Serial Derotational Casting in Congenital Scoliosis as a Time-buying Strategy

Transcription

1 ORIGINAL ARTICLE Serial Derotational ing in Congenital Scoliosis as a Time-buying Strategy Halil G. Demirkiran, MD,* Senol Bekmez, MD,w Rustem Celilov, MD,* Mehmet Ayvaz, MD,* Ozgur Dede, MD,z and Muharrem Yazici, MD* Background: Serial casting is an effective treatment modality in early-onset idiopathic scoliosis; however, the role of this method in congenital scoliosis is not well studied. Methods: A total of 11 patients with progressive congenital scoliosis were treated with serial cast application. Age at initial cast application, magnitudes of the congenital, compensatory and sagittal deformities, coronal balance, T1 to T12 height, number of casts and time-in cast per patient, subsequent surgical interventions, and complications were evaluated. Results: Mean age at the first cast application was 40 months, and the average number of cast changes was 6.2 per patient. There were no major complications. The average precasting curve magnitude was 70.7 degrees (range, 44 to 88 degrees) and was significantly reduced to 55.1 degrees (range, 16 to 78 degrees) at the latest follow-up (P = 0.005). The average precasting compensatory curve was 55.8 degrees (range, 38 to 72 degrees) and was significantly reduced to 39.8 degrees (range, 23 to 62 degrees) at the latest follow-up (P = 0.017). Average T1 to 12 height increased from 12.8 cm at post-first cast to a 14.6 cm at the latest follow-up (P = 0.04). Average time in cast was 26.3 months (range, 13 to 49 mo). During the treatment period, none of the patients required surgery for curve progression. Conclusions: Serial derotational casting is a safe and effective time-buying strategy to delay the surgical interventions in congenital deformities in the short-term follow-up. Level of Evidence: Level IV, case series. Key Words: congenital scoliosis, serial casting, early-onset (J Pediatr Orthop 2015;35:43 49) In congenital spinal deformity, surgery is the preferred method of treatment to stop progression, as bracing and casting methods are believed to be ineffective to control the curve progression in most cases. 1,2 As the serious From the *Department of Orthopaedics and Traumatology, Hacettepe University Faculty of Medicine; wdepartment of Orthopaedics and Traumatology, Dr Sami Ulus Children s Hospital, Ankara, Turkey; and zdepartment of Orthopaedic Surgery, Children s Hospital of Pittsburgh, University of Pittsburgh Medical Center (UPMC), Pittsburgh, PA. The authors declare no conflicts of interest. Reprints: Muharrem Yazici, MD, Department of Orthopaedics and Traumatology, Hacettepe University Faculty of Medicine, Sihhiye, Ankara, Turkey. mimyazici@gmail.com. Copyright r 2014 by Lippincott Williams & Wilkins long-term consequences of early spinal fusion became apparent, growth-friendly modalities have been gaining popularity in most types of early-onset progressive scoliosis, as well as congenital deformity, to control the curve progression and maintain natural growth of the spine and the thorax However, the high complication rate of growing rod techniques is a concern, and there has been an increasing interest in delaying the start of growing rod surgery in young children. 11 Recently, serial casting regained popularity in the treatment of early-onset scoliosis (EOS). Serial cast correction has been demonstrated to provide complete resolution of mild-to-moderate idiopathic curves (< 50 to 60 degrees) if started at a very early age (< 2 y). 12,13 For older children with more severe curves, there is recent evidence showing that serial casting may also help delay the first growing rod surgery in noncongenital EOS, which would theoretically decrease the number of complications related to recurrent surgical procedures Although the aforementioned reports include a few patients with congenital deformity, the role of serial casting in congenital scoliosis has not been sufficiently studied. Therefore, the aim of this study was to report the outcomes of serial casting in young children with congenital scoliosis and investigate whether serial casting may be used effectively to delay surgical procedures. METHODS A total of 11 patients with progressive congenital scoliosis were managed with serial casting at a single institution between 2008 and The decision to start cast treatment versus