DOI: https://doi.org/10.21203/rs.3.rs-2677652/v1
Objective: To describe the technique and evaluate the results of “skipping” posterior hemivertebra resection with short segmental fusion for the treatment of progressive complex congenital spinal deformities due to nonadjacent hemivertebrae.
Methods: This study was a retrospective case series. Twelve patients (M/6, F/6) with congenital scoliokyphosis due to nonadjacent fully segmented hemivertebrae and with an average age of 4.6 (3-9) years were enrolled. Whole standing spine radiographs were used to measure the Cobb angle of the segmental curve and the compensatory curve; segmental kyphosis; thoracic kyphosis; lumbar lordosis; trunk shift; sagittal vertical alignment; and T1-S1 length before surgery, after surgery and at the latest follow-up evaluation. The hemivertebral location, fused segment, operation time and blood loss were assessed.
Results: All patients were followed for at least 2 years. The average number of fused segments was 2.6 for each patient and 1.3 for each hemivertebra. The segmental scoliosis measurement was 43.0° before surgery, 4.7° after surgery and 7.8° at the latest follow-up evaluation. Segmental kyphosis measurements improved from 15.4° to 6.5°. The correction of the compensatory cranial and caudal curves was 86.7% and 83.5%, respectively. Trunk shift improved from 22.3 mm to 7.9 mm. The T1-S1 length was 25.3 cm before surgery, 27.5 cm after surgery, and 34.7 cm at the latest follow-up evaluation. Revision surgery was indicated for 2 patients who developed decompensation during the follow-up period.
Conclusions: “Skipping” posterior hemivertebra resection with short segmental fusion could provide satisfactory correction with limited fusion. However, decompensation may occur during follow-up. The prognosis of the discs between two fusion masses needs to be evaluated in the future.
Hemivertebrae are the most frequent cause of congenital scoliosis, which may lead to progressive local scoliokyphosis, and followed by compensatory deformities during spinal growth. As for the natural history, the scoliosis due to hemivertebrae increased by 1-3.5° per year[1]. Delayed intervention of congenital spine deformities due to the hemivertebrae may cause the severe and rigid spine deformities, which developed the complication of pulmonary compromise and neurologic dyfuction and usually required long spine fusions(Fig. 1).
Hemivertebra resection for patients at a younger age, when the local scoliokyphosis and compensatory deformities are mild, has proven to be an ideal procedure for progressive deformities due to a hemivertebra. Hemivertebra resection was performed by a combined anterior and posterior approach for good visualization of neural structures at first[2–6]. With the development of spine implant and surgical skills, especially the use of pedicle screws in younger children, posterior-only hemivertebra resection has become more popular with a correction rate almost equal to that of the combined anterior and posterior approach but with much lower surgery-related morbidity[7–13].
Posterior hemivertebra resection with short segmental fusion has been proven to be an effective and safe procedure for progressive congenital scoliokyphosis due to a single hemivertebra. The treatment of multiple nonajacent hemivertebrae has not been well studied. Traditionally, either long fusion or fusionless techniques were used, which may have the draback to the growth potential and mobility of the spine. In this study, we introduced a technique of “´skipping´ hemivertebra resection with short segmental fusion” for selected patients with this kind of deformity; we also tried to correct and control the deformities with limited fusion levels. This study was conducted to evaluate the preliminary results of this technique.
Twelve consecutive patients with progressive complex congenital scoliokyphosis due to two nonadjacent hemivertebrae treated by “skipping” hemivertebra resection with short segmental fusion were retrospectively studied.
The patients included 6 boys and 6 girls with a mean age of 4.6 years at surgery (Table 1).
