- Email: [email protected]
Hemivertebra Resection With Instrumented Fusion by Posterior Approach in Children
Spine Deformity 3 (2015) 541e548 www.spine-deformity.org
Hemivertebra Resection With Instrumented Fusion by Posterior Approach in Children Lucas Piantoni, MD*, Ida A. Francheri Wilson, MD, Carlos A. Tello, MD, PhD, Mariano A. Noel, MD, Eduardo Galaretto, MD, Rodrigo G. Remondino, MD, Ernesto S. Bersusky, MD Servicio de Patologıa Espinal, Hospital Nacional de Pediatrıa, Prof. Dr. Juan P. Garrahan C.A.B.A., Buenos Aires, Argentina Received 10 December 2014; revised 13 March 2015; accepted 26 April 2015
Abstract Study Design: We conducted a retrospective study of patients with congenital scoliosis due to hemivertebra (HV) and performed resection with instrumentation through posterior approach-only with long term follow-up. Objectives: The objective of this study was to assess results of HV resection by posterior approach-only with instrumentation between 2002 and 2011. Summary and Background Data: Hemiepiphysiodesis, arthrodesis in situ and resection without instrumentation had been performed in the past with different results. Hemivertebra resection with spinal instrumentation through anterior and posterior approach has been advocated as the treatment of choice. Methods: A total of 67 patients with 78 HV and 70 surgical procedures were evaluated. Thirty-five of the patients were females and 32 were males. The mean age of the patients was 5.5 years (from 0.8 to 16 yr), and the mean follow-up period was 6.55 years (from 2.1 to 10.8 yr). Eighteen patients presented additional pathologies: specific syndromes, cardiopathies, thoracic, abdominal, and bone malformations. Sixteen patients had partial HV while 51 had full mobility HV. Twenty-eight of the HV were thoracic, 16 thoracolumbar, 28 lumbar, and six lumbosacral. Results: The scoliosis mean preoperative angular value was 38.55 , and the mean postoperative angular value was 19.89 . The kyphosis mean preoperative angular value was 29.98 and the mean postoperative angular value was 15.41 . One rod was used in 38 surgical procedures (54.28%) and 2 rods in 32 procedures (45.72%). We used monoaxial screws in 45 patients, poliaxial screws in 17 patients, and both monoaxial screws and poliaxial screws in one patient. Screws and hooks were used in one patient and hooks in three patients. Arthrodesis was performed in 64 patients, and no complications were registered in those without arthrodesis. Postoperative orthesis was used in 57 opportunities. Several complications during or after HV resection by posterior approach-only were reported. Two patients suffered from intraoperative neurophysiological complications, which were resolved during surgery with minimum gestures and presented no sequelae. Two patients developed superficial infections and needed toilette and IV antibiotic therapy. Two patients needed an additional surgery due to secondary curves, 4.9 yr after primary surgery on average. One patient developed crankshaft. Conclusions: We concluded that HV resection by posterior approach-only with instrumentation is a simple, secure, reliable, less invasive and well tolerated technique that can successfully resolve this kind of congenital scoliosis in children. Ó 2015 Scoliosis Research Society. Keywords: Hemivertebra resection; Vertebrae malformation; Pediatric spine surgery; Congenital scoliosis; Early-onset scoliosis
Introduction Author disclosures: LP (none), IAFW (none), CAT (grants from Biomet, during the conduct of the study), MAN (none), EG (none), RGR (grants from Johnson & Johnson, during the conduct of the study), ESB (grants from Johnson & Johnson, during the conduct of the study). *Corresponding author. Diego Palma 81, 1 A, San Isidro, CP 1642, Buenos Aires, Argentina. Tel.: þ540117429926; fax: þ54111226000. E-mail address: [email protected] (L. Piantoni). 2212-134X/$ - see front matter Ó 2015 Scoliosis Research Society. http://dx.doi.org/10.1016/j.jspd.2015.04.008
Hemivertebra (HV) is a formation defect produced during embryonic development [1]. Scoliosis due to HV is considered a severe pathology of slow and progressive process, associated with coronal and sagittal imbalance. Without treatment, HV can develop into an unacceptable deformity [2,3].
542
L. Piantoni et al. / Spine Deformity 3 (2015) 541e548
The severity of the curve will depend on different variables: type and location of the defect, number of malformed vertebras, number of healthy incorporated vertebras, and the potential growth of the patient [4]. Nonsegmented and bisegmented incarcerated HV are not related to severe curves and/or could lead to a patron of balanced curve. It was demonstrated that patients with segmented, multiple, unilateral nonsequential or adjacent bilateral, or bilateral distant HV associated with contralateral bone bar have the worst prognosis, with higher chances to produce curves of severe angular value, trunk decompensation, and neurological compromise, especially when the kyphosis component is predominant. Surgical treatment of scoliosis associated with HV is often necessary to achieve a balanced spine [5-7]. The objective of surgery is to prevent potential and severe deformity or to resolve already established deformity. Generally, the first situation is more common in young patients, and the second, in adults. Posterior arthrodesis without instrumentation, anterioreposterior arthrodesis, and hemiepiphysiodesis can stabilize curves, but their outcomes are suboptimal in children. HV resection with instrumentation by the combined approach or posterior approach only with instrumented fusion accomplishes the aforementioned objectives and corrects the initial curve with better outcomes [8,9]. Previously, the removal of the HV was performed by a combined anterior and posterior approach [10-12]. But during the last decade, posterior approach has been advocated as the procedure of choice [7,13,14]. Indication for surgery should be based on the following: semisegmented and fully segmented HVs, thoracic region, lumbar or lumbosacral region, solitary defect, unbalanced trunk, and children older than 1 year 6 months [15,16]. It is very difficult to predict the natural history of each HV; one must consider the type or pattern of deformity, location, age of presentation, and progressive deformity. A complete history and physical and imaging evaluation (coronal and sagittal spinal balance) is needed that includes head and pelvic tilt, shoulder orientation, as well as complete neurological evaluation. Therefore, surgery indication is a compound or combination of two or more elements. The objective of this study was to assess the results of HV resection by posterior approach only with instrumentation by the Hospital Spine Division between 2002 and 2011. Materials and Methods Sixty-seven patients (35 females and 32 males) with 78 HVs were evaluated. Seventy surgical HV resection procedures with instrumentation through posterior approach only were performed between 2002 and 2011 by the spine department. Procedures were performed in that hospital or in other health care centers. This is a retrospective study, with Level IV evidence.