surgery was made based on the type and extent of congenital vertebral anomalies. At our institution, patients with short and sharp angular congenital deformities are treated with hemivertebrectomy or in situ fusion, depending on the anomaly. Therefore, only long curves with multiple anomalous vertebrae were started on cast treatment, if deformity progression of >5 degrees was documented over a 6-month period. According to these criteria, all patients who were suitable for growing rods were started on cast treatment. Congenital medical comorbidities accompanying congenital scoliosis were recorded in 3 patients. There was atrial septal defect in 1 patient, horseshoe kidney in 1 patient, and cleft palate in another patient. J Pediatr Orthop Volume 35, Number 1, January

2 Demirkiran et al J Pediatr Orthop Volume 35, Number 1, January 2015 applications were completed under general endotracheal anesthesia, on a specially designed frame. Mehta modification of Cotrel-Morel technique was used for the application of casts. 12,18 Apexes of the kyphotic deformities were meticulously well padded to avoid skin ulcers. Routine cast changes were carried out in 2- to 4- month intervals, depending on the patient s age. Spinal alignment measurements were made on standard standing posteroanterior and lateral radiographs taken in the cast on the day after the cast application. The procured data included age, Cobb angles of congenital and compensatory coronal curves, sagittal deformity magnitude, coronal balance, thoracic height (T1 to T12), number of cast applications, follow-up time, requirement of surgery, and complications. The criteria for discontinuing cast treatment were: documented insufficient deformity control with the cast (> 5 degrees increase in deformity), skin problems that preclude cast application, and refusal of cast application by family. The data were analyzed using the SPSS 15.0 statistical software. The Friedman test was used for nonparametric 2-way analysis of variance and multiple comparisons. The Wilcoxon signed ranks tests were used to compare the measurement variations between the parameters with the level of significance set at P < RESULTS The clinical and radiographic data are summarized in the Table 1. There were 7 female and 4 male patients. All patients had long congenital curves with mixed formation and segmentation anomalies involving 5 to 10 segments. The average age at the index cast application was 39.5 ± 23 months (range, 9 to 80 mo). The average number of cast applications was 6.2 ± 2.1 (range, 3 to 10) per patient. All patients had a minimum of 3 cast changes and a minimum follow-up of 1 year. The mean precasting Cobb angle of the congenital deformity was 70.7 ± 14.4 degrees (range, 44 to 88 degrees). After initial cast application, the main curve magnitude significantly reduced to a mean 54.4 ± 14.7 degrees (range, 22 to 72 degrees) (P = 0.003). At the latest follow-up, the mean Cobb angle of the congenital deformity was 55.1 ± 17.2 degrees (range, 16 to 78 degrees). (Figs. 1, 2). A mean precasting Cobb angle of the compensatory curve of 55.8 ± 13.3 degrees (range, 38 to 72 degrees) was significantly reduced to a mean 43.5 ± 13.6 degrees (range, 27 to 64 degrees) after initial cast application (P = 0.012). At the latest follow-up, the mean Cobb angle of the compensatory curve was 39.8 ± 14.6 degrees (range, 23 to 62 degrees) (Fig. 3). Correction of both the congenital and the compensatory curves were maintained at the latest follow-up, with the average deformity magnitude being still significantly lower than the precasting values (P = and 0.017, respectively). TABLE 1. General Overview on the Patient Data T1-T12 Height (cm) Sagittal Deformityw (deg.) Compensatory Curve* (deg.) Congenital Curve (deg.) Present Status No. s Time in (mo) nl (cm/y) F/U F/U F/U Level Precast F/U Precast Age at (mo) Precast Sex (F/M) Patient # 1 F F F T12-L F T5-T F M Growing rod 7 F T9-L F M M Growing rod 11 M T6-T *The angle of deformity for the compensatory curve if it exists. wthe level and angle of the sagittal deformity if it exists. F indicates female; F/U, follow-up; M, male r 2014 Lippincott Williams & Wilkins