Sex | Age | Follow-up (month) | Location of the resected hemivertebra | Op-time (min) | Blood loss (ml) | SRS-22 | |
---|---|---|---|---|---|---|---|
1 | F | 7 | 123 | L1/2, L4/5 | 260 | 610 | 3.7 |
2 | M | 3 | 96 | T7/8, T11/12 | 210 | 450 | 4.3 |
3 | M | 3 | 87 | T12, L3 | 230 | 400 | 4.5 |
4 | F | 9 | 84 | L2/3, L4/5 | 200 | 550 | 4.1 |
5 | F | 4 | 54 | T3, L2 | 240 | 500 | - |
6 | F | 2 | 50 | T12/L1, L4/5 | 180 | 350 | - |
7 | M | 3 | 38 | T10/11, L3 | 150 | 650 | - |
8 | F | 9 | 36 | T12/L1, L5 | 250 | 500 | 3.9 |
9 | M | 3 | 30 | T1, T12 | 185 | 380 | - |
10 | M | 3 | 28 | T8/9, L4/5 | 195 | 390 | - |
11 | F | 6 | 26 | L3/4, L5/S1 | 170 | 300 | - |
12 | M | 3 | 24 | T10/11, L2 | 190 | 400 | - |
The locations and number of resected hemivertebrae were as follows: cervicothoracic spine (C7-T1), 1; upper thoracic spine (T2-5), 1; thoracic spine (T6-9), 2; thoracolumbar region (T10-L2), 11; lumbar spine (L3-4), 8; and lumbosacral region (L5-S1), 1.
Measurements were taken from standing long-cassette anterior-posterior and lateral radiographs. Segmental scoliosis was measured from the upper endplate of the vertebra above the hemivertebra to the lower endplate of the vertebra below the hemivertebra. The cranial and caudal regions were also measured as described by Ruf in 2002[8]. Trunk shift was defined as the perpendicular distance (mm) from the sacrum center to the plumb line drawn from the midpoint of the C7 vertebra body. Segmental kyphosis was measured in the same way as segmental scoliosis on the coronal plane; differences between the measured values and the normal alignment were recorded[14]. T1–S1 length was used to calculate the growth of the spine. T1–S1 length was measured from the middle of the superior endplate of T1 to the middle of the superior endplate of S1. A Scoliosis Research Society-22 (SRS-22) questionnaire was used as an outcome measure at the latest follow-up evaluation. However, we did not obtain the results of the SRS-22 questionnaire from 4 patients because they were too young to be evaluated with this questionnaire.
All patients were treated by “skipping” posterior hemivertebra resection with short segmental fusion. All surgeries were performed by the same surgeon.
Each patient was placed prone on a frame after general anesthesia. A midline skin incision was made after baseline spinal cord monitoring recordings and administration of intravenous antibiotic prophylaxis. Two separate midline incisions were marked according to the fused levels.
We started the procedure with the deformity caused by the larger hemivertebra. Subperiosteal dissection was performed to expose the posterior structures of the hemivertebra and adjacent levels, including the lamina, transverse processes and facet joints. Fluoroscopy was used to confirm the hemivertebra and trajectory of the screw. Then, pedicle screws were inserted after tapping. Posterior elements of the hemivertebra, including the lamina, upper and lower facets, and the transverse process were removed to expose the pedicle and the nerve roots above and below the hemivertebra. The dura was cautiously elevated posteriorly, and the epidural veins were precauterized by bipolar cautery. After that, dissection was performed to expose the lateral cortex of the hemivertebra. A temporary precontoured rod was connected to the screws on the concave side to stabilize the spine. Osteotomies were used to remove the vertebral body along the cartilage endplate. The cancellous bone was preserved for grafting. At this point, the upper and lower discs should have been removed completely. The contralateral facet joint, bar or disc of the hemivertebra were also resected. A “V” shaped space was made after the osteotomy. After that, the temporary rod was removed, and a precontoured rod was connected to the screws on both sides. Gradual compression was applied to correct the deformity. Fluoroscopy was used to confirm the correction results. Decortication of the posterior elements and posterolateral fusion with autogenous bone were performed. The wound was closed with subfascial drainage.