The following syndromes and associated pathologies were observed in 18 patients: bone malformations, VACTERL association, myelomeningocele, interventricular communication, renal hypoplasia, radius bone bilateral agenesis, tethered cord, and Goldenhar syndrome. We recorded the HV location, limits of the curve, arthrodesed levels, side and type, kyphosis, and scoliosis angular values before and after surgery, type of implant used, use of bone graft, use of postoperative orthosis, and years of follow-up. Complications included infection, hemodynamic decompensation, neurological findings, and others associated with the graft. All patients were periodically assessed. The assessment consisted of physical and imaging evaluation (X-Ray, magnetic resonance imaging, and complete computed tomographic scan, electrocardiograph, renal ultrasonography) in order to discard associated pathologies and located 3D computed tomographic scans in order to plan the surgery. Two different specialists performed the postoperative periodical follow-up, which consisted of a physical and XRay assessment (at 1, 3, 6, 12, 36, and 60 months post surgery) as well as the sporadic follow-up. HVs were classified as Mobile 0 (M0), Mobile 1 (M1), or Mobile 2 (M2) [4,17]. Mobile 0 HV is the HV that is fused or unsegmented. Mobile 1 HV refers to an HV that is fused or monosegmented. Mobile 2 HV is a bisegmented HV that has two types: incarcerated, which would not produce deformity, and unincarcerated, which could lead to deformity. The classic Winter and McEween classification based on the type of malformation distinguishes HVs as unsegmented (US), partially segmented or semisegmented (SS), and fully segmented (FS). In our study, we found 17 semisegmented HVs and 61 fully segmented HVs (Table). Surgical technique HV resection was performed by a posterior approach only in one step, with instrumentation of adjacent segments using pedicular screws and/or hooks. Unilateral or bilateral instrumentation was used depending on the age of the patient, severity of the curve, bilateral pathology, body mass index, and surgeon’s criteria [4]. The patient was placed in prone position and neurologically controlled with somatosensory evoked potentials and motor evoked potentials. A median posterior longitudinal incision was performed (with a uni- or bilateral approach). HV levels and screw placement were intraoperatively assessed through X-ray control. Screws were placed in the upper and lower levels of the HV. In some patients, the HV was resected from the inside, a screw was placed in the HV to be resected with increasing diameter until enough space, to start the removal procedure, was achieved. Then the curettage of the vertebra was performed with the insideeoutside technique. External, superior, and inferior walls of the pedicle were resected. Finally,
L. Piantoni et al. / Spine Deformity 3 (2015) 541e548
543
Table Patient demographics. Cases
Age at surgery
HV level HV side
Type MAV MAV Implant Rod Arthrodesis Postoperative Kyphosis Scoliosis number control
Postoperative Follow-up orthosis
MAV MAV kyphosis scoliosis Case Case Case Case Case Case Case Case Case Case Case Case
1 2 3 4 5 6 7 8 8 8 9 10
7 y 10 mo 4y 4 y 4 mo 5 y 11 mo 4 y 3 mo 1 y 5 mo 4 y 4 mo 10 y 10 y 10 y 1 mo 9 y 11 mo 8 mo
L1 T12 L4 T10 T7 T10 T9 L3 T4 L1 L3 T4, T6
Case Case Case Case Case Case Case
11 12 13 14 15 16 17
9 2 4 2 9 1 2
y y y y y y y
7 4 7 2 7 7 4
T10 L1 L4 L3 T10 L1 L1
Case Case Case Case
18 19 20 21
9 1 5 2
y y y y
3 mo 1 mo 5 mo 10 mo
mo mo mo mo mo mo mo
L1 L1 L1 T6, T10
Case 22 6 y 3 mo L3 Case 23 14 y 0 mo T10 Case 24 3 y 4 mo Case 25 4 y 1 mo
L2 T12, L5
Case Case Case Case Case Case Case Case Case Case Case Case Case Case
26 27 28 29 30 31 32 33 34 35 36 37 38 39
4 y 1 mo 14 y 5 mo 4y 3 y 7 mo 4 y 10 mo 8 y 8 mo 3 y 1 mo 14 y 7 mo 4 y 1 mo 3 y 2 mo 8 y 1 mo 6 y 1 mo 6 y 10 mo 8 y 11 mo
L3 L5 L2 L3 T11 L4 T10 L5 L2 L3 L3 L2 L5 L4, L5
Case Case Case Case Case Case Case
40 41 42 43 44 45 46
8 y 1 mo 2 y 2 mo 4 y 2 mo 16 y 11 y 4 mo 2 y 4 mo 9 y 9 mo
T11 L3 T5 L2 L4 L1 L3
Left Right Right Right Left Left Right Right Right Left Right Bilateral double ipsilateral Left Left Right Left Right Left Left Left Right Left Bilateral double ipsilateral Right Left Right Bilateral double contralateral Left Right Left Left Left Left Left Right Left Left Right Left Left Bilateral double ipsilateral Left Right Right Right Left Right Left
M1 M2 M2 M2 M2 M1 M2 M2 M1 M2 M2 M1
20 36 11 27 10 45 21 42 15 14 30 42
20 22 25 30 39 34 45 34 20 12 34 80
MS PS MS MS MS MS PS MS PM MS MS H, PS
2 2 2 1 2 1 1 1 1 1 1 2
T12eL2 T12eL1 L3eL5 T9eT11 T5eT7 T8eT10 T7eT11 L2eL4 T2eT10 T12eL3 L3eL4 T2eT10
11 20 0 10 8 32 11 15 14 12 8 50
0 23 12 25 35 30 38 20 12 5 20 48
No Cast Cast No No TLSO No TLSO TLSO Cast Cast Cast
7 y 4 mo 5 y 9 mo 10 y 3 mo 6 y 9 mo 6 y 3 mo 6 y 7 mo 6 y 8 mo 9 y 9 mo 9 y 9 mo 9 y 8 mo 6 y 1 mo 3 y 5 mo
M2 M2 M2 M2 M2 M2 M2
21 22 30 15 26 29 10
20 38 45 32 20 80 40
MS PS PS MS PS MS PS
1 1 2 1 1 1 1
T9eT11 L1eL3 L3eL5 L1eL4 T9eT11 T12eL3 L2eL3
18 4 8 4 10 10 4
22 5 13 12 22 37 27
2 8 5 5 6 6 7
M2 M2 M2 M2
10 11 26 41
45 40 45 55
MS, PS 2 H 1 PS 2 MS 2
L1eL2 L2eL4 T12eL3 T6eT11
8 5 12 18
4 30 15 28
TLSO Cast No Cast TLSO Cast Cast and TLSO No Cast TLSO TLSO
M1 M2
20 3
34 40
MS MS
1 1
L3eL4 T8eT11
0 2
30 19
2 y 1 mo 6 y 5 mo
M2 M2
8 40
20 51
MS MS
1 1
L2eL4 T9eS1
1 25
0 30
TLSO Cast and TLSO TLSO TLSO
6 y 5 mo 7 y 5 mo
M2 M1 M2 M2 M2 M1 M2 M1 M2 M2 M2 M2 M1 M1
18 11 22 42 32 18 15 45 44 26 7 14 40 49
25 45 30 40 46 35 23 45 30 32 40 36 45 40
MS MS MS MS MS MS MS MS MS H, PS MS MS MS PS
1 1 1 1 1 1 1 1 1 1 1 1 1 1
L3eL4 T12eS1 L2eL3 L2eL4 T10eT11 L3eL4 T9eT12 L1eS1 L2eL4 L3eL4 L1eL4 L1eL2 L2eL4 L2eS1
8 18 11 19 19 22 11 5 20 7 1 5 15 0
3 20 12 20 10 25 20 15 12 26 6 20 12 17
No TLSO TLSO TLSO TLSO TLSO TLSO TLSO TLSO No Cast No Cast TLSO
6 y 9 mo 6 y 7 mo 5 y 6 mo 6 y 5 mo 18 y 6 mo 10 y 7 mo 6 y 4 mo 4 y 4 mo 5 y 9 mo 5 y 8 mo 5 y 11 mo 6 y 7 mo 7 y 8 mo 5 y 7 mo
M2 M2 M1 M1 M1 M2 M2
38 29 21 23 39 42 36
31 40 35 40 22 45 42
MS MS MS PS MS MS MS
1 2 2 2 1 2 2
T11eT12 L2eL4 T2eT7 L1eL4 L5eS1 T12eL1 L2eL3
4 21 6 17 22 21 20
10 16 20 4 25 20 30
Cast TLSO No No TLSO TLSO YESO
10 y 1 mo 5 y 1 mo 4 y 10 mo 5 y 2 mo 10 y 4 mo 3 y 9 mo 8 y 10 mo
y y y y y y y
7 7 9 9 5 2 5
mo mo
mo mo mo
5y2 10 y 8 mo 5 y 5 mo 5 y 2 mo
(Continued on next page)
544
L. Piantoni et al. / Spine Deformity 3 (2015) 541e548
Table (Continued ) Cases
Age at surgery
HV level HV side
Type MAV MAV Implant Rod Arthrodesis Postoperative Kyphosis Scoliosis number control
Postoperative Follow-up orthosis
MAV MAV kyphosis scoliosis Case 47 6 y 2 mo Case 48 1 y 1 mo Case Case Case Case Case
49 50 51 52 53
5 7 2 2 7
y y y y y
5 7 3 4 6
mo mo mo mo mo
Case 53 2 y 9 mo Case 54 2 y 9 mo Case 55 7 y 6 mo Case 56 2 y 1 mo Case 57 4 y 2 mo
L5 L3
Right Right
M2 M2
50 10
34 46
PS MS
2 1
L5eS1 L3eL4
25 0
15 15
T11 L2 L4 L1 T9 left, T10 right
Left Left Right Left Bilateral double contralateral Right Left Right
M2 M1 M1 M2 M2
24 32 41 28 42
21 43 30 32 50
TM TM TM PS PS
2 2 2 1 2
T10eT12 L1eL3 L3eL5 T12eL2 T8eL4
8 9 19 2 20
2 21 22 24 32
M2 M2 M1
40 37 60
37 20 32
PS MS PS
2 2 1
T10eT12 T12eL2 T9eT11
30 10 11
15 4 10
M2 M2
43 11
32 35
MS MS
2 1
T10eT11 T9eL2
35 10
M2 M2
41 10
35 44
MS H, PS
2 1
T12eL2 T8eL3
M2 M2 M2 M2 M2 M2 M2
41 65 30 38 43 35 42
30 40 25 35 28 25 110
H PS PS MS PS MS H, PS
2 2 2 1 2 1 2
M2
10
32
H
2
T11 L4 T10
T8 Left T10 left, Bilateral L2 right double contralateral Case 58 5 y 6 mo L1 Right Case 59 4 y 6 mo T10, L3 Bilateral double ipsilateral Case 60 7 y 8 mo L2 Left Case 61 2 y 11 mo T11 Right Case 62 2 y 4 mo L1 Left Case 63 2 y 1 mo T12 Right Case 64 5 y 6 mo T8 Left Case 65 4 y 1 mo L1 Right Case 66 3 y 7 mo T10, L1 Bilateral double contralateral Case 67 7 y 8 mo L2 Left
Cast Cast and TLSO Cast No Cast TLSO Cast
5 y 3 mo 9 y 6 mo 5 y 2 mo 10 y 2 mo 8 y 3 mo 8 y 3 mo 5 y 11 mo
10 15
Cast TLSO Cast and TLSO Cast Cast
4 y 1 mo 5 y 1 mo 4 y 4 mo 4 y 10 mo 6 y 3 mo
48 4
19 12
TLSO TLSO
6 y 5 mo 4 y 8 mo
L1eL3 T9eL1 T12eL1 T12eL1 T7eT9 T12eL1 T9eL2
21 75 13 18 35 10 22
22 40 7 4 15 15 50
TLSO TLSO TLSO No No TLSO Cast and TLSO
4 5 4 5 5 6 6
L1eL3
16
59
Cast and TLSO
8 y 6 mo
y y y y y y y
4 mo 3 mo 5 mo 4 mo 4 mo 10 mo 11 mo
FS, fully segmented; H, hooks; HV, hemivertebra; M0, Mobile 0; M1, Mobile 1; M2, Mobile 2; MAV, mean angular value; mo, month(s); MS, monoaxial screws; PMS, poly- and monoaxial screws; PS, polyaxial screws; SH, screws and hooks; SS, semisegmented; TLSO, thoracolumbar orthosis; US, unsegmented; y, year(s).