3 J Pediatr Orthop Volume 35, Number 1, January 2015 Serial Derotational ing in Congenital Scoliosis A B C D E F FIGURE 1. A 36-month-old girl with congenital scoliosis. A, Precasting congenital and compensatory curves. B, The absence of any sagittal plane deformity. The posteroanterior (C) and lateral (D) radiographs after the initial cast application show improvement in the congenital and compensatory deformities. At sixth cast application, the congenital and compensatory curve magnitudes have remained stable after 24 months in cast (E and F). Sagittal deformity was present in 4 patients. An average precasting local kyphosis angle of 75.5 ± 11.3 degrees (range, 61 to 85 degrees) was reduced to 62.7 ± 11.9 degrees (range, 48 to 73 degrees) after the first cast application and it was 63.5 ± 11.5 degrees (range, 52 to 78 degrees) at the latest follow-up. r 2014 Lippincott Williams & Wilkins 45

4 Demirkiran et al J Pediatr Orthop Volume 35, Number 1, January 2015 Congenital Deformity (degrees) max min mean The mean precasting 27.4 ± 20.8 mm (range, 4 to 74 mm) of coronal off-balance was reduced to 14.3 ± 9.4 mm (range, 0 to 30 mm) after initial cast application (P = 0.005). At the latest follow-up, mean offbalance was 11.9 ± 10.6 mm (range, 0 to 33 mm). (Fig. 4) The mean T1 to T12 height after the initial cast application was 12.8 ± 3 cm (range, 8 to 18.8 cm). At the latest follow-up, it was significantly improved to a mean 14.6 ± 3 cm (range, 9.1 to 20.3 cm) (P = 0.04). The improvement in T1 to T12 height per year was 0.81 ± 0.14 cm (range, 0.62 to 1.06 cm) per patient. At the time of the last follow-up, surgical intervention was delayed for an average of 26.3 ± 10.8 months (range, 13 to 49 mo). Two patients underwent growing rod surgery. In both patients, the deformity was well controlled with casting. However, in 1 patient, the decision to convert to growing rod was made because of the patient s increased age and size, and for the other, the precast post 1st cast final f/u FIGURE 2. Initial mean angle of deformity for congenital curves and changes after the first and last cast applications. Compensatory Deformity (degrees) max min mean precast post 1st cast final f/u FIGURE 3. Initial mean angle of deformity for compensatory curves and changes after the first and last cast applications. 23 parents decided to switch to growing rod surgery. The remaining 9 patients were still managed with serial casting treatment. For none of the patients in this group, curve progression exceeded >5 degrees within cast. There were no instances of neurological impairment or thoracic wall deformity. Mild skin irritation in 3 patients resolved with local wound care and well padding of the cast. ing was temporarily discontinued in 1 patient because of pneumonia and resumed after resolution of the infection. DISCUSSION Congenital spinal deformities present as a large spectrum varying from mild asymptomatic curves to progressive deformities with neurological and cardiopulmonary sequelae. The natural course of deformity may be predicted according to the type, localization of malformation, and the age of the patient. 19 Although natural history studies are limited, the available data suggest that if left untreated, the combination of mixed vertebral and rib anomalies can restrict the space available for the lung and cause severe disability or even death. 20,21 Therefore, treatment is warranted. There is scant evidence with regard to nonsurgical treatment in congenital scoliosis. 22 Traditionally, conservative treatment methods in congenital scoliosis such as bracing and casting have been thought to be ineffective for the congenital curves because of the rigid nature of the deformity and may only have a corrective effect on compensatory curves. Because of these reasons, surgery is the preferred treatment method if progression is documented, or if the anomaly has high risk of progression or predicted as high risk for progression. 1,2 On the contrary, early definitive fusion is known to be associated with other problems, such as crankshaft phenomenon 23 and impairment of respiratory function. 24,25 The staggering evidence of compromised pulmonary function after extensive thoracic fusions 26 compromises the major impetus to seek for treatment methods that spare spinal growth in the treatment of early-onset spinal deformities. Convex hemiepiphysiodesis may be considered as a growth-friendly surgical method in selected patients with congenital scoliosis. Although historically convex hemiepiphyseodesis has been shown to have unpredictable results, 27,28 recent evidence suggests that a posterior instrumented hemiepiphysiodesis may achieve better deformity control. 