Complications of Open Anterior Instrumentation in Adolescent Idiopathic Scoliosis

Michael W. Peelle

Lawrence G. Lenke

HISTORICAL PERSPECTIVE

The anterior approach to the thoracic and thoracolumbar spine has been broadly utilized since its initial development for the treatment of spinal tuberculosis (1,2,3,4). Twenty-five years ago, Dwyer and Schafer published the first results of segmental anterior spinal instrumentation for the correction of scoliosis using flexible cables (5). Modifications by Zielke advanced the use of segmental instrumentation using a single screw and semirigid, threaded rod designed to allow rod derotation maneuvers (6,7,8,9). Problems with implant failure led to suboptimal clinical outcomes; therefore, more rigid anterior systems have become widely implemented in the last decade (8,9,10,11,12,13). This chapter will review the complications of open anterior scoliosis corrections with the emphasis on instrumentation and fusion techniques.

Indications for Anterior Surgery

The most critical step in avoidance of complications when performing an anterior spinal instrumentation and fusion in a patient with adolescent idiopathic scoliosis is proper patient selection. In our practice, we limit anterior-only surgery to those scoliosis patients with Lenke type I and type V curves (13,14). Although several authors have expanded the indications for anterior surgery to include Lenke type II, III, and VI curves (15), our institutional bias favors posterior instrumentation in these patients, given the excellent results afforded by segmental pedicle screw constructs. A second indication for anterior surgery is the first stage in a two-stage approach to large, inflexible curves (15). Again, as the magnitude of curve correction and implant strength with pedicle screw constructs continues to improve, the need for anterior releases and anterior instrumentation continues to decline, particularly since the avoidance of crankshaft phenomenon appears possible by posterior rigid screw fixation (17,18,19). Considerable debate continues concerning the role of anterior versus posterior surgery; however, certain merits of anterior instrumentation will continue its utilization by scoliosis surgeons (12,15,20). The emerging role of thoracoscopic procedures continues to redefine the role of anterior surgery (16), as does renewed ideas of scoliosis correction including anterior convex disc staples. In short, the familiarity of anterior procedures continues its importance for the modern scoliosis surgeon.

PULMONARY COMPLICATIONS

One major concern of an anterior approach to the spine is both the short- and long-term effects on pulmonary function. In fact, pulmonary complications typically account for up to 50% of the morbidity of anterior approaches to the thoracic spine (21,22). Although complication rates for adolescent idiopathic scoliosis (AIS) patients are typically lower than neuromuscular scoliosis, congenital scoliosis, or adult scoliosis patients, careful preoperative planning and postoperative monitoring remains essential in this patient population. O’Brien and Jones compiled data from several series regarding complication rates after anterior surgery in adults and adolescents (21). Minor pulmonary complications following open anterior thoracic spine correction included pleural effusions (2.7%), atelectasis (4.7%), pneumothorax (2.2%), pneumonia (2.5%), and other pulmonary difficulties (2.2%). Major pulmonary complications requiring intubation included pneumothorax (0.17%), hemothorax (0.33%), and pneumonia (0.17%). Overall, pulmonary issues accounted for 4.9% of major complications.

The study by Grossfield et al., an extensive review (599 anterior procedures) of pediatric patients, reported a major complication rate of 7.5% and minor complication rate of 32.6% (22). They identified two important risk factors predisposing patients to a higher complication rate with regards to pulmonary function. Curves of large magnitude— in particular, curves greater than 100 degrees—had more than double the rate of minor and major complications. Preexisting pulmonary disease, as evidenced by pulmonary function tests less than 40% of predicted values, also increased complication rates by 50% or more. Interestingly, the detachment of the diaphragm did not directly influence overall complication rates in their assessment. Despite the prevalence of pulmonary complications, the routine use of preoperative pulmonary function tests is not necessarily recommended since the majority of AIS patients will have normal studies (23,24). However, patients with large curves (>70 degrees) and/or medical comorbities—such as asthma, pneumonia, or a history of respiratory distress requiring hospitalization—should be evaluated if anterior surgery is considered.

A series of publications from our institution evaluated the long-term effects of anterior surgery via prospective measurements of pulmonary function tests (PFTs) in four groups of AIS patients: two groups undergoing posterior spinal fusion (PSF; with or without thoracotomy/thoracoplasty) and two groups undergoing anterior spinal fusion (ASF; with or without thoracotomy/thoracoplasty) (25,26,27,28). At minimum 5-year follow-up, only the PSF without thoracotomy/thoracoplasty group had a significant increase in absolute forced vital capacity (FVC) and forced expiratory volume in 1 second (FEV-1), whereas the three other groups experienced a significant decrease in percent predicted values for the same measures (25). The conclusions from these studies cannot be understated; any chest cage disruption for surgical correction of AIS may result in suboptimal long-term pulmonary function recovery. Some authors tout the benefit of thoracoscopic approaches to preserve chest wall anatomy and subsequent pulmonary function; however, the long-term outcomes are not proven (16,29). Lenke et al. examined 39 AIS patients regarding the effects of open versus thoracoscopic anterior procedures when combined with a posterior fusion. Both groups had modest improvement in pulmonary function tests at two-year follow-up; however, no statistical difference was detected between the two groups (29).

VASCULAR COMPLICATIONS

As the use of thoracic instrumentation becomes more widespread, the proximity of thoracic screws to the aorta has gained recent attention. A comparison of patients with AIS versus a straight spine noted that the aorta is positioned more posteriorly and laterally with increasing Cobb angle of the thoracic curve (30). They reported an increase in the average distance from the aorta wall to vertebral body at the curve apex (2 to 4 mm). Sucato et al. utilized computed tomography (CT) examination of 14 dextroscoliotic adolescents treated with single-rod, thoracoscopic instrumentation to assess a total of 106 screws. The average screw-to-spinal canal distance was 5.3 mm (1.2 to 11.4 mm) (31). Thirteen screws (12.3%) were noted to deform the aorta and 15 (14.2%) were immediately adjacent to the aorta wall, but no patients were noted to have neurologic or vascular compromise. A useful anatomic reminder supported by their study is the fact that thoracic vertebral bodies increase in width an average of 1.2 mm per level (24 mm at T4 and 34 mm at T12) and increase in depth from 17.7 mm at T4 to 25.5 mm at T12.

Although the long-term outcomes of implants adjacent to vascular structures are not known, the incidence of perioperative complications remains extremely low, with only several case reports depicting catastrophic events (32). For an open anterior approach, our experience dictates that major vessel damage can best be avoided by thorough release of the vasculature off the spine by blunt dissection and contralateral screw tip palpation to ensure optimal screw lengths (33).

SINGLE-ROD INSTRUMENTATION

Early reports of anterior fusions with modified Zielke instrumentation yielded variable results. Using Harms-MOSS instrumentation (flexible threaded rods), Betz et al. compared 78 ASF to 100 PSF patients with multisegmented hook instrumentation (34). Twenty of 78 patients (26%) had a coronal Cobb correction loss of greater than 10 degrees, with an average loss of 15 degrees. Implant breakage
occurred in 24 of 78 patients (31%). Although the majority (75%) remained asymptomatic and successfully fused, eight patients required reoperations for implant-related complications, including proximal screw pull-out and four pseudarthroses (Fig. 10.1). Of particular concern is the noted thoracic (T5-T12) hyperkyphosis of more than 40 degrees in 40% of patients whose preoperative kyphosis was greater than 20 degrees.

Only gold members can continue reading. Log In or Register to continue