Commentary In all of medicine, we strive to identify whether the treatments that we provide for our patients have a long-lasting effect or have any effect at all. If we could combine a randomized clinical trial with long-term follow-up, it would provide the ultimate test to gauge these results. The study by Diarbakerli et al. provides long-term data in a cross-sectional study to help to understand patient-reported outcomes following the current treatment options for these patients: observation, bracing, and surgery. Our treatment algorithms come from the natural history studies that have defined the risk of curve progression for both skeletally immature patients1 and skeletally mature patients2. These treatments are predicated on the idea that patients are better off when curves are small heading into adulthood or when they are corrected and are fused if they exceed a coronal curve threshold of 50°. The authors studied a large number of patients (1,187) with patient-reported outcomes analyzed at a long-term follow-up (averaging approximately 18 years). When reviewing this study, the following should be remembered. First, the radiographs used in the study were made during the patient’s last clinical evaluation (at the age of 20 years) prior to contacting the patient for the study (at the mean ages of 39 years, 36 years, and 40 years). Second, only coronal radiographs were made, without evaluation of the important lateral radiograph. Finally, the brace and surgical groups had treatments that were current at the time but not relevant today, especially as they involved the use of first-generation techniques and instrumentation. A deep dive into the data demonstrates some interesting findings related to treatment type, curve magnitude, and outcome of the surgical procedure. For those patients who avoided a surgical procedure, curve magnitude correlated inversely with outcome, with significantly better scores in all of the Scoliosis Research Society-22r (SRS-22r) domains and the EuroQol 5-Dimensions (EQ-5D) index when the Cobb angle was <30°. The finding in the brace-treated patients that the self-image scores were better for those who had recently undergone this treatment may be a reflection of their memories of the bracing experience. A surgical procedure seems to have a detrimental effect on outcome, with worse function and self-image when compared with the untreated group and the brace-treated group. Despite similar curve magnitudes in the 3 groups (24°, 31°, and 29°), the surgical patients had worse scores in these 2 domains, which most likely reflects the various effects of the surgical procedure including the stiffness imparted by the arthrodesis of the spine, the posterior scarring and stiffness of the soft tissues, and the presence and awareness of implants and a surgical incision in these patients. The EQ-5D domain scores highlight this, as the mobility and usual activities scores are lower in the surgical group. When focusing on the surgically treated patients, the authors demonstrate a relationship between remaining motion segments and functional scores, with worse scores seen in those patients fused below L2 (both SRS-22r function and pain scores and EQ-5D index). Although this continues to be an important finding, many would suggest that a balanced spine in both the coronal and sagittal planes is essential, and these parameters were not analyzed in this study. There are several strengths to this study: it involves a large series of patients (n = 1,187) with excellent return (65%) for long-term follow-up (age between 35 and 40 years), and there are good-quality patient-reported outcomes. These strengths lead me to be confident concluding that, first, keeping curve magnitude small (preferably below 30°) when possible with bracing provides for maximum long-term outcome, and, second, attempting to avoid a surgical procedure for non-progressive curves continues to be a good idea, and when a surgical procedure is performed, it is important to preserve lumbar motion segments. There are questions that remain and could not be completely answered with this study. Can modern brace treatment improve the results of treatment, maintaining smaller curve magnitudes and ultimately improving outcomes? Will the use of modern surgical techniques and implants improve on the results of the operative group seen in this study? With large-magnitude surgical correction, does the improvement in radiographic correction provide for improved patient scores? More specifically, will this improved correction close the gap between the patients undergoing a surgical procedure and those not undergoing a surgical procedure in the function and self-image domains? This study provides valuable information to the reader confirming the detrimental effects of scoliosis on patient-reported outcomes, the value of treatment to limit curve progression, and the effectiveness of surgical treatment, especially with preservation of motion segments. The authors should be congratulated on their outstanding effort to provide us with long-term, patient-focused outcomes and help to set the stage for future studies to determine whether modern treatment improves on these results.
Read full abstract