, s.d. = 583.97) and the mean duration of surgery was 6.78 h (range 3.5–10 h, s.d. = 1.89). The mean preoperative kyphosis was 90.08⁰ (Range 56.5⁰ - 130.8⁰, s.d. = 22.86). Mean kyphosis in the immediate postoperative X-ray was 38.06⁰ (range 17.9⁰ - 65.5⁰, s.d. = 14.30). Mean percentage correction was 57.59% (range 38.46%–76.16%, s.d. = 11.39). The difference in preoperative and postoperative deformity was found to be highly significant. (P < 0.000) Mean duration of hospital stay was 18 days (range 16–20, s.d. = 1.26). No pseudoarthrosis was seen on X-rays. There was no evidence of loosening or breakage of pedicle screws. The mean loss of correction at minimum five year follow-up was 2.56⁰ (range 1–6, s.d. = 1.09). There was a significant decrease in mean preoperative Oswestry's Disability Index from 55.43 (range 44–68, s.d. = 6.15) to 10.06 (range 8–16, s.d. = 2.64) (P < 0.000). There were complications in three patients, two of which had neurological and one had wound complication. There was no superficial or deep infection. One patient had neurological complication in the form of postoperative complete paraplegia. He was kept in ICU for postoperative monitoring after which he developed bedsores and succumbed to death over three months. Other patient developed incomplete paraplegia (Frankel grade B) postoperatively. He recovered after rehabilitation over period of 5 months from Frankel grade B to D. None of the perioperative events could explain the cause of his neurological deficit. Third patient suffered from wound complication in the form of wound dehiscence. We managed it with antibiotics, debridement and daily dressing. His wound healed in 4 weeks.

Table 1.

Personal characteristics and surgical details of 16 patients operated for post TB kyphotic deformity.

Sr.no	Gender	Age (years)	Resection Segment level	Instrumented vertebra	Pre-operative Cobbs angle (⁰)	Post-operative Cobbs angle (⁰)	Complications
1.	Female	18	D9,D10,D11	D6-D8, D12-L1	98.6	32.4	-
2.	Female	20	D7,D8	D4-D6, D9-D11	75.4	23.2	-
3.	Female	33	L3	D12-L2, L4-L5	108.8	34.4	-
4.	Female	16	D10,D11	D6-D9, D12-L3	58.4	23	-
5.	Male	34	D8, D9, D10	D5-D7, D11-D12	120.4	65.5	Neurological injury and death
6.	Female	20	D9,D10	D6-D8, D11-L1	74.4	35	-
7.	Male	8	D7	D5-D6, D8-D9	79.4	35.7	-
8.	Female	18	D11,D12	D8-D9, L1-L3	80.3	37.4	-
9.	Female	37	D11,D12	D6-D10, L1-L3	130.8	50.1	Neurological injury, recovered completely later
10.	Female	18	D12,L1	D9-D11, L2-L5	89.3	29.9	-
11.	Female	35	D8,D9	D5-D7,D10-D12	70.8	41.2	-
12.	Female	28	D8,D9	D5-D7, D10-D12	56.5	33.2	-
13.	Female	3	L2	D12-L1,L3-L4	75.1	17.9	-
14.	Female	5	D4,D5,D6	D1-D3, D7-D9	89.2	38.4	-
15.	Male	4	D5, D6, D7,D8, D9, D10	D3-D4,D11-D12	116.9	39.7	-
16.	Male	14	D10,D11,D12	D7-D9, L1-L3	117	72	Wound dehiscence
Average			2.31	5.68	90.08	38.06

4. Discussion

The most common cause of kyphotic deformity in developing world is tuberculosis of spine. It destroys the anterior column in 90% of patients which results in fusion of vertebra and kyphotic deformity.⁶ Patients with kyphotic deformity may be neurologically intact even without deformity correction.⁹ Also, neurological detoriation can occur many years after the cure of spinal TB. Occurrence of late onset neurological deficit has been reported at mean of 18 years and 26 years by Hsu et al. and Wong et all respectively.^10,11 A report by British medical council working party reported favorable outcome of spinal TB with nonsurgical treatment. But they took follow-up for only 15 years and severity of kyphosis was not included in the outcome measures.¹² Therefore, we also operated patients without neurological deficit and then data was used in this study.

