Treatment of Scoliosis by Hartshill Rectangle and Sublaminar Wiring

Background: Scoliosis is a complex three dimensional deformity characterized by coronal, sagittal and horizontal plane deviation. Treatment of scoliosis has made significant development in techniques and modalities of internal fixation. Objective: To see the efficacy of Hartshill rectangle and sublaminar wiring for correction of scoliotic deformity, clinical outcome and patient’s satisfaction. Methods: During the period of July 2008 to June 2011, 14 patients of scoliotic spinal deformities of various etiologies (Congenital-02, idiopathic-12) were operated. Age ranged from 11 to 23 years with mean 16.71 ± 03.22 years. Female 09 and male 05 patients were included where Hartshill rectangles with sublaminar wiring were used for every case. Results: At least 12 months (range 12-36 months) follow up regarding clinico-radiological evaluation and patient’s satisfaction were accounted as per the SRS-24 questionnaire. The mean scoliosis in the coronal plane (Cobb’s angle) measured 53.50o ± 03.98o preoperatively reducing significantly to 21.14o ± 02.77o (p<0.05) and the mean preoperative sagittal plane angle 60.14o ± 07.30o reducing significantly to average 42.14o ± 05.28o (p<0.05) at 1 year as determined by paired t-test. No neurological complication was seen except transient hyperesthesia in 02 patients, superficial wound infection in 01 patient and 01 wire breakage. All the patients were happy to live with their present back status. Conclusion: Overall outcome in segmental spinal fixation with sublaminar wire and Hartshill rectangle is satisfactory and effective.

L-shaped rods in1970s.Later the wiring of the apical vertebrae to the Harrington rod became a popular modality to correct coronal spinal deformities in the early 1980s.Hartshill rectangle was introduced by Dove 9 in 1986.Although posterior surgical techniques have been shown to provide a limited correction of the deformity in only one plane and leaving the rotational hump, 10 significant correctability and ability to maintain correction in all planes can be achieved by locally made stainless steel Hartshill loop rectangle and sub laminar wires. 11though segmental spinal instrumentation by sublaminar wiring and Hartshill loop rectangle, had fallen out of favor in comparison to the newer third generation implants 11 , it is still considered a method of choice of surgical correction of scoliotic deformities. 12,13,14 Te field of scoliotic surgery has long been unexplored in our country due to lack of expertise as well as costly and relatively unaffordable price of the implants.As cost of implants had been a major issue for most of our patients, in our socio-economic aspect we have chosen to evaluate the efficacy of low cost locally made Hartshill rectangle and sublaminar wires as a modality for correction of scoliotic deformity and its co-relation with patients' satisfaction and clinical outcome.

Methods:
The study is performed prospectively from July 2008 to June 2011 in Bangabandhu Sheikh Mujib Medical University and a private hospital in Dhaka.We have evaluated the surgery on 14 consecutive scoliotic patients of different etiologies which includes 12 cases of Adolescent Idiopathic Scoliosis (AIS) and 02 cases of congenital scoliosis.There were 09 female and 05 male presenting within an age ranging from 11 to 23 years with a mean age of 16.71 ± 03.22 years.The patients in idiopathic type were further sub classified according to King's and Moe's classification system as Type-2 (04 patients) and Type-3 (08 patients).There were no patients with King's and Moe's Type-1, Type-4 and Type-5.Ten patients presented with right convex curve and 04 with left sided convexity.The patients with the following criteria had been included for the study, i) Cobb's angle > 40°; ii) Progressive increase of the Cobb's angle > 10° in successive 6 months follow up; iii) Patients demand for surgery for cosmetic purpose.The exclusion criteria for the patients had been, i) Cobb's angle < 40°; ii) Cobb's angle > 60°.Preoperative SRS 24 satisfaction questionnaires were completed by all the 14 patients, while Cobb's angles and radiological sagittal profiles were measured by qualified radiologists.Preoperative MRI scans were done in all patients to exclude any intraspinal anomaly.The Rib hump was evaluated clinico-radiologically and was corrected by posterior costoplasty in selected cases.Patients routinely followed up at 3, 6 and 12 months in the first year after surgery, and then once a year.At every visit follow up evaluation was performed by measuring the coronal Cobb's angle (anteroposterior view) and evaluating the sagittal balance (lateral view) [figure-1(g)].All the 14 patients filled the SRS 24 satisfaction questionnaire 15 and Pain status of the patients was evaluated by the Visual Analogue Score (VAS) 16 in every follow up.