resection of the hemi-laminae was performed, and the internal wall of the pedicle was removed. The technique was completed with the resection of the discs. The egg-shell procedure was the mother surgical technique during these 70 procedures. We tried to resect as much of the contralateral side of the disc as possible concerning the HV when a unilateral approach was used. This was probably one of the most demanding steps of the surgical technique. But crucial to prevent pseudarthrosis, the contralateral disc should be removed when all of the HV is resected. It is extremely difficult to close the wedge from behind without a complete body and disc removal. In other patients, however, the HV was resected from outside, and this was done according to surgeon’s criteria. In both cases, bleeding was carefully controlled. One or two rods were placed. When curves where not severe, instrumentation was limited to adjacent vertebras; however, when curves were severe, instrumentation included more
levels. Compression and/or distraction was performed when necessary. Lastly, the surgeon proceeded with final closure. Before the patient left the operating room, neurophysiology and neurology experts verified that nerve roots and spinal cord were unharmed (Fig.) [4,18]. One of the most important indications of a bilateral approach are kyphotic cases, kyphosis over scoliosis curve pattern, children with high BMI, older children, multiple HVs, HV with contralateral bone bar, and HV with other associated vertebral anomalies; in addition, we have the unilateral indications [19]. In clinical practice, when kyphosis is present, we recommend a bilateral approach with instrumentation. And we do agree that when kyphosis is present, two-level instrumentation might not be good enough to achieve a secure as well as solid level of fusion. We strongly recommend against performing unilateral instrumentation in kyphotic cases.
L. Piantoni et al. / Spine Deformity 3 (2015) 541e548
545
Fig. Image 1. Spine X-ray, anteroposterior (1A) and lateral (1B). One-year-10-month-old female patient, lumbar scoliosis of 45 , left L2eL3 hemivertebra Mobile 2 or fully segmented (Tello et al. [4-14,17]), preoperative. Image 2. Same patient, computed tomographic scan 3D image reconstruction, anterior (2A) and (2B) posterior view, same findings, preoperative. Image 3. Spine X-ray anteroposterior (3B) and lateral (3A), 7 months after surgery, left L2eL3 instrumented posterior arthrodesis, satisfactory outcome.
546
L. Piantoni et al. / Spine Deformity 3 (2015) 541e548
Results According to HV classification [4,17], 17 of the total 78 were HV type M1 (21.79%) and 61 were HV type M2 (78.21%). In relation to HV location, 28 were thoracic (35.9%), 16 thoracolumbar (20.51%), 28 lumbar (35.9%), and six lumbosacral (7.69%). Mean age of the patients at surgery was 5 years 5 months (from 8 months to 16 years). Eight patients had double HVs, which were resected in the same surgical procedure, and one patient had three HVs and required a two-step surgery. Mean postoperative follow-up was 6.55 years, with a maximum of 10 years 8 months and a minimum of 2 years 1 month. Implants used were monoaxial screws in 45 patients (67.17%), polyaxial screws in 17 patients (25.38%), hooks in 3 patients (4.47%), screws and hooks in 1 patient (1.49%), and poly- and monoaxial screws in 1 patient (1.49%). There was only 1 case of rib resection (1.49%) among the 70 procedures performed, which was required for HV excision. One rod was used in 38 surgical procedures (54.28%) and 2 rods in 32 procedures (45.72%). Intraoperative arthrodesis with allograft and autograft was performed in 64 patients (91.42%). Fifty-seven patients were immobilized with some kind of orthosis (85.07%). Scoliosis preoperative mean angular value (MAV) was 38.55 (range 5 12 e110 ), and scoliosis postoperative MAV was 19.89 (range 5 0 e59 ). Mean correction for scoliosis was 48.37%. Kyphosis preoperative MAV was 29.98 (range 5 3 e65 ), and kyphosis postoperative MAV was 15.41 (range 5 0 e75 ). Mean correction for kyphosis was 48.73%. Two patients suffered from a neurophysiological event (2.98%) that was reverted with minimal maneuvers (reduction of compression, distraction, reorientation of the screw, or rod remodeling) and presented no postoperative sequelae [20]. Medium-term postoperative complications were observed in 4 patients (5.63%): one patient presented ipsilateral/unilateral sciatica associated with the resected HV, whereas two patients developed superficial infections that were resolved with surgical toilet and intravenous antibiotic therapy. Eight patients (11.26%) presented incorporation of vertebras into the primary curve. Two of these patients required revision surgeries in the long term, one due to decompensation of the arthrodesed curve and proximal junction kyphosis and the second due to crankshaft effect.
Discussion HV resection is a preventive and corrective procedure performed in children before they develop a severe deformity, which is a practice accepted worldwide [21-23]. The complete resection of the HV is also performed in older patients in order to correct and stabilize the trunk. Several types of HV resections have been described in the past [24,25].
The first HV resection technique was described by Royle in 1928, and this was followed by Compere et al. In 1968, Winter and colleagues studied 234 patients in order to establish the diagnostic and prognostic criteria. However, not all patients were surgically treated. In 1975, with similar goals, Nasca evaluated a total of 60 patients. In 1979, Leatherman was the first to report a two-step surgery. Initially, the excision was performed by anterior and posterior approach [17], which allows for a better visualization of the defect and also the possibility to resect the adjacent discs with better outcomes and lower risks [26-28]. However, the anterior approach, either by thoracophreno-lumbotomy or thoracotomy, turns the procedure into a more aggressive one with a higher complication rate. The posterior approach allows the surgeon to reach the anterior spinal cord through the technique described by Harms et al., transforming the HV excision into a less aggressive treatment [7,13,29,30]. In addition, this approach allows for the complete removal of the neural arch with a more limited exposure of the HV body, which is less developed anteriorly and occupies only the posterior part. The posterior removal of the pedicle and vertebral body is done using either the outsideeinside or the egg-shell technique. Final removal of the laminae and inner wall of the pedicle is required to avoid or delay the epidural bleeding. We did not resect systematically the posterior part of the rib in patients with thoracic HV. Because did not perform comparative studies, we are not sure about its benefits. Our approach has been unilateral and, in some cases, bilateral including instrumentation. The demographic data of our study shows 38 cases with unilateral instrumentation (54.28%) and 32 with bilateral (45.72%). Bilateral or unilateral instrumentation was selected according to age of the patient, presence of opposite bone bar, angular value of the curve, and surgeon preference. Young children, without contralateral pathology, low body mass index, and when scoliosis angular value was more important than kyphosis angular value, were treated unilaterally [4,8,17]. Older patients, with a more severe deformity, a predominant kyphosis component, bone bars, or contralateral malformations, were treated bilaterally [4,8,17,31,32]. Current implants that allow the placement of pedicular screws and cervical implants can sometimes be used in very young children. The use of hooks has been previously described but located near the center of vertebral rotation may cause minor correction. However, our first choice is the placement of pedicular screws in the adjacent vertebras of the HV to be removed [30,33-36]. When we detected difficulties using the screws, we used sublaminar hooks. This was justified by careful observation of the surgical anatomy with images of 3D computed tomographic scan before surgery and intraoperative X-ray assessment. However, there were times when intraoperative anatomical findings did not match the computed
L. Piantoni et al. / Spine Deformity 3 (2015) 541e548
tomographic scans and preoperative X-ray. In order to save fusion levels, careful anatomical assessment is crucial to the instrumentation. When placing the pedicular screws, it is preferable to direct them to the median sagittal of the instrumentation, with convergent inclination. When the compression was done, we observed a lordotic effect. The corrections observed were satisfactory. Scoliosis preoperative MAV was 38.55 , and postoperative MAV was 19.89 , which represents a correction of 48.37%. Moreover, kyphosis preoperative MAV was 29.98 , and the postoperative was 15.41 , which represents a correction of 48.73%. It is important to note that in very young children where the objective of the surgery is preventive, the correction should be evaluated over time. Not all HVs have to be excised, only those that are completely mobile (M2) [4,17], nonincarcerated, and provoking decompensation (ie, located in lumbar and/or lumbosacral area) frequently progress. Therefore, we indicate the removal of the HV with preventive criteria. Patients could undergo surgery once they are 18 months old approximately, but there is no evidence to establish this age limit. However, the volume and bone size are parameters to take into account in order to remove the HV successfully before it produces greater deformity. The majority of our patients have been treated with instrumentation and arthrodesis, and some of them with a unilateral approach and without bone graft. We could not establish if those patients would have a better or worse outcome. We did not detect a relationship between nofusion and complications. However, a larger number of patients with longer follow-up period would be necessary to evaluate this. Importantly, the number of patients who maintained the correction or required subsequent surgeries must be evaluated over time and at the end of the spine growth. There were no significant complications, either intraoperative or immediate postoperative.
