Abstract
The aim of the prospective, comparative radiographic analysis was to determine the role of the fulcrum-bending radiograph (FBR) for the assessment of the proximal thoracic (PT), main thoracic (MT), and the thoracolumbar/lumbar (TL/L) curves in patients undergoing posterior spinal pedicle screw fixation and fusion for adolescent idiopathic scoliosis (AIS). The FBR demonstrated statistically better correction than other preoperative methods for the assessment of frontal plane correction of the MT curves. The fulcrum-bending correction index (FBCI) has been considered a superior method than the correction rate for comparing curve correction undergoing posterior spinal fusion because it accounts for the curve flexibility. However, their applicability to assess the PT and TL/L curves in AIS patients remains speculative. The relation between FBR and correction obtained by pedicle screws fixation is still unknown. Thirty-eight consecutive AIS patients who underwent pedicle screw fixation and posterior fusion were included in this study. The assessment of preoperative radiographs included standing posterior–anterior (PA), FBR, supine side-bending, and postoperative standing PA and lateral plain radiographs. The flexibility of the curve, as well as the FBCI, was calculated for all patients. Postoperatively, radiographs were assessed at immediate (i.e. 1 week), 3-month, 6-month, 12-month, and 2-year follow-up. Cobb angles were obtained from the PT, MT, and TL/L curves. The study consisted of 9 PT, 37 MT, and 12 TL/L curves, with a mean age of 15.1 years. The mean FBR flexibility of the PT, MT, and the TL/L curves was 42.6, 61.1, and 66.2%, respectively. The mean operative correction rates in the PT, MT, and TL/L curves were 43.4, 69.3, and 73.9%, respectively, and the mean FBCI was 103.8, 117.0, and 114.8%, respectively. Fulcrum-bending flexibility was positively correlated with the operative correction rate in PT, MT, and TL/L curves. Although the correction rate in MT and TL/L curves was higher than PT curves, the FBCI in PT, MT, and TL/L curves was not significantly different (p < 0.05). The FBR can be used to assist in the assessment of PT, MT, and TL/L curve corrections in AIS patients. When curve flexibility is taken into account by FBR, the ability of pedicle screws to correct PT, MT, and TL/L curves is the same.
Keywords: Adolescent idiopathic scoliosis, Spine, Fusion, Radiograph, Flexibility, Fulcrum-bending radiograph
Introduction
Determination of curve flexibility in patients with adolescent idiopathic scoliosis (AIS) is an integral consideration in the preoperative planning. As such, various techniques have been developed to evaluate preoperative coronal curve flexibility [1–4]. One of which is the fulcrum-bending radiograph (FBR). The FBR has been shown to allow the identification of structural changes, assist in the selection of fusion levels, and predict the degree of correction [1, 5–7].
The FBR has been shown to favorably reflect the flexibility of the main thoracic (MT) curves in AIS patients in comparison to other techniques, such as the supine side-bending radiograph [8]. No significant difference was found between supine side-bending radiographs and the FBR for assessing TL/L curve flexibility, while supine left side-bending radiographs have demonstrated greater efficacy in flexibility assessment of PT curves [8]. The ability of the FBR in predicting the postoperative correction of curves by pedicle screws is currently unknown. Although this relationship was evaluated in a previous study [8], the corrective procedure included both anterior and posterior fusions, and the specific instrumentation utilized was not clarified. As such, the predictive capacity of curve flexibility upon postoperative curve correction of stand-alone posterior spine fusion with pedicle screw fixation and fusion remains speculative. Therefore, the purpose of this study was to assess the role of the FBR in the evaluation of PT, MT, and TL/L curves in AIS patients undergoing posterior fusion with pedicle screw fixation.
Materials and methods
Setting and patient population
From January 2004 to September 2006, 38 AIS patients undergoing posterior spinal fusion with pedicle screws at a single institution were included in this study for prospective assessment after an ethical approval was obtained. Inclusion criteria were as follows: individuals with AIS requiring an one-stage posterior procedure that has no previous additional flexibility-modifying surgery. According to the Lenke et al. [9] classification scheme, there were 18 patients with Type I curves, 8 patients with Type II curves, 6 patients with Type III curves, 1 patient with a Type IV curve, 1 patient with a type V curve, and 4 patients with Type VI curves. There were 10 patients with lumbar spine modifier A, 13 patients with B, and 15 patients with C. All patients received autologous bone graft harvested from the posterior iliac crest. The specific pedicle screw instrumentation system utilized was the CD Horizon M8 System (Medtronic Sofamor Danek, Minneapolis, MN, USA) in 15 cases and the MossMiami system (Depuy Spine, Raynham, MA, USA) in 23 cases. All surgical procedures were performed by one senior surgeon (ML). The fusion levels were chosen according to the recommendation by Lenke et al. [9, 10].
Radiographic and clinical assessment
Preoperative, immediate postoperative (i.e. the first week), and follow-up radiographs were obtained on long cassettes by certified radiology technicians in the standardized fashion. Side-bending radiographs were performed by asking patients to maximally bend while in the supine position. A supine side-bending radiograph was considered optimal when the ipsilateral rib touched the pelvis. The FBRs were obtained by suspending the patients in a lateral position over a radiolucent fulcrum that contained a radio-opaque marker which allowed radiographic identification of the level of placement, as previously described by Cheung and Luk [1]. The thoracic fulcrum was placed at the rib of the corresponding apex of the curve. In the lumbar spine, fulcrums were placed at the apex of the curves. The smallest-sized fulcrum that resulted in the shoulder (in thoracic curves) or pelvis (in lumbar curves) being lifted away from the table was the fulcrum used for obtaining radiographs. For proximal thoracic (PT) curves, the fulcrum was placed in the axilla, while the ipsilateral arm was placed below the patient’s head.