After that, the other hemivertebra was resected, and short segmental fusion was performed in the same way.
Sensory evoked potential (SEP) and motor evoked potential (MEP) were used intraoperatively.
The patient usually started to walk 2 or 3 days after the surgery. Also, the patient used a plastic brace for at least 3 months.
A total of 24 hemivertebrae were resected in 12 patients. The average follow-up period was 56.3 (24–123) months. The average operation time was 205.0 (150–260) minutes, and the average intraoperative blood loss was 456.7 (350–650) ml (Table 1).
The correction in the coronal and sagittal planes is presented in Table 2. There was a correction rate of 89.1% for segmental scoliosis correction with a 3.1° loss of correction. In the sagittal plane, the correction rate was 57.1%, and 0.5° loss occurred at the latest follow-up evaluation. Trunk shift showed progressive improvement during the follow-up period. T1-S1 length was 25.3 cm before surgery, 27.5 cm after surgery, and 34.7 cm at the latest follow-up evaluation (Table 2).
Fused segments | Segmental scoliosis (°) | Cranial compensatory curve (°) | Caudal compensatory curve (°) | Trunk shift (mm) | Segmental kyphosis (°) | T1-S1 (cm) | ||||||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
Pre op | Post op | Follow -up | Pre op | Post op | Follow -up | Pre op | Post op | Follow -up | Preop | Postop | Follow-up | Preop | Postop | Follow-up | Preop | Postop | Follow-up | |||
1 | Upper | T12-L2 | 86 | 40 | 60 | 50 | 20 | 52 | 45 | 20 | 15 | 27 | 10 | 27 | 54 | 22 | 62 | 30 | 33.1 | 42.5 |
Lower | L4-5 | 50 | 3 | 20 | 45 | 10 | 30 | 5 | 3 | 20 | 15 | 12 | 10 | |||||||
2 | Upper | T7-8 | 40 | 8 | 10 | 25 | 5 | 7 | 13 | 3 | 3 | 25 | 11 | 5 | 15 | 2 | 2 | 21.2 | 22.9 | 41.5 |
Lower | T11-12 | 46 | 1 | 7 | 28 | 1 | 2 | 20 | 0 | 5 | 22.5 | -0.5 | -1.5 | |||||||
3 | Upper | T11-L1 | 45 | 2 | 2 | 15 | 1 | 1 | 33 | 1 | 1 | 10 | 15 | 3 | 1 | -3 | -4 | 22.3 | 24.5 | 32 |
Lower | L3-4 | 50 | 5 | 6 | 25 | 4 | 4 | 30 | 1 | 2 | 13 | 11 | 9 | |||||||
4 | Upper | L2-3 | 40 | 2 | 3 | 19 | 0 | 1 | 25 | 2 | 1 | 40 | 10 | 8 | 13 | 9 | 10 | 27.3 | 29 | 41.5 |
Lower | L4-5 | 48 | 3 | 5 | 38 | 2 | 4 | 15 | 1 | 3 | 14 | 11 | 8 | |||||||
5 | Upper | T2-5 | 36 | 5 | 7 | 10 | 2 | 4 | 22 | 3 | 3 | 15 | 8 | 10 | 2 | 0 | 0 | 23.5 | 25.8 | 33 |
Lower | L1-3 | 38 | 0 | 2 | 13 | 0 | 2 | 22 | 0 | 0 | 17 | 12 | 12 | |||||||
6 | Upper | T12-L1 | 45 | 2 | 3 | 27 | 0 | 0 | 19 | 2 | 3 | 20 | 8 | 4 | 6 | 1 | 0 | 27.2 | 28.8 | 38.7 |
Lower | L4-5 | 35 | 0 | 1 | 21 | 1 | 2 | 18 | 1 | 2 | 15 | 13 | 12 | |||||||
7 | Upper | T9-12 | 38 | 7 | 10 | 15 | 2 | 2 | 25 | 7 | 10 | 28 | 10 | 7 | 31.5 | 2.5 | 1.5 | 24 | 26 | 30 |
Lower | L3-5 | 55 | 2 | 3 | 38 | 1 | 2 | 20 | 2 | 3 | 15 | 10 | 8 | |||||||
8 | Upper | T11-L1 | 45 | 5 | 8 | 28 | 10 | 11 | 20 | 7 | 9 | 21 | 9 | 3 | 5 | 1 | 1 | 28.5 | 30.5 | 36.5 |
Lower | L4-S1 | 35 | 2 | 4 | 31 | 4 | 4 | 5 | 1 | 2 | 42 | 23 | 18 | |||||||
9 | Upper | C7-T2 | 37 | 3 | 5 | 18 | 1 | 3 | 10 | 2 | 4 | 20 | 8 | 5 | -1 | -3 | -4 | 24.7 | 27.5 | 30.6 |