Conclusion Complete resection of the HV by a posterior approach only tends to be a simple, fast, and reliable procedure that allows for the complete removal of the formation defect with satisfactory results in the long term. This technique has been demonstrated to be less invasive and to achieve an accepted correction in the sagittal and coronal planes.
References [1] Hedequist D, Emans J. Congenital scoliosis. J Am Acad Orthop Surg 2004;12:266e75. [2] Ruf M, Jensen R, Harms J. Hemivertebra resection in the cervical spine. Spine (Phila Pa 1976) 2005;30:380e5.
547
[3] Hedequist DJ, Hall JE, Emans JB. Hemivertebra excision in children via simultaneous anterior and posterior exposures. J Pediatr Orthop 2005;25:60e3. [4] Tello C. Posterior hemivertebrae resection. Scoliosis Research Society 43rd Annual Meeting. San Antonio: Instructional Regional Course; 2010. [5] Benli IT, Aydin E, Alanay A, et al. Results of complete hemivertebra excision followed by circumferential fusion and anterior or posterior instrumentation in patients with type-IA formation defect. Eur Spine J 2006;15:1219e29. [6] Nakamura H, Matsuda H, Konishi S, Yamano Y. Single-stage excision of hemivertebrae via the posterior approach alone for congenital spine deformity: follow-up period longer than ten years. Spine 2002;27:110e5. [7] Ruf M, Harms J. Hemivertebra resection by a posterior approach: innovative operative technique and first results. Spine 2002;27: 1116e23. [8] Klemme WR, Polly DW, Orchowski JR. Hemivertebral excision for congenital scoliosis in very young children. J Pediatr Orthop 2001;21:761e4. [9] Shono Y, Abumi K, Kaneda K. One-stage posterior hemivertebra resection and correction using segmental posterior instrumentation. Spine 2001;26:752e7. [10] Bollini G, Docquier PL, Launay F, et al. Results at skeletal maturity after double-approach hemivertebral resection. Spine 2008;33: 2001e6. [11] Lin RM, Huang KY, Lai KA. Mini-open anterior spine surgery for anterior lumbar diseases. Eur Spine J 2008;17:691e7. [12] Garrido E, Tome-Bermejo F, Tucker SK, et al. Short anterior instrumented fusion and posterior convex non-instrumented fusion of hemivertebra for congenital scoliosis in very young children. Eur Spine J 2008;17:1507e14. [13] Ruf M, Harms J. Posterior hemivertebra resection with transpedicular instrumentation: early correction in children aged 1 to 6 years. Spine 2003;29:1593e4. [14] Winter RB, Burger EL. L5 hemivertebra resection and T12-S1 fusion in a 14-year-old female with a 36-year follow-up. Spine 2012;37: E445e50. [15] Bollini G, Docquier PL, Viehweger E, Launay F, Jouve JL. Lumbar hemivertebra resection. J Bone Joint Surg Am 2006;88:1043e52. [16] Li X, Luo Z, Li X, et al. Hemivertebra resection for the treatment of congenital lumbarspinal scoliosis with lateral-posterior approach. Spine (Phila Pa 1976) 2008;33:2001e6. [17] Tello C. Hemivertebra excision: Proceedings of the 18th Annual Meeting, Scoliosis Research Society, 1983:168. [18] Yaszay B, O’Brien M, Shufflebarger HL, et al. Efficacy of hemivertebra resection for congenital scoliosis: a multicenter retrospective comparison of three surgical techniques. Spine 2011;36:2052e60. [19] Peng X, Chen L, Zou X. Hemivertebra resection and scoliosis correction by a unilateral posterior approach using single rod and pedicle screw instrumentation in children under 5 years of age. J Pediatr Orthop B 2011;20:397e403. [20] Zhang J, Shengru W, Qiu G, et al. The efficacy and complications of posterior hemivertebra resection. Eur Spine J 2011;20:1692e702. [21] Winter RB, Lonstein JE. Scoliosis secondary to a hemivertebra: seven patients with gradual improvement without treatment. Spine 2010;35:E49e52. [22] Obeid I, Bourghli A, Vital JM. Lumbar hemivertebra resection by posterior approach for congenital scoliosis. Eur Spine J 2012;21:2721e3. [23] Obeid I, Bourghli A, Vital JM. Thoracic hemivertebra resection by posterior approach for congenital scoliosis. Eur Spine J 2013;22: 678e80. [24] Jeszenszky D, Fekete TF, Kleinstueck FS, et al. Fusionless posterior hemivertebra resection in a 2-year-old child with 16 years follow-up. Eur Spine J 2012;21:1471e6. [25] Wang S, Zhang J, Qiu G, et al. Posterior hemivertebra resection with bisegmental fusion for congenital scoliosis: more than 3 year
548
[26]
[27]
[28]
[29]
L. Piantoni et al. / Spine Deformity 3 (2015) 541e548 outcomes and analysis of unanticipated surgeries. Eur Spine J 2013;22:387e93. Jalanko T, Rintala R, Puisto V, Helenius I. Hemivertebra resection for congenital scoliosis in young children: comparison of clinical, radiographic, and health-related quality of life outcomes between the anteroposterior and posterolateral approaches. Spine 2011;36: 41e9. Xu W, Yang S, Wu X, Claus C. Hemivertebra excision with shortsegment spinal fusion through combined anterior and posterior approaches for congenital spinal deformities in children. J Pediatr Orthop B 2010;19:545e50. Mladenov K, Kunkel P, Stuecker R. Hemivertebra resection in children, results after single posterior approach and after combined anterior and posterior approach: a comparative study. Eur Spine J 2012;21:506e13. Deviren V, Berven S, Smith JA, et al. Excision of hemivertebrae in the management of congenital scoliosis involving the thoracic and thoracolumbar spine. J Bone Joint Surg Br 2001;83: 496e500.
[30] Ruf M, Jensen R, Letko L, Harms J. Hemivertebra resection and osteotomies in congenital spine deformity. Spine 2009;34:1791e9. [31] Aydogan M, Ozturk C, Tezer M, et al. Posterior vertebrectomy in kyphosis, scoliosis and kyphoscoliosis due to hemivertebra. J Pediatr Orthop B 2008;17:33e7. [32] Li XF, Liu ZD, Hu GY, et al. Posterior unilateral pedicle subtraction osteotomy of hemivertebra for correction of the adolescent congenital spinal deformity. Spine J 2011;11:111e8. [33] Freeman BJ, Oullet JA, Webb JK. Excision of hemivertebrae in the management of congenital scoliosis. J Bone Joint Surg Br 2002;83: 496e500. [34] Hedequist D, Emans J, Proctor M. Three rod technique facilitates hemivertebra wedge excision in young children through a posterior only approach. Spine 2009;34:E225e9. [35] Zarghooni K, Sobotrke R, Schmidt H, et al. Dorsal resection of a thoracic hemivertebra in a 4-year-old boy with endochondral gigantism. A case report. Acta Orthop Belg 2010;76:699e705. [36] Halm H. Transpedicular hemivertebra resection and instrumented fusion for congenital scoliosis. Eur Spine J 2011;20:993e4.