29 However, hemiepiphysiodesis is still an invasive procedure regardless of the technique and does not always result in curve correction. Growth-friendly techniques such as growing rods and VEPTR gained popularity in management of all types of EOS, including congenital scoliosis. 3,6,10 However, both of these 2 methods require recurrent and frequent surgical interventions and have a high incidence of complications including implant failure, infection, and wound-healing problems. 11,30 Sankar at al 30 reported complication rates of 2.3 per patient treated with growing rods and 2.7 per patient treated with VEPTR. The rate of complications increases as the treatment period with 46 r 2014 Lippincott Williams & Wilkins

5 J Pediatr Orthop Volume 35, Number 1, January 2015 Serial Derotational ing in Congenital Scoliosis A B C 3.4 cm 7.4 cm D E F 2 cm FIGURE 4. A 19-month-old girl with a significant off-balance (A) and sagittal plane deformity (B). C and D, Improvement in the off-balance and local kyphosis angle. At the time of eighth cast application, posteroanterior (E) and lateral (F) radiographs demonstrate that the deformity remained stable after 35 months in cast. growing rods increase. Bess and colleagues showed that the complication rate is increased by 24% for each of the additional lengthening procedure. In addition, they showed a 13% complication rate decrease for each year of increase in the patient s age at the initial growing rod surgery. 11 These studies suggest that, if the start of growing rod treatment can be delayed, complication rates will drastically decrease. ing had been widely used in the management of scoliosis regardless of etiology or age for postoperative r 2014 Lippincott Williams & Wilkins 47

6 Demirkiran et al J Pediatr Orthop Volume 35, Number 1, January 2015 corrective immobilization until spinal instrumentation methods developed so as to obviate external corrective bracing. 31 Although the role of serial casting as a standalone treatment method for scoliosis in young children had been suggested about 50 years ago by Cotrel and Morel, 18 it took some time for convincing evidence to appear in the literature. Most current studies support the role of serial casting as a definitive treatment modality in the management of mild curves in patients with idiopathic EOS. 12,13 More severe idiopathic and most nonidiopathic curves do not respond to cast treatment as well; however, recent studies showed that casting may still have a role in the treatment of such deformities. Fletcher et al 15 reported their results of cast treatment in 12 idiopathic and 17 nonidiopathic patients. They concluded that spinal growth was maintained, 39 months of delay in surgery was achieved, and growing rod surgery was avoided in 72.4% of their patients. Baulesh et al 14 reported on a cohort of 39 patients including 17 nonidiopathic diagnoses. They concluded that, although nonidiopathic curves did not respond to cast treatment as well as the idiopathic curves, the thoracic growth was maintained and the index surgical procedure was delayed for an average of 2 years. These recent reports on EOS include a small number of congenital deformities; however, there is no report solely evaluating the effectiveness of this method in congenital scoliosis. In this study, we investigated the efficiency of serial casting with Mehta modification of Cotrel method in the management of congenital spinal deformities. There were statistically significant reductions in mean Cobb angles of both congenital and compensatory curves after serial casting. The mean precasting congenital deformity of 70.7 ± 14.4 degrees decreased to 54.4 ± 14.7 degrees, and a mean precasting compensatory curve of 55.8 ± 13.3 degrees decreased to 43.5 ± 13.6 degrees after the initial cast application. The results show that the correction occurred after the first cast application and was maintained over the treatment period. This finding is similar to growing rod application where the major part of correction is achieved during the index implantation. The main goal of growthfriendly approach is to enhance thoracic growth to allow for close to normal pulmonary function in children with EOS. To test how much of this goal could be achieved, we measured