Severe angular kyphotic deformities of thoracic or thoraco-lumbar can cause compromise in cardiorespiratory function and late neurological deficit. Lumbar deformity can cause postural imbalance, shortening of trunk, back pain and iliocostal impingement. Childhood kyphotic deformity may present with restrictive lung disease, respiratory failure and right heart failure in adulthood.¹³ The shape of spinal canal and the cord conditions differ individually even in patients with same degree of deformity.¹⁴ Also the relationship between the late onset neurological deficit and severity of deformity is not certain.¹⁵ Cord atrophy or syringomyelia develops by the time patient presents with neurological deficit.¹⁶ Hence, it is better to correct angular deformities rather than waiting for late neurological deficit to occur which doesn't have good prognosis.

Multiple techniques have been described to correct severe kyphosis due to spinal tuberculosis.¹⁷ Old technique like sequential, multistage procedure is quite morbid and has mortality rate of 10%.¹⁸ Single stage posterior procedures like Vertebral column resection (VCR) or posterior osteotomies are also described for correction of post TB kyphosis. SPO (Smith Peterson Osteotomy) requires a mobile disc anteriorly which is usually not present in severe rigid deformity. PSO (Pedicle subtraction osteotomy) can be applied to angular deformity but correction that can be achieved is 30–40° and the shortening of the cord can be dangerous so both the techniques are not recommended for Konstam's angle more than 90°.^19,20 VCR emerged as popular technique for correction of severe rigid kyphotic deformity with fixed trunk translation.²¹ Traditional VCR with both anterior and posterior approach is morbid and associated with major complications like vascular or neurological injury, increased blood loss, higher infection rates and prolonged surgical time. Suk et al. developed all posterior VCR so as to avoid complication of anterior approach. In this study 70 patients were operated with 61.9% correction of deformity and 8.57% complication rate.⁸ This was further modified by Kawahara et al. by adding interbody cage through posterior approach.¹⁸

The percentage deformity correction in our study was 57.59% as compared to Rajasekaran et al. (56.8%) and Wang et al. (82.3%). A closing-opening wedge posterior osteotomy to correct severe kyphotic deformities was first described by Kawahara et al. in which he reported good correction of 73.13% at 7.5 years without neurological injuries.¹⁸ The average blood loss in our procedure was 1013 ml, with range from 160 to 1800 ml as compared to other studies, Suk et al. (2980 ml), Bezer et al. (2780 ml) and Wang et al. (2933 ml).^8,22,23 The lower blood loss in our study can be attributed to good hemostasis with cauterization, use of hemostatic agents like Tranexamic acid infusion, Thrombin soaked Gelatin sponge and SURGIFLO™ and use of and hypotensive anesthesia except during the phase of deformity correction. Four out of 16 patients in our study were children. As children have lower amount of absolute blood loss, it could have affected average blood loss of our study.

Complications described in multiple studies on kyphotic deformity correction are neurological injury, infection, pulmonary embolism, thrombophlebitis, pseudoarthrosis, bed sores, wound dehiscence, hardware failure or prominence and superior mesenteric artery syndrome.^17,24, 25, 26 Except mortality in one patient, complications of our procedure are less than other reported studies with vertebral column resection. Suk et al. had 10 complications of 25 patients in his study.⁸ In other series of nine patients, four had pneumothorax and one neurological injury.²⁷ In our study of 16 patients, only two patients had neurological injury, one of which recovered upto Frankel grade D. In a study of 33 patients, Bridwell et al. reported 7 cases of pseudoarthrosis.¹⁹ There was no case of pseudoarthrosis in 15 patients we followed up. The neurological complication rate of our study was 12.5% which is lower than the two stage procedures (21.1%).²⁵ The reported complication rate of multiple studies after pedicle subtraction and closed wedge osteotomies is upto 15.2%, which is higher than our study.^14,25,28

5. Conclusion

The safety and efficacy of Posterior VCR technique for post tuberculosis severe kyphotic deformity is favorable with no severe late stage complications. Excision of ribs, careful handling of cord and gradual correction of deformity with good hemostasis is important. However, this technique is technically challenging and can be associated with severe life threatening complications. Nevertheless, Posterior VCR can result in high patient satisfaction and functional outcome with correct application and meticulous surgical technique.

Patient declaration statement

“The authors certify that they have obtained all appropriate patient consent forms. In the form the patient(s) has/have given his/her/their consent for his/her/their images and other clinical information to be reported in the journal. The patients understand that their names and initials will not be published and due efforts will be made to conceal their identity, but anonymity cannot be guaranteed.”

Declaration of competing interest

None declared.