Hemivertebra Resection With Instrumented Fusion by Posterior Approach in Children Lucas Piantoni, MD*, Ida A. Francheri Wilson, MD, Carlos A. Tello, MD, PhD, Mariano A. Noel, MD, Eduardo Galaretto, MD, Rodrigo G. Remondino, MD, Ernesto S. Bersusky, MD Servicio de Patologıa Espinal, Hospital Nacional de Pediatrıa, Prof. Dr. Juan P. Garrahan C.A.B.A., Buenos Aires, Argentina Received 10 December 2014; revised 13 March 2015; accepted 26 April 2015
Abstract Study Design: We conducted a retrospective study of patients with congenital scoliosis due to hemivertebra (HV) and performed resection with instrumentation through posterior approach-only with long term follow-up. Objectives: The objective of this study was to assess results of HV resection by posterior approach-only with instrumentation between 2002 and 2011. Summary and Background Data: Hemiepiphysiodesis, arthrodesis in situ and resection without instrumentation had been performed in the past with different results. Hemivertebra resection with spinal instrumentation through anterior and posterior approach has been advocated as the treatment of choice. Methods: A total of 67 patients with 78 HV and 70 surgical procedures were evaluated. Thirty-five of the patients were females and 32 were males. The mean age of the patients was 5.5 years (from 0.8 to 16 yr), and the mean follow-up period was 6.55 years (from 2.1 to 10.8 yr). Eighteen patients presented additional pathologies: specific syndromes, cardiopathies, thoracic, abdominal, and bone malformations. Sixteen patients had partial HV while 51 had full mobility HV. Twenty-eight of the HV were thoracic, 16 thoracolumbar, 28 lumbar, and six lumbosacral. Results: The scoliosis mean preoperative angular value was 38.55 , and the mean postoperative angular value was 19.89 . The kyphosis mean preoperative angular value was 29.98 and the mean postoperative angular value was 15.41 . One rod was used in 38 surgical procedures (54.28%) and 2 rods in 32 procedures (45.72%). We used monoaxial screws in 45 patients, poliaxial screws in 17 patients, and both monoaxial screws and poliaxial screws in one patient. Screws and hooks were used in one patient and hooks in three patients. Arthrodesis was performed in 64 patients, and no complications were registered in those without arthrodesis. Postoperative orthesis was used in 57 opportunities. Several complications during or after HV resection by posterior approach-only were reported. Two patients suffered from intraoperative neurophysiological complications, which were resolved during surgery with minimum gestures and presented no sequelae. Two patients developed superficial infections and needed toilette and IV antibiotic therapy. Two patients needed an additional surgery due to secondary curves, 4.9 yr after primary surgery on average. One patient developed crankshaft. Conclusions: We concluded that HV resection by posterior approach-only with instrumentation is a simple, secure, reliable, less invasive and well tolerated technique that can successfully resolve this kind of congenital scoliosis in children. Ó 2015 Scoliosis Research Society. Keywords: Hemivertebra resection; Vertebrae malformation; Pediatric spine surgery; Congenital scoliosis; Early-onset scoliosis
Introduction Author disclosures: LP (none), IAFW (none), CAT (grants from Biomet, during the conduct of the study), MAN (none), EG (none), RGR (grants from Johnson & Johnson, during the conduct of the study), ESB (grants from Johnson & Johnson, during the conduct of the study). *Corresponding author. Diego Palma 81, 1 A, San Isidro, CP 1642, Buenos Aires, Argentina. Tel.: þ540117429926; fax: þ54111226000. E-mail address: [email protected] (L. Piantoni). 2212-134X/$ - see front matter Ó 2015 Scoliosis Research Society. http://dx.doi.org/10.1016/j.jspd.2015.04.008
Hemivertebra (HV) is a formation defect produced during embryonic development [1]. Scoliosis due to HV is considered a severe pathology of slow and progressive process, associated with coronal and sagittal imbalance. Without treatment, HV can develop into an unacceptable deformity [2,3].
542
L. Piantoni et al. / Spine Deformity 3 (2015) 541e548
The severity of the curve will depend on different variables: type and location of the defect, number of malformed vertebras, number of healthy incorporated vertebras, and the potential growth of the patient [4]. Nonsegmented and bisegmented incarcerated HV are not related to severe curves and/or could lead to a patron of balanced curve. It was demonstrated that patients with segmented, multiple, unilateral nonsequential or adjacent bilateral, or bilateral distant HV associated with contralateral bone bar have the worst prognosis, with higher chances to produce curves of severe angular value, trunk decompensation, and neurological compromise, especially when the kyphosis component is predominant. Surgical treatment of scoliosis associated with HV is often necessary to achieve a balanced spine [5-7]. The objective of surgery is to prevent potential and severe deformity or to resolve already established deformity. Generally, the first situation is more common in young patients, and the second, in adults. Posterior arthrodesis without instrumentation, anterioreposterior arthrodesis, and hemiepiphysiodesis can stabilize curves, but their outcomes are suboptimal in children. HV resection with instrumentation by the combined approach or posterior approach only with instrumented fusion accomplishes the aforementioned objectives and corrects the initial curve with better outcomes [8,9]. Previously, the removal of the HV was performed by a combined anterior and posterior approach [10-12]. But during the last decade, posterior approach has been advocated as the procedure of choice [7,13,14]. Indication for surgery should be based on the following: semisegmented and fully segmented HVs, thoracic region, lumbar or lumbosacral region, solitary defect, unbalanced trunk, and children older than 1 year 6 months [15,16]. It is very difficult to predict the natural history of each HV; one must consider the type or pattern of deformity, location, age of presentation, and progressive deformity. A complete history and physical and imaging evaluation (coronal and sagittal spinal balance) is needed that includes head and pelvic tilt, shoulder orientation, as well as complete neurological evaluation. Therefore, surgery indication is a compound or combination of two or more elements. The objective of this study was to assess the results of HV resection by posterior approach only with instrumentation by the Hospital Spine Division between 2002 and 2011. Materials and Methods Sixty-seven patients (35 females and 32 males) with 78 HVs were evaluated. Seventy surgical HV resection procedures with instrumentation through posterior approach only were performed between 2002 and 2011 by the spine department. Procedures were performed in that hospital or in other health care centers. This is a retrospective study, with Level IV evidence.