the T1 to T12 height in our children. Our data show that the thoracic spine growth averaged 0.81 cm (range, 0.62 to 1.06 cm) per year, despite the presence of congenital anomalies. Although this is less than the normal growth velocity for the 0- to 5-year age group, 32 given the presence of congenital deformities, we conclude that casting preserved the growth potential of the spine. In our group of patients, although 2 patients eventually underwent growing rod surgery, none of the patients required conversion to growing rod treatment because of curve progression and the need for surgery was delayed for an average of 26.3 months. Although modern cast application requires general endotracheal anesthesia, the surgical trauma, risk of infection, and neural complications were avoided. If all patients had an index growing rod implantation and recurrent lengthenings every 6 months, at least 55 surgical procedures would be required in 11 patients (an average of 5 surgical procedures per patient). This number does not take into account the potential additional procedures for implant-related complications such as anchor pull-outs, rod fractures, and wound-related problems, which are inherent to growing rod treatment. Cost and benefit analyses are becoming a major part of health care, and although we did not analyze economic burden of cast treatment versus growing rods, this comprises an area that deserves attention. The limitations of this study mainly stem from the retrospective design. A potential drawback of the serial casting procedure may be the negative effects of the cast on pulmonary function as the patients with congenital scoliosis may already have some pulmonary compromise because of thoracic cage involvement. Recent data suggest that although pulmonary parameters deteriorate after application of the first cast, they return back to baseline values by the time of the application of the second cast. 33 However, the authors of that study did not specify the diagnoses of their patients with EOS, and the effects of a cast in patients with congenital deformities might be different. We did not measure the pulmonary functions before and after the cast applications, and the net effects of cast application on the pulmonary functions of these patients are not known. However, in our patient cohort, casts were well tolerated and we did not encounter any pulmonary insufficiency. Another criticism that can be done for this study is that all radiographic measurements were taken in-cast during follow-ups. However, in our point of view, casting is only a delaying tactic instead of a definitive treatment method, for congenital scoliosis. As we documented an effective control of deformity during treatment period, the absence of out-cast radiographs during follow-ups did not affect the message given in this study. Although the average follow-up is short, curve correction has been maintained over the treatment period, and surgery was delayed for an average of 26.3 months. Eventually, all patients in this cohort will most likely need a definitive fusion procedure; however, the goal of serial casting in this patient population is to decrease the number of recurrent surgical procedures. Our results suggest that in patients with congenital scoliosis serial casting, using the derotation method is a safe and effective time-buying strategy to delay the need for surgical interventions in the short-term follow-up. However, longer follow-up and larger cohorts will be necessary to define the limits of cast treatment in congenital scoliosis. REFERENCES 1. Yazici M, Yilmaz G. Congenital scoliosis. In: Akbarnia B, Yazici M, Thompson GH, eds. The Growing Spine Management of Spinal Disorders in Young Children. Berlin: Springer; 2011: Richards BS, Sucato DJ, Johnston CE. Scoliosis. In: Herring JA, ed. Tachdjian s Pediatric Orthopaedics. Philadelphia, PA: Elsevier Saunders; 2014: Akbarnia BA, Breakwell LM, Marks DS, et al. Dual growing rod technique followed for three to eleven years until final fusion: the 48 r 2014 Lippincott Williams & Wilkins