Lower | T11-L1 | 40 | 2 | 3 | 15 | 2 | 4 | 25 | 4 | 5 | 26.5 | 6.5 | 4.5 | |||||||
10 | Upper | T7-10 | 42 | 3 | 6 | 25 | 3 | 8 | 28 | 2 | 10 | 17 | 7 | 10 | 1 | -1 | -1 | 24.1 | 26.7 | 29.5 |
Lower | L4-5 | 28 | 2 | 4 | 12 | 2 | 5 | 20 | 2 | 3 | 16 | 12 | 10 | |||||||
11 | Upper | L3-4 | 40 | 4 | 4 | 20 | 1 | 2 | 10 | 3 | 5 | 20 | 10 | 5 | 16 | 5 | 3 | 27.5 | 29.3 | 31.9 |
Lower | L4-5 | 35 | 3 | 3 | 28 | 2 | 3 | 5 | 1 | 2 | 16 | 6 | 5 | |||||||
12 | Upper | T10-11 | 30 | 4 | 5 | 20 | 2 | 3 | 10 | 2 | 2 | 25 | 12 | 8 | 1 | -3 | -3 | 23.5 | 26.3 | 29.2 |
Lower | L1-3 | 47 | 5 | 7 | 10 | 1 | 2 | 35 | 9 | 11 | 14 | 9 | 8 | |||||||
Average | 43.0 | 4.7 | 7.8 | 24 | 3.2 | 6.6 | 20.0 | 3.3 | 5.2 | 22.3 | 9.8 | 8.0 | 15.4 | 6.6 | 7.1 | 25.3 | 27.5 | 34.7 |
Most hemivertebrae have normal growth plates and create a wedge-shaped deformity that progresses during as the spine grows. The location of the hemivertebra is a significant factor in the progression of the deformity. The potential for progression of the curve was found to be higher in hemivertebrae located in the thoracolumbar transition zone or the lumbar area[1]. Braces and casts have been proven to be less effective for this kind of deformity, and early surgical intervention is indicated to prevent the development of severe local deformities and structural secondary curves.
Several surgical procedures have been introduced for the treatment of congenital scoliosis due to hemivertebrae. Among these procedures, hemivertebra resection is the only method that enables complete correction of the deformity by removing the pathology and that has predictable results. Hemivertebra resection was first performed via a combined anterior and posterior approach[2–6]. During the early years of the patient’s life, the local deformity due to hemivertebrae was mild, and compensatory structural curves had not yet developed. Pedicle screws could provide stronger force for correction than other implants, including those with hooks and wires. Ruf and Harms reported that it was safe to use pedicle screws even in very young children[15]. With the wide use of pedicle screws in children, hemivertebra resection with short fusion at a younger age has become an ideal procedure for congenital kyphoscoliosis due to hemivertebrae. In recent years, some surgeons have presented successful results of posterior-only hemivertebra resection and instrumentation with pedicle screws[7–13]. Ruf first reported successful results of posterior hemivertebra resection in 2003[9] and 2009[10]. However, only some of the patients were treated with short segmental fusion in both of their studies. After that, several studies reported good results after posterior hemivertebra resection with short segmental fusion[12, 13]. The advantages of hemivertebra resection by the posterior approach with transpedicular instrumentation were excellent correction of the local and compensatory curve in the frontal and sagittal planes, short fusion, early mobilization with high stability, no requirement for anterior access, and low neurologic risk.