The following syndromes and associated pathologies were observed in 18 patients: bone malformations, VACTERL association, myelomeningocele, interventricular communication, renal hypoplasia, radius bone bilateral agenesis, tethered cord, and Goldenhar syndrome. We recorded the HV location, limits of the curve, arthrodesed levels, side and type, kyphosis, and scoliosis angular values before and after surgery, type of implant used, use of bone graft, use of postoperative orthosis, and years of follow-up. Complications included infection, hemodynamic decompensation, neurological findings, and others associated with the graft. All patients were periodically assessed. The assessment consisted of physical and imaging evaluation (X-Ray, magnetic resonance imaging, and complete computed tomographic scan, electrocardiograph, renal ultrasonography) in order to discard associated pathologies and located 3D computed tomographic scans in order to plan the surgery. Two different specialists performed the postoperative periodical follow-up, which consisted of a physical and XRay assessment (at 1, 3, 6, 12, 36, and 60 months post surgery) as well as the sporadic follow-up. HVs were classified as Mobile 0 (M0), Mobile 1 (M1), or Mobile 2 (M2) [4,17]. Mobile 0 HV is the HV that is fused or unsegmented. Mobile 1 HV refers to an HV that is fused or monosegmented. Mobile 2 HV is a bisegmented HV that has two types: incarcerated, which would not produce deformity, and unincarcerated, which could lead to deformity. The classic Winter and McEween classification based on the type of malformation distinguishes HVs as unsegmented (US), partially segmented or semisegmented (SS), and fully segmented (FS). In our study, we found 17 semisegmented HVs and 61 fully segmented HVs (Table). Surgical technique HV resection was performed by a posterior approach only in one step, with instrumentation of adjacent segments using pedicular screws and/or hooks. Unilateral or bilateral instrumentation was used depending on the age of the patient, severity of the curve, bilateral pathology, body mass index, and surgeon’s criteria [4]. The patient was placed in prone position and neurologically controlled with somatosensory evoked potentials and motor evoked potentials. A median posterior longitudinal incision was performed (with a uni- or bilateral approach). HV levels and screw placement were intraoperatively assessed through X-ray control. Screws were placed in the upper and lower levels of the HV. In some patients, the HV was resected from the inside, a screw was placed in the HV to be resected with increasing diameter until enough space, to start the removal procedure, was achieved. Then the curettage of the vertebra was performed with the insideeoutside technique. External, superior, and inferior walls of the pedicle were resected. Finally,
L. Piantoni et al. / Spine Deformity 3 (2015) 541e548
543
Table Patient demographics. Cases
Age at surgery
HV level HV side
Type MAV MAV Implant Rod Arthrodesis Postoperative Kyphosis Scoliosis number control
Postoperative Follow-up orthosis
MAV MAV kyphosis scoliosis Case Case Case Case Case Case Case Case Case Case Case Case
1 2 3 4 5 6 7 8 8 8 9 10
7 y 10 mo 4y 4 y 4 mo 5 y 11 mo 4 y 3 mo 1 y 5 mo 4 y 4 mo 10 y 10 y 10 y 1 mo 9 y 11 mo 8 mo
L1 T12 L4 T10 T7 T10 T9 L3 T4 L1 L3 T4, T6
Case Case Case Case Case Case Case
11 12 13 14 15 16 17
9 2 4 2 9 1 2
y y y y y y y
7 4 7 2 7 7 4
T10 L1 L4 L3 T10 L1 L1
Case Case Case Case
18 19 20 21
9 1 5 2
y y y y
3 mo 1 mo 5 mo 10 mo
mo mo mo mo mo mo mo
L1 L1 L1 T6, T10
Case 22 6 y 3 mo L3 Case 23 14 y 0 mo T10 Case 24 3 y 4 mo Case 25 4 y 1 mo
L2 T12, L5
Case Case Case Case Case Case Case Case Case Case Case Case Case Case
26 27 28 29 30 31 32 33 34 35 36 37 38 39
4 y 1 mo 14 y 5 mo 4y 3 y 7 mo 4 y 10 mo 8 y 8 mo 3 y 1 mo 14 y 7 mo 4 y 1 mo 3 y 2 mo 8 y 1 mo 6 y 1 mo 6 y 10 mo 8 y 11 mo
L3 L5 L2 L3 T11 L4 T10 L5 L2 L3 L3 L2 L5 L4, L5
Case Case Case Case Case Case Case
40 41 42 43 44 45 46
8 y 1 mo 2 y 2 mo 4 y 2 mo 16 y 11 y 4 mo 2 y 4 mo 9 y 9 mo
T11 L3 T5 L2 L4 L1 L3
Left Right Right Right Left Left Right Right Right Left Right Bilateral double ipsilateral Left Left Right Left Right Left Left Left Right Left Bilateral double ipsilateral Right Left Right Bilateral double contralateral Left Right Left Left Left Left Left Right Left Left Right Left Left Bilateral double ipsilateral Left Right Right Right Left Right Left
M1 M2 M2 M2 M2 M1 M2 M2 M1 M2 M2 M1
20 36 11 27 10 45 21 42 15 14 30 42
20 22 25 30 39 34 45 34 20 12 34 80
MS PS MS MS MS MS PS MS PM MS MS H, PS
2 2 2 1 2 1 1 1 1 1 1 2
T12eL2 T12eL1 L3eL5 T9eT11 T5eT7 T8eT10 T7eT11 L2eL4 T2eT10 T12eL3 L3eL4 T2eT10
11 20 0 10 8 32 11 15 14 12 8 50
0 23 12 25 35 30 38 20 12 5 20 48
No Cast Cast No No TLSO No TLSO TLSO Cast Cast Cast
7 y 4 mo 5 y 9 mo 10 y 3 mo 6 y 9 mo 6 y 3 mo 6 y 7 mo 6 y 8 mo 9 y 9 mo 9 y 9 mo 9 y 8 mo 6 y 1 mo 3 y 5 mo
M2 M2 M2 M2 M2 M2 M2
21 22 30 15 26 29 10
20 38 45 32 20 80 40
MS PS PS MS PS MS PS
1 1 2 1 1 1 1
T9eT11 L1eL3 L3eL5 L1eL4 T9eT11 T12eL3 L2eL3
18 4 8 4 10 10 4
22 5 13 12 22 37 27
2 8 5 5 6 6 7
M2 M2 M2 M2
10 11 26 41
45 40 45 55
MS, PS 2 H 1 PS 2 MS 2
L1eL2 L2eL4 T12eL3 T6eT11
8 5 12 18
4 30 15 28
TLSO Cast No Cast TLSO Cast Cast and TLSO No Cast TLSO TLSO
M1 M2
20 3
34 40
MS MS
1 1
L3eL4 T8eT11
0 2
30 19
2 y 1 mo 6 y 5 mo
M2 M2
8 40
20 51
MS MS
1 1
L2eL4 T9eS1
1 25
0 30
TLSO Cast and TLSO TLSO TLSO
6 y 5 mo 7 y 5 mo
M2 M1 M2 M2 M2 M1 M2 M1 M2 M2 M2 M2 M1 M1
18 11 22 42 32 18 15 45 44 26 7 14 40 49
25 45 30 40 46 35 23 45 30 32 40 36 45 40
MS MS MS MS MS MS MS MS MS H, PS MS MS MS PS
1 1 1 1 1 1 1 1 1 1 1 1 1 1
L3eL4 T12eS1 L2eL3 L2eL4 T10eT11 L3eL4 T9eT12 L1eS1 L2eL4 L3eL4 L1eL4 L1eL2 L2eL4 L2eS1
8 18 11 19 19 22 11 5 20 7 1 5 15 0
3 20 12 20 10 25 20 15 12 26 6 20 12 17
No TLSO TLSO TLSO TLSO TLSO TLSO TLSO TLSO No Cast No Cast TLSO
6 y 9 mo 6 y 7 mo 5 y 6 mo 6 y 5 mo 18 y 6 mo 10 y 7 mo 6 y 4 mo 4 y 4 mo 5 y 9 mo 5 y 8 mo 5 y 11 mo 6 y 7 mo 7 y 8 mo 5 y 7 mo
M2 M2 M1 M1 M1 M2 M2
38 29 21 23 39 42 36
31 40 35 40 22 45 42
MS MS MS PS MS MS MS
1 2 2 2 1 2 2
T11eT12 L2eL4 T2eT7 L1eL4 L5eS1 T12eL1 L2eL3
4 21 6 17 22 21 20
10 16 20 4 25 20 30
Cast TLSO No No TLSO TLSO YESO
10 y 1 mo 5 y 1 mo 4 y 10 mo 5 y 2 mo 10 y 4 mo 3 y 9 mo 8 y 10 mo
y y y y y y y
7 7 9 9 5 2 5
mo mo
mo mo mo
5y2 10 y 8 mo 5 y 5 mo 5 y 2 mo
(Continued on next page)
544
L. Piantoni et al. / Spine Deformity 3 (2015) 541e548
Table (Continued ) Cases
Age at surgery
HV level HV side
Type MAV MAV Implant Rod Arthrodesis Postoperative Kyphosis Scoliosis number control
Postoperative Follow-up orthosis
MAV MAV kyphosis scoliosis Case 47 6 y 2 mo Case 48 1 y 1 mo Case Case Case Case Case
49 50 51 52 53
5 7 2 2 7
y y y y y
5 7 3 4 6
mo mo mo mo mo
Case 53 2 y 9 mo Case 54 2 y 9 mo Case 55 7 y 6 mo Case 56 2 y 1 mo Case 57 4 y 2 mo
L5 L3
Right Right
M2 M2
50 10
34 46
PS MS
2 1
L5eS1 L3eL4
25 0
15 15
T11 L2 L4 L1 T9 left, T10 right
Left Left Right Left Bilateral double contralateral Right Left Right
M2 M1 M1 M2 M2
24 32 41 28 42
21 43 30 32 50
TM TM TM PS PS
2 2 2 1 2
T10eT12 L1eL3 L3eL5 T12eL2 T8eL4
8 9 19 2 20
2 21 22 24 32
M2 M2 M1
40 37 60
37 20 32
PS MS PS
2 2 1
T10eT12 T12eL2 T9eT11
30 10 11
15 4 10
M2 M2
43 11
32 35
MS MS
2 1
T10eT11 T9eL2
35 10
M2 M2
41 10
35 44
MS H, PS
2 1
T12eL2 T8eL3
M2 M2 M2 M2 M2 M2 M2
41 65 30 38 43 35 42
30 40 25 35 28 25 110
H PS PS MS PS MS H, PS
2 2 2 1 2 1 2
M2
10
32
H
2
T11 L4 T10
T8 Left T10 left, Bilateral L2 right double contralateral Case 58 5 y 6 mo L1 Right Case 59 4 y 6 mo T10, L3 Bilateral double ipsilateral Case 60 7 y 8 mo L2 Left Case 61 2 y 11 mo T11 Right Case 62 2 y 4 mo L1 Left Case 63 2 y 1 mo T12 Right Case 64 5 y 6 mo T8 Left Case 65 4 y 1 mo L1 Right Case 66 3 y 7 mo T10, L1 Bilateral double contralateral Case 67 7 y 8 mo L2 Left
Cast Cast and TLSO Cast No Cast TLSO Cast
5 y 3 mo 9 y 6 mo 5 y 2 mo 10 y 2 mo 8 y 3 mo 8 y 3 mo 5 y 11 mo
10 15
Cast TLSO Cast and TLSO Cast Cast
4 y 1 mo 5 y 1 mo 4 y 4 mo 4 y 10 mo 6 y 3 mo
48 4
19 12
TLSO TLSO
6 y 5 mo 4 y 8 mo
L1eL3 T9eL1 T12eL1 T12eL1 T7eT9 T12eL1 T9eL2
21 75 13 18 35 10 22
22 40 7 4 15 15 50
TLSO TLSO TLSO No No TLSO Cast and TLSO
4 5 4 5 5 6 6
L1eL3
16
59
Cast and TLSO
8 y 6 mo
y y y y y y y
4 mo 3 mo 5 mo 4 mo 4 mo 10 mo 11 mo
FS, fully segmented; H, hooks; HV, hemivertebra; M0, Mobile 0; M1, Mobile 1; M2, Mobile 2; MAV, mean angular value; mo, month(s); MS, monoaxial screws; PMS, poly- and monoaxial screws; PS, polyaxial screws; SH, screws and hooks; SS, semisegmented; TLSO, thoracolumbar orthosis; US, unsegmented; y, year(s).