Operative procedure:
The extent of exposure is determined preoperatively by radiological or fluoroscopic assessment of both the upper and lower target segment and marked over the skin accordingly.All the patients underwent surgery by posterior midline approach but the incision was designed by keeping the postoperative scar in mind.The dissection was carried out meticulously with proper haemostasis upto the transverse processes.The spinous processes were excised, prepared and kept for grafting.The sublaminar spaces are created and cleared by cutting the interspinous ligaments.The midline ligamentum flava was removed with care.The extent of vertebral level involvement is followed according to the basic principles of scoliosis surgery proposed by Goldstein 17 and Moe 18 .We prefer to involve two vertebra above and below the corresponding upper and lower end vertebra.
After exposure of the sublaminar spaces 20 gauge stainless steel double looped wires are passed at each level in caudal to cephalad direction.The facetal preparation for fusion is performed.The size of Hartshill loop rectangle is measured and adequately contoured.Wires are tied onto the loop rectangle with cephalad wire ends inside the loop rectangle and caudal wire ends outside the loop rectangle at each level.This maneuver is reversed at the lower end of the construct to prevent the caudal migration of the construct.Tightening of concave wires are performed first at each level which causes translation of the curve on to the loop rectangle and then the wires of the convex side is tightened which essentially derotates the curve.All the wires were tightened 1-2 times before final tightening is done.This corrects the spinal deformity onto the loop rectangle [figure-1(f)].The sublaminar spaces are packed with spongostum.To enhance the posterior fusion process thorough shingling is done over the posterior surface of laminae and graft bed is prepared where bone chips collected from the excised portion of the spinous processes are laid down.The entire wound is irrigated and then closed in layers keeping a drain in situ.

Results:
All fourteen patients underwent posterior surgery and posterior instrumentation with Hartshill rectangle and sublaminar wiring.Posterior costoplasty was done in two patients.Autogenous bone graft taken from excised spinous process, excised rib mixed with freeze dried allograft taken from Savar Atomic energy centre was used in all patients.We used 20 gauge, double looped, precontour stainless steel sublaminar wires and a 6 mm thick stainless steel spinal rectangle for the spinal instrumentation.As there is un-availability of intraoperative monitoring system of spinal cord, we sometimes used 'wake up' test.Postoperative mobilization was done at an average of 6 days (5 to 12 days) .We did not use any postoperative bracing.
Follow-up ranged from 12 to 36 months, the average being 14.5 month.As showed in (p<0.05) with 29.77% correction at 1 year follow up remaining within the near normal anatomical range and all the cases had a normal balanced spine based on radiological evaluation and one of the case is illustrated in figure 1(a)-(g).
Surgical time for posterior surgery ranged from three hours to four hours 15 minutes (average of three hours and 40 minutes and average blood loss was 800cc.The lower extent of posterior spinal arthrodesis was a L2 (n= 2) patients, L3 (n= 5) patients, L4 (n= 6) patients, L5 (n= 1) patients.Total 384 sublaminar wire were introduced in 192 laminar levels.The pain status improved but not significantly from preoperative average of 3.73 to 2.93 at 1 year follow-up.The patients' satisfaction index as per the SRS24 questionnaire in 14 patients revealed that 80.48% of patients were satisfied and 71.42% were happy to live with their present back status lifelong.No permanent neurological complication was seen, only transient hyperesthesia developed in 02 (14.29%) patients who were improved after 6 months.Superficial wound infection developed in 01 (07.14%) patient which was managed conservatively.
Although segmental spinal instrumentation, done by sublaminar wiring and Hartshill loop rectangle, had fallen out of favor in comparison to the third generation implants due to various drawbacks, 21 significant correction and ability to maintain it is still possible in all planes, at a comparatively low cost with this technique. 11 An ideal system of spinal fixation would have several characteristics.It would be safe and easy, provide rigid fixation and predictable correction with a minimum post operative external support required, be applicable to a wide variety of spinal deformities, and have a high rate of fusion with a low incidence of failure of the instrumention.Segmental Spinal instrumentation with wires appears to meet the majority of these criteria and offers several advantages over other forms of segmental and non segmental fixation. 22There is an even distribution in corrective forces, with two lateral fixation points on each segment which provide a good rotatory control.Sagittal plane correction is also comparable to the third generation implants, as care is taken in contouring the loop rectangle before tightening the wires. 23,24However, the loop rectangle and wire constructs aim at translation and coronal and sagittal balancing rather than derotation as its principle, so extent of derotation is not the purpose that has been intended to achieve.
As like as the socioeconomic condition, the awareness of availability of surgical solution is not yet well established in our country because of the unavailability of the appropriate surgical set up.We are the only institute in the country performing the deformity surgery with a relatively better anaesthetic facility and surgical setup but most of the cases are late and with advanced deformity.
We  11 which is a little bit higher than ours as because they had surgeries on Neuromuscular Scoliosis which we did not.The patients' satisfaction index as per the SRS24 questionnaire in our 14 patients revealed that 80.48% of patients were satisfied with the surgery and ready to undergo the same surgery if required.On, whether they were happy to live with their present back status lifelong, 71.42% agreed.This observation is also comparable to Bhojraj et al. 11 Neurological deficit rates up to 17% have been reported with sublaminar wiring. 28,29The SRS Committee has reported neurological deficit risk to be 0.03%. 30Asher 31 reported neurological deficit score as 0.03% in his series of 355 patients, but Benli et al. 25 have reported no neurological deficit in patients operated with ISOLA Instrumentation.In our study no neurological deficit has been observed with sublaminar wiring, only 02 (14.29%) cases developed transient hyperaesthesia which was improved after 06 months.We have observed only 01 (07.14%) case developing superficial wound infection which was improved by conservative measures.Asher 31 reported 1-2% early and 1 % late infection rate in his series and Richards et al. 32 reported 10% late infection rate that was managed with debridement.We observed only 01 wire breakage whereas Bhojraj et al. 11 had 06 broken wires and none of them needed re-exploration as they were asymptomatic as like as ours.