Although posterior hemivertebra resection with short segmental fusion has been proven to be a safe and effective procedure for progressive congenital scoliokyphosis caused by a single hemivertebra, the early treatment of progressive complex congenital spinal deformities due to multiple nonadjacent hemivertebrae remains controversial. Early long fusion of the spine should be avoided, as it will lead to a short trunk and thoracic insufficiency syndrome. A fusionless technique may be an option. The use of a vertically expanded prosthetic titanium rib has been effective in patients with progressive congenital scoliosis to control spinal deformities and to expand the deformed chest wall[16–18]. The growing rod technique has been proven to be effective in the treatment of early-onset scoliosis[19]. However, reports on the results of the growing rod technique for congenital scoliosis are limited. In 2011, Elsebai et al reported the results of 19 patients diagnosed with congenital scoliosis treated with the growing rod technique. In that study, there were 12 patients with a single rod and 7 with dual rods; all patients had an average of 4 years of follow-up. The correction rate of the major curve was 27.8%. The mean T1-S1 length increase was 11.7 mm/y. The space available for lung (SAL) ratio increased from 0.81 preoperatively to 0.96 at the latest follow-up evaluation[20]. In 2012, Wang reported their results of the dual growing rod technique for 30 patients with congenital scoliosis. The mean scoliosis curvature improved from 72.3° to 34.9° after initial surgery and was 35.2° at the last follow-up or postfinal fusion. The increase in T1–S1 length was 1.49 cm per year (range, 0.75–2.50). The SAL improved from 0.84 preoperatively to 0.96 at the latest follow-up evaluation[21]. A hybrid technique of osteotomy with short segmental fusion and a dual growing rod for the treatment of severe and rigid congenital scoliokyphosis was introduced by Wang in 2013[22]. Their preliminary results showed that this technique could help to improve the correction of severe and rigid congenital spinal deformities and decrease the risk of implant failure by eliminating the large asymmetric growth potential around the apex, with little influence on the length of the spine.
Although fusionless techniques can correct and control complex early-onset scoliosis while preserving the growth potential of the spine, patients treated with these techniques still suffer many complications, multiple repeated surgeries and several anesthesia sessions. Several techniques, including the Shilla technique[23, 24], tethering technique[25, 26] and magnetic controlled growing rod technique[27, 28], have been introduced to address these problems in the treatment of patients with early-onset scoliosis. However, most of the reported results of these techniques were only for relatively flexible deformities such as idiopathic scoliosis and syndromic scoliosis. The outcomes of these techniques for the treatment of complex congenital spinal deformities remain to be investigated.
Two nonincarcerated nonadjacent hemivertebrae will cause two progressive segmental curves or a progressive long curve according to the location of the hemivertebrae. Early surgical intervention is mandatory to prevent the formation of severe deformities that require long fusions and that have complication risks. Traditionally, long fusion or fusionless techniques have been the only options. In this study, we introduced a technique of “skipping” posterior hemivertebra resection with short segmental fusion for the treatment of these children, aiming to correct and control progressive deformities without long fusion and complications due to a fusionless technique. The correction of segmental scoliosis and kyphosis was 89.1% and 57.1%, respectively. Significant growth of the spine was observed during the follow-up, and the T1–S1 length increased from 27.5 cm after the surgery to 34.7 cm at the latest follow-up evaluation.