resection of the hemi-laminae was performed, and the internal wall of the pedicle was removed. The technique was completed with the resection of the discs. The egg-shell procedure was the mother surgical technique during these 70 procedures. We tried to resect as much of the contralateral side of the disc as possible concerning the HV when a unilateral approach was used. This was probably one of the most demanding steps of the surgical technique. But crucial to prevent pseudarthrosis, the contralateral disc should be removed when all of the HV is resected. It is extremely difficult to close the wedge from behind without a complete body and disc removal. In other patients, however, the HV was resected from outside, and this was done according to surgeon’s criteria. In both cases, bleeding was carefully controlled. One or two rods were placed. When curves where not severe, instrumentation was limited to adjacent vertebras; however, when curves were severe, instrumentation included more
levels. Compression and/or distraction was performed when necessary. Lastly, the surgeon proceeded with final closure. Before the patient left the operating room, neurophysiology and neurology experts verified that nerve roots and spinal cord were unharmed (Fig.) [4,18]. One of the most important indications of a bilateral approach are kyphotic cases, kyphosis over scoliosis curve pattern, children with high BMI, older children, multiple HVs, HV with contralateral bone bar, and HV with other associated vertebral anomalies; in addition, we have the unilateral indications [19]. In clinical practice, when kyphosis is present, we recommend a bilateral approach with instrumentation. And we do agree that when kyphosis is present, two-level instrumentation might not be good enough to achieve a secure as well as solid level of fusion. We strongly recommend against performing unilateral instrumentation in kyphotic cases.
L. Piantoni et al. / Spine Deformity 3 (2015) 541e548
545
Fig. Image 1. Spine X-ray, anteroposterior (1A) and lateral (1B). One-year-10-month-old female patient, lumbar scoliosis of 45 , left L2eL3 hemivertebra Mobile 2 or fully segmented (Tello et al. [4-14,17]), preoperative. Image 2. Same patient, computed tomographic scan 3D image reconstruction, anterior (2A) and (2B) posterior view, same findings, preoperative. Image 3. Spine X-ray anteroposterior (3B) and lateral (3A), 7 months after surgery, left L2eL3 instrumented posterior arthrodesis, satisfactory outcome.
546
L. Piantoni et al. / Spine Deformity 3 (2015) 541e548
Results According to HV classification [4,17], 17 of the total 78 were HV type M1 (21.79%) and 61 were HV type M2 (78.21%). In relation to HV location, 28 were thoracic (35.9%), 16 thoracolumbar (20.51%), 28 lumbar (35.9%), and six lumbosacral (7.69%). Mean age of the patients at surgery was 5 years 5 months (from 8 months to 16 years). Eight patients had double HVs, which were resected in the same surgical procedure, and one patient had three HVs and required a two-step surgery. Mean postoperative follow-up was 6.55 years, with a maximum of 10 years 8 months and a minimum of 2 years 1 month. Implants used were monoaxial screws in 45 patients (67.17%), polyaxial screws in 17 patients (25.38%), hooks in 3 patients (4.47%), screws and hooks in 1 patient (1.49%), and poly- and monoaxial screws in 1 patient (1.49%). There was only 1 case of rib resection (1.49%) among the 70 procedures performed, which was required for HV excision. One rod was used in 38 surgical procedures (54.28%) and 2 rods in 32 procedures (45.72%). Intraoperative arthrodesis with allograft and autograft was performed in 64 patients (91.42%). Fifty-seven patients were immobilized with some kind of orthosis (85.07%). Scoliosis preoperative mean angular value (MAV) was 38.55 (range 5 12 e110 ), and scoliosis postoperative MAV was 19.89 (range 5 0 e59 ). Mean correction for scoliosis was 48.37%. Kyphosis preoperative MAV was 29.98 (range 5 3 e65 ), and kyphosis postoperative MAV was 15.41 (range 5 0 e75 ). Mean correction for kyphosis was 48.73%. Two patients suffered from a neurophysiological event (2.98%) that was reverted with minimal maneuvers (reduction of compression, distraction, reorientation of the screw, or rod remodeling) and presented no postoperative sequelae [20]. Medium-term postoperative complications were observed in 4 patients (5.63%): one patient presented ipsilateral/unilateral sciatica associated with the resected HV, whereas two patients developed superficial infections that were resolved with surgical toilet and intravenous antibiotic therapy. Eight patients (11.26%) presented incorporation of vertebras into the primary curve. Two of these patients required revision surgeries in the long term, one due to decompensation of the arthrodesed curve and proximal junction kyphosis and the second due to crankshaft effect.
Discussion HV resection is a preventive and corrective procedure performed in children before they develop a severe deformity, which is a practice accepted worldwide [21-23]. The complete resection of the HV is also performed in older patients in order to correct and stabilize the trunk. Several types of HV resections have been described in the past [24,25].
The first HV resection technique was described by Royle in 1928, and this was followed by Compere et al. In 1968, Winter and colleagues studied 234 patients in order to establish the diagnostic and prognostic criteria. However, not all patients were surgically treated. In 1975, with similar goals, Nasca evaluated a total of 60 patients. In 1979, Leatherman was the first to report a two-step surgery. Initially, the excision was performed by anterior and posterior approach [17], which allows for a better visualization of the defect and also the possibility to resect the adjacent discs with better outcomes and lower risks [26-28]. However, the anterior approach, either by thoracophreno-lumbotomy or thoracotomy, turns the procedure into a more aggressive one with a higher complication rate. The posterior approach allows the surgeon to reach the anterior spinal cord through the technique described by Harms et al., transforming the HV excision into a less aggressive treatment [7,13,29,30]. In addition, this approach allows for the complete removal of the neural arch with a more limited exposure of the HV body, which is less developed anteriorly and occupies only the posterior part. The posterior removal of the pedicle and vertebral body is done using either the outsideeinside or the egg-shell technique. Final removal of the laminae and inner wall of the pedicle is required to avoid or delay the epidural bleeding. We did not resect systematically the posterior part of the rib in patients with thoracic HV. Because did not perform comparative studies, we are not sure about its benefits. Our approach has been unilateral and, in some cases, bilateral including instrumentation. The demographic data of our study shows 38 cases with unilateral instrumentation (54.28%) and 32 with bilateral (45.72%). Bilateral or unilateral instrumentation was selected according to age of the patient, presence of opposite bone bar, angular value of the curve, and surgeon preference. Young children, without contralateral pathology, low body mass index, and when scoliosis angular value was more important than kyphosis angular value, were treated unilaterally [4,8,17]. Older patients, with a more severe deformity, a predominant kyphosis component, bone bars, or contralateral malformations, were treated bilaterally [4,8,17,31,32]. Current implants that allow the placement of pedicular screws and cervical implants can sometimes be used in very young children. The use of hooks has been previously described but located near the center of vertebral rotation may cause minor correction. However, our first choice is the placement of pedicular screws in the adjacent vertebras of the HV to be removed [30,33-36]. When we detected difficulties using the screws, we used sublaminar hooks. This was justified by careful observation of the surgical anatomy with images of 3D computed tomographic scan before surgery and intraoperative X-ray assessment. However, there were times when intraoperative anatomical findings did not match the computed
L. Piantoni et al. / Spine Deformity 3 (2015) 541e548
tomographic scans and preoperative X-ray. In order to save fusion levels, careful anatomical assessment is crucial to the instrumentation. When placing the pedicular screws, it is preferable to direct them to the median sagittal of the instrumentation, with convergent inclination. When the compression was done, we observed a lordotic effect. The corrections observed were satisfactory. Scoliosis preoperative MAV was 38.55 , and postoperative MAV was 19.89 , which represents a correction of 48.37%. Moreover, kyphosis preoperative MAV was 29.98 , and the postoperative was 15.41 , which represents a correction of 48.73%. It is important to note that in very young children where the objective of the surgery is preventive, the correction should be evaluated over time. Not all HVs have to be excised, only those that are completely mobile (M2) [4,17], nonincarcerated, and provoking decompensation (ie, located in lumbar and/or lumbosacral area) frequently progress. Therefore, we indicate the removal of the HV with preventive criteria. Patients could undergo surgery once they are 18 months old approximately, but there is no evidence to establish this age limit. However, the volume and bone size are parameters to take into account in order to remove the HV successfully before it produces greater deformity. The majority of our patients have been treated with instrumentation and arthrodesis, and some of them with a unilateral approach and without bone graft. We could not establish if those patients would have a better or worse outcome. We did not detect a relationship between nofusion and complications. However, a larger number of patients with longer follow-up period would be necessary to evaluate this. Importantly, the number of patients who maintained the correction or required subsequent surgeries must be evaluated over time and at the end of the spine growth. There were no significant complications, either intraoperative or immediate postoperative.