Conclusion:
Overall outcome in segmental spinal fixation with sublaminar wire and hartshill rectangle were satisfactory.
No decompensation and imbalance have been observed and almost totally balanced vertebral column has been obtained in all patients.Given these results, in our opinion, this technique is a successful alternative and effective modality of treatment for scoliotic patients where advanced surgical options are not available or attainable.
the patients were evaluated clinically [Figure-1(a), (c)] as well as radiologically by doing a scoliogram film in standing posture P/A view extending from C 2 to S 2 as well as a lateral view in supine position of the same extent.The right and left lateral bending views were also evaluated preoperatively [Figure-1(e)].These key radiological images were helpful in determining the Cobb's angle, sagittal profile, flexibility of the spinal column as well as determining the extent of fusion.The average preoperative Cobb's angle was 53.50 o .The saggital profiles of all the patients were measured between T 2 -T 12 for thoracic kyphosis and L 1 -S 1 for lumbar lordosis.The normal thoracic kyphosis was taken as 20 o -40 o and lumbar lordosis as 40 o -60 o .In our patients the thoracic curves averaged 60.14° and the lumbar curves 44°.

Fig.- 1 :
Fig.-1: Clinical and radiological assessment of scoliotic patients.Preoperative clinical assessment of the patients regarding the curve, flexibility, correctivity and assessment of the rib hump, (a) (c), the postoperarive assessment shows a good cosmetic correction of the deformity (b) (d), the radiological assessment of the spine in standing and bending scoliogram (e), the peroperative view after instrumentation (f) and the postoperative X-ray showing correction of the curve in both sagittal and coronal plane (g).
table-I and figure-2, the mean scoliosis in the coronal plane (Cobb's angle) measured 53.50 o ± 03.98 o preoperatively and 18.86 o ± 02.86 o postoperatively (p<0.05).After 1 year of follow up the mean Cobb's angle was 21.14 o ± 02.77 o which was statistically significant (p<0.05) with an average loss of correction of 02.21 o .The average correction immediate postoperatively was 32.14 o which reduced to 29.87 o at 1 year follow up.We have observed average 60.58% correction at 1 year follow up.The younger age group had a better percentage of correction 62.60% than the 57.85% correction in older age group.The preoperative mean kyphotic angle was 60.14 o ± 07.30 o which was significantly corrected to mean 39.50 o ± 02.46 o (p<0.05) at immediate postoperative period that reduced to 42.14 o ± 05.28 o

Table - I
Demographic characteristics, peroperative vertebral level of fixation and postoperative outcome of surgery.
22e mean scoliosis in the coronal plane (Cobb's angle) measured 53.50 o ± 03.98 o pre-operatively and 18.86 o ± 02.86 o immediate post operatively.We observed average percentage Coronal Cobb's angle correction of 60.58% which was statistically significant (p<0.05).In the series of 36 patients Bhojraj et al.11showed 73.25° preoperative Cobb's angle and 26.12° immediate postoperatively with average angle of correction is 64.34%.Although we have selected patients with relatively less steep curves the average degree of correction is almost comparable.Studies conducted by Mueller et al.26and Remes et al.27showed 62.50% and 62-67% correction which is also comparable to our results, although Benli et al.25showed a larger degree of correction of 79.9°.The loss of correction at 24 months follow-up was 02.21° which is also comparable to the findings of the study conducted by Bhojraj et al.11who found 2.2°, moreover Herndon et al.22showed range of 0°-10° loss of correction in 2-7 years follow up.The preoperative average kyphotic angle in our series was 60.14 o ± 07.30 o which was corrected to average 39.50 o at immediate postoperative period that reduced to 42.14 o ± 05.28 o with 29.77% correction at 1 year follow-up.The correction was also statistically significant (p<0.05) and remaining within the near normal anatomical range.The average correction was 18 o (range 12 o -25 o ) which is comparable to the study byBhojraj et al. 11where he showed 22.60 o correction with a range of 04°-32°.The patients' pain status as assessed by the VAS does not depict a significant change as the same scenario described by other studies.Average preoperative pain score according to VAS was 0.50 that had not been changed even at 1 year follow up.Average pain 4.2 -4.5 was observed byBhojraj et al.