Two patients in our study required revision surgeries for progressive decompensation during the follow-up: one with a dual growing rod and the other with posterior instrumented fusion. Junctional deformities after hemivertebra resection have been reported in several studies. Wang found that proximal kyphosis occurred in 7 out of their 37 patients. The risk factors included greater immediate postoperative segmental kyphosis, proximal junctional angle and screw malposition on the upper instrumented vertebra hemivertebra located on the lower thoracic or thoracolumbar region[29]. Yang et al reported emerging S-shaped curves in congenital scoliosis patients after hemivertebra resection and short segmental fusion. They found postoperative-emerging S-shaped scoliosis in 9 of their 128 patients. The reasons for this outcome remain unknown. The features of these curves were similar to those of adolescent idiopathic scoliosis, and brace or revision surgeries may have been needed[30]. In our series, progressive proximal junctional scoliokyphosis occurred in patient 1, and she underwent revision surgeries of posterior osteotomy and fusion 9 years after the initial surgery (Fig. 2). The reason for this occurrence may have been insufficient correction due to the improper upper instrumented vertebra at T13, which was near the apex of the scoliokyphosis. causing by the upper hemivertebra. Resection of one more hemivertebra and the dual growing rod technique was used for patient 5 five years after the surgery for progressive deformities caused by the growth of a hemivertebra left by the initial surgery.
As a result of stress concentration, the discs near a fusion mass tended to degenerate faster. Recently, Nohara et al found that distal disc degeneration occurred in 62.7% of their patients who received lumbar fusion for the treatment of adolescent idiopathic scoliosis[31]. In our study, 8 patients had two separate fusion masses in the thoracolumbar and lumbar regions, which were relatively mobile, after “skipping” posterior hemivertebra resection with short segmental fusion. The degeneration of the discs between two fusion masses remains unclear. No complaints or signs related to disc degeneration were found until the latest follow-up evaluation. MRI scans were arranged for 2 patients with two adjacent fusion masses in the lumbar spine at the 10-year follow-up. No significant degenerative changes in the discs between the two adjacent fusion masses were noted (Fig. 3). However, the patients in our series were still young even after 10 years of follow-up. They need to be further investigated through follow-up in the future.
“Skipping” posterior hemivertebra resection with short segmental fusion could provide satisfactory correction with limited fusion. This approach may help to avoid long fusion or fusionless techniques. Also, this technique may be an ideal choice for early intervention of congenital scoliokyphosis due to 2 nonadjacent fully segmented hemivertebrae before long, severe and rigid deformities occur. However, decompensation may occur as the children grow up. The prognosis of discs with two fusion masses in the lumbar spine needs to be evaluated in the future.
Ethics approval and consent to participate: This research was approved by the ethics committee of Peking Union Medical College Hospital, all procedures performed in studies involving human participants were in accordance with the 1964 Declaration of Helsinki and its later amendments or comparable ethical standards. All participants and their legal guardians agreed with the data and publication of the manuscript. All patients and their legal guardians provided written informed consent for the study.
Consent for publication: Not applicable.
Availability of data and materials: The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.
Competing interests: The authors declare that they have no competing interests.
Funding: This work was supported by the National Natural Science Foundation of China (81171673, 81972037, 81902178), Natural Science Foundation of Beijing Municipality( L222096, L192015), National High Level Hospital Clinical Research Funding (2022-PUMCH-B-122).
Authors' contributions: Y.D. and S. W. wrote the main manuscript text and joined the most of the surgeries. J. Z. performed surgeries and supervised the whole study. Y.Y., N.W. and Q.Z. collected the data and joined the most of the surgeries.G.L. joined the most of the surgeries and prepared the all figures.All authors reviewed the manuscript. All authors read and approved the final manuscript.
Acknowledgements: Not Applicable.
The Manuscript submitted does not contain information about medical device(s)/drug(s).
Note: No portions of this paper have been presented previously.