7 J Pediatr Orthop Volume 35, Number 1, January 2015 Serial Derotational ing in Congenital Scoliosis effect of frequency of lengthening. Spine (Phila Pa 1976). 2008;33: Akbarnia BA, Campbell RM, Dimeglio A, et al. Fusionless procedures for the management of early-onset spine deformities in 2011: what do we know? J Child Orthop. 2011;5: Akbarnia BA, Cheung K, Noordeen H, et al. Next generation of growth-sparing techniques: preliminary clinical results of a magnetically controlled growing rod in 14 patients with early-onset scoliosis. Spine (Phila Pa 1976). 2013;38: Flynn JM, Emans JB, Smith JT, et al. VEPTR to treat nonsyndromic congenital scoliosis: a multicenter, mid-term followup study. J Pediatr Orthop. 2013;33: Karatas AF, Dede O, Rogers K, et al. Growth-sparing spinal instrumentation in skeletal dysplasia. Spine (Phila Pa 1976). 2013; 38: McCarthy RE, Luhmann S, Lenke L, et al. The shilla growth guidance technique for early-onset spinal deformities at 2-year follow-up: a preliminary report. J Pediatr Orthop. 2014;34: Tis JE, Karlin LI, Akbarnia BA, et al. Early onset scoliosis: modern treatment and results. J Pediatr Orthop. 2012;32: Elsebai HB, Yazici M, Thompson GH, et al. Safety and efficacy of growing rod technique for pediatric congenital spinal deformities. J Pediatr Orthop. 2011;31: Bess S, Akbarnia BA, Thompson GH, et al. Complications of growingrod treatment for early-onset scoliosis: analysis of one hundred and forty patients. J Bone Joint Surg Am. 2010;92: Mehta MH. Growth as a corrective force in the early treatment of progressive infantile scoliosis. J Bone Joint Surg Br. 2005;87: Sanders JO, D Astous J, Fitzgerald M, et al. Derotational casting for progressive infantile scoliosis. JPediatrOrthop. 2009;29: Baulesh DM, Huh J, Judkins T, et al. The role of serial casting in early-onset scoliosis (EOS). J Pediatr Orthop. 2012;32: Fletcher ND, McClung A, Rathjen KE, et al. Serial casting as a delay tactic in the treatment of moderate-to-severe early-onset scoliosis. J Pediatr Orthop. 2012;32: Waldron SR, Poe-Kochert C, Son-Hing JP, et al. Early onset scoliosis: the value of serial risser casts. J Pediatr Orthop. 2013;33: Johnston CE, McClung AM, Thompson GH, et al. Compairison of growing rod instrumentation versus serial cast treatment for earlyonset scoliosis. Spine Deformity. 2013;1: Cotrel Y, Morel G. The elongation-derotation-flexion technic in the correction of scoliosis. Rev Chir Orthop Reparatrice Appar Mot. 1964;50: McMaster MJ, Ohtsuka K. The natural history of congenital scoliosis. A study of two hundred and fifty-one patients. J Bone Joint Surg Am. 1982;64: Bergofsky EH, Turino GM, Fishman AP. Cardiorespiratory failure in kyphoscoliosis. Medicine (Baltimore). 1959;38: Nachemson A. A long term follow-up study of non-treated scoliosis. Acta Orthop Scand. 1968;39: Winter RB, Moe JH, MacEwen GD, et al. The Milwaukee brace in the nonoperative treatment of congenital scoliosis. Spine. 1976;1: Terek RM, Wehner J, Lubicky JP. Crankshaft phenomenon in congenital scoliosis: a preliminary report. J Pediatr Orthop. 1991; 11: Goldberg CJ, Gillic I, Connaughton O, et al. Respiratory function and cosmesis at maturity in infantile-onset scoliosis. Spine (Phila Pa 1976). 2003;28: Karol LA, Johnston C, Mladenov K, et al. Pulmonary function following early thoracic fusion in non-neuromuscular scoliosis. J Bone Joint Surg Am. 2008;90: Karol LA. Early definitive spinal fusion in young children: what we have learned. Clin Orthop Relat Res. 2011;469: Marks DS, Sayampanathan SR, Thompson AG, et al. Long-term results of convex epiphysiodesis for congenital scoliosis. Eur Spine J. 1995;4: Uzumcugil A, Cil A, Yazici M, et al. Convex growth arrest in the treatment of congenital spinal deformities, revisited. J Pediatr Orthop. 2004;24: Demirkiran G, Yilmaz G, Kaymaz B, et al. Safety and efficacy of instrumented convex growth arrest in treatment of congenital scoliosis. J Pediatr Orthop. 2014;34: Sankar WN, Acevedo DC, Skaggs DL. Comparison of complications among growing spinal implants. Spine (Phila Pa 1976). 2010;35: Risser JC. Scoliosis treated by cast correction and spine fusion. Clin Orthop Relat Res. 1976;116: Dimeglio A, Bonnel F, Canavese F. Normal growth of the spine and thorax. In: Akbarnia B, Yazici M, Thompson GH, eds. The Growing Spine Management of Spinal Disorders in Young Children. Berlin: Springer; 2011: Jenson RD, Shafer C, Stasic A, et al. EDF casting preserves pulmonary compliance in early onset scoliosis: a preliminary report, 6th International Congress on Early Onset Scoliosis and Growing Spine (ICEOS), November 15-16, 2012, Dublin, Ireland. J Child Orthop. 2012;6: r 2014 Lippincott Williams & Wilkins 49