Conclusion Complete resection of the HV by a posterior approach only tends to be a simple, fast, and reliable procedure that allows for the complete removal of the formation defect with satisfactory results in the long term. This technique has been demonstrated to be less invasive and to achieve an accepted correction in the sagittal and coronal planes.
References [1] Hedequist D, Emans J. Congenital scoliosis. J Am Acad Orthop Surg 2004;12:266e75. [2] Ruf M, Jensen R, Harms J. Hemivertebra resection in the cervical spine. Spine (Phila Pa 1976) 2005;30:380e5.
547
[3] Hedequist DJ, Hall JE, Emans JB. Hemivertebra excision in children via simultaneous anterior and posterior exposures. J Pediatr Orthop 2005;25:60e3. [4] Tello C. Posterior hemivertebrae resection. Scoliosis Research Society 43rd Annual Meeting. San Antonio: Instructional Regional Course; 2010. [5] Benli IT, Aydin E, Alanay A, et al. Results of complete hemivertebra excision followed by circumferential fusion and anterior or posterior instrumentation in patients with type-IA formation defect. Eur Spine J 2006;15:1219e29. [6] Nakamura H, Matsuda H, Konishi S, Yamano Y. Single-stage excision of hemivertebrae via the posterior approach alone for congenital spine deformity: follow-up period longer than ten years. Spine 2002;27:110e5. [7] Ruf M, Harms J. Hemivertebra resection by a posterior approach: innovative operative technique and first results. Spine 2002;27: 1116e23. [8] Klemme WR, Polly DW, Orchowski JR. Hemivertebral excision for congenital scoliosis in very young children. J Pediatr Orthop 2001;21:761e4. [9] Shono Y, Abumi K, Kaneda K. One-stage posterior hemivertebra resection and correction using segmental posterior instrumentation. Spine 2001;26:752e7. [10] Bollini G, Docquier PL, Launay F, et al. Results at skeletal maturity after double-approach hemivertebral resection. Spine 2008;33: 2001e6. [11] Lin RM, Huang KY, Lai KA. Mini-open anterior spine surgery for anterior lumbar diseases. Eur Spine J 2008;17:691e7. [12] Garrido E, Tome-Bermejo F, Tucker SK, et al. Short anterior instrumented fusion and posterior convex non-instrumented fusion of hemivertebra for congenital scoliosis in very young children. Eur Spine J 2008;17:1507e14. [13] Ruf M, Harms J. Posterior hemivertebra resection with transpedicular instrumentation: early correction in children aged 1 to 6 years. Spine 2003;29:1593e4. [14] Winter RB, Burger EL. L5 hemivertebra resection and T12-S1 fusion in a 14-year-old female with a 36-year follow-up. Spine 2012;37: E445e50. [15] Bollini G, Docquier PL, Viehweger E, Launay F, Jouve JL. Lumbar hemivertebra resection. J Bone Joint Surg Am 2006;88:1043e52. [16] Li X, Luo Z, Li X, et al. Hemivertebra resection for the treatment of congenital lumbarspinal scoliosis with lateral-posterior approach. Spine (Phila Pa 1976) 2008;33:2001e6. [17] Tello C. Hemivertebra excision: Proceedings of the 18th Annual Meeting, Scoliosis Research Society, 1983:168. [18] Yaszay B, O’Brien M, Shufflebarger HL, et al. Efficacy of hemivertebra resection for congenital scoliosis: a multicenter retrospective comparison of three surgical techniques. Spine 2011;36:2052e60. [19] Peng X, Chen L, Zou X. Hemivertebra resection and scoliosis correction by a unilateral posterior approach using single rod and pedicle screw instrumentation in children under 5 years of age. J Pediatr Orthop B 2011;20:397e403. [20] Zhang J, Shengru W, Qiu G, et al. The efficacy and complications of posterior hemivertebra resection. Eur Spine J 2011;20:1692e702. [21] Winter RB, Lonstein JE. Scoliosis secondary to a hemivertebra: seven patients with gradual improvement without treatment. Spine 2010;35:E49e52. [22] Obeid I, Bourghli A, Vital JM. Lumbar hemivertebra resection by posterior approach for congenital scoliosis. Eur Spine J 2012;21:2721e3. [23] Obeid I, Bourghli A, Vital JM. Thoracic hemivertebra resection by posterior approach for congenital scoliosis. Eur Spine J 2013;22: 678e80. [24] Jeszenszky D, Fekete TF, Kleinstueck FS, et al. Fusionless posterior hemivertebra resection in a 2-year-old child with 16 years follow-up. Eur Spine J 2012;21:1471e6. [25] Wang S, Zhang J, Qiu G, et al. Posterior hemivertebra resection with bisegmental fusion for congenital scoliosis: more than 3 year
548
[26]
[27]
[28]
[29]
L. Piantoni et al. / Spine Deformity 3 (2015) 541e548 outcomes and analysis of unanticipated surgeries. Eur Spine J 2013;22:387e93. Jalanko T, Rintala R, Puisto V, Helenius I. Hemivertebra resection for congenital scoliosis in young children: comparison of clinical, radiographic, and health-related quality of life outcomes between the anteroposterior and posterolateral approaches. Spine 2011;36: 41e9. Xu W, Yang S, Wu X, Claus C. Hemivertebra excision with shortsegment spinal fusion through combined anterior and posterior approaches for congenital spinal deformities in children. J Pediatr Orthop B 2010;19:545e50. Mladenov K, Kunkel P, Stuecker R. Hemivertebra resection in children, results after single posterior approach and after combined anterior and posterior approach: a comparative study. Eur Spine J 2012;21:506e13. Deviren V, Berven S, Smith JA, et al. Excision of hemivertebrae in the management of congenital scoliosis involving the thoracic and thoracolumbar spine. J Bone Joint Surg Br 2001;83: 496e500.
[30] Ruf M, Jensen R, Letko L, Harms J. Hemivertebra resection and osteotomies in congenital spine deformity. Spine 2009;34:1791e9. [31] Aydogan M, Ozturk C, Tezer M, et al. Posterior vertebrectomy in kyphosis, scoliosis and kyphoscoliosis due to hemivertebra. J Pediatr Orthop B 2008;17:33e7. [32] Li XF, Liu ZD, Hu GY, et al. Posterior unilateral pedicle subtraction osteotomy of hemivertebra for correction of the adolescent congenital spinal deformity. Spine J 2011;11:111e8. [33] Freeman BJ, Oullet JA, Webb JK. Excision of hemivertebrae in the management of congenital scoliosis. J Bone Joint Surg Br 2002;83: 496e500. [34] Hedequist D, Emans J, Proctor M. Three rod technique facilitates hemivertebra wedge excision in young children through a posterior only approach. Spine 2009;34:E225e9. [35] Zarghooni K, Sobotrke R, Schmidt H, et al. Dorsal resection of a thoracic hemivertebra in a 4-year-old boy with endochondral gigantism. A case report. Acta Orthop Belg 2010;76:699e705. [36] Halm H. Transpedicular hemivertebra resection and instrumented fusion for congenital scoliosis. Eur Spine J 2011;20:993e4.