Cobb angles were manually measured on all the radiographs by one investigator using the standard technique [11]. An experienced spine surgeon reviewed the medical records and plain radiographs of all patients. Radiographic assessment included the use of pre- and postoperative PA and lateral standing and PA supine side-bending radiographs, in addition to preoperative FBR. Postoperative radiographic evaluation included immediate (i.e. 1 week), 3-month, 6-month, 12-month, and 2-year follow-up radiographs. Assessment of fusion was based on the radiographic evidence of instrumentation failure, segmental motion, or radiolucency at any level of the fusion construct on follow-up radiographs. Additional patient assessments included demographics, intra- and postoperative complications, as well as postoperative cosmetic balance. Radiographic shoulder height (RSH) was measured for those patients in whom PT and/or MT curves needed to be fixed. It was determined from the standing AP radiograph and defined as the graded height difference of soft tissue shadows directly superior to the acromioclavicular joints. This was graded as balanced (<1 cm, sid-to-side difference, grade 0), minimal imbalance (1–2 cm, grade 1), moderate imbalance (2–3 cm, grade 2), and significant imbalance (>3 cm, grade 3).
The correction rate, fulcrum-bending flexibility, and fulcrum-bending correction index (FBCI) were calculated as follows:
Formula 1
 |
Formula 2
 |
Formula 3
 |
Statistical analyses
Statistical analyses were performed utilizing SPSS statistical software vr. 17.0 (Chicago, IL, USA). Descriptive statistics were performed. Pearson’s correlation test was utilized for bivariate analyses. Paired t tests were utilized to assess differences between pre- and postoperative curves within group samples, whereas a single factor ANOVA was employed in comparing different groups. Correlation values of r were defined as follows: high correlation 0.80–1.00, marked degree of correlation 0.60–0.79, moderate degree of correlation 0.40–0.59, and less than 0.40 was noted as low to no degree of correlation [12]. To determine the best fit trend (e.g. linear, quadratic, cubic, etc.) for the correlation between the fulcrum-bending flexibility and the operative correction rate, the r2 and adjusted r2 values were assessed. Statistical significance was defined as p < 0.05.
Results
A total of 58 curves in 38 patients were considered structural according to the methods recommended by Lenke et al. [9], including 9 PT, 37 MT, and 12 thoracolumbar/lumbar (TL/L) curves. There were 32 females (84.2%) and 6 males (15.8%). The mean age at the time of surgery was 15.1 years (SD ±2.1 years; range 10.0–18.0 years).
The mean preoperative Cobb angles of the PT, MT, and TL/L curves were 44.9°, 54.6°, and 52.8°, respectively (Table 1). The mean FBR flexibilities of the PT, MT, and TL/L curves were 42.6, 61.1, and 66.2%, respectively. At immediate postoperative assessment, the operative correction rates of the PT, MT, and TL/L curves were 43.4, 69.3, and 73.9%, respectively. The FBCIs of the PT, MT, and TL/L curves were 103.8, 117.0, and 114.8%, respectively. There was no significant difference between FBR and postoperative Cobb angle in PT curves (p = 0.36); however, a statistically significant difference was found between FBR and postoperative Cobb angle in MT curves (p = 0.0001) and TL/L curves (p = 0.04). Correlation analysis indicated that the correction rate was positively correlated with the fulcrum-bending flexibility of the PT (r = 0.91, r2 = 0.84, p = 0.0006), MT (r = 0.65, r2 = 0.42, p < 0.001), and TL/L curves (r = 0.65, r2 = 0.42, p = 0.02), and that the correlation exhibited a quadratic trend with marked to high correlation (Fig. 1a–c). There was no statistically significant difference in correction rates between MT and TL/L curves (p > 0.05); however, such values were greater than that of PT curves. Moreover, there was no significant difference in FBCI between the PT, MT, and TL/L curves (p > 0.05).
Table 1.
Descriptive analysis of various radiographic parameters to curves based on preoperative, immediate postoperative, and 2-year follow-up assessment
| PT | MT | TL/L | |
|---|---|---|---|
| Preoperative | |||
| Standing PA Cobb angle (°) | 44.9 (9.3, 34.0–58.0) | 54.6 (9.0, 19.0–74.0) | 52.8 (10.7, 33.0–67.0) |
| FBR Cobb angle (°) | 25.7 (8.5, 14.0–40.0) | 21.4 (8.7, 6.0–42.0) | 18.3 (9.6, 3.0–38.0) |
| FBR flexibility (%) | 42.6 (17.2, 11.8–60.0) | 61.1 (13.6, 33.3–89.1) | 66.2 (14.5, 43.3–94.0) |
| Immediate postoperative | |||
| Standing PA Cobb angle (°) | 25.2 (8.6, 9.0–37.0) | 16.6 (6.5, 4.0–30.0) | 13.8 (6.0, 3.0–23.0) |
| Correction rate (%) | 43.4 (18.7, 14.7–75.7) | 69.3 (11.0, 48.3–92.0) | 73.9 (9.8, 57.7–94.0) |
| FBCI (%) | 103.8 (16.6, 83.3–127.3) | 117.0 (24.5, 87.2–219.0) | 114.8 (21.2, 87.8–153.6) |
| 2-year follow-up | |||
| Standing PA Cobb angle (°) | 26.7 (7.6, 13.0–38.0) | 17.8 (6.3, 5.0–29.0) | 15.3 (5.0, 8.0–23.0) |
| Correction rate (%) | 40.1 (16.1, 5.9–64.9) | 67.1 (10.9, 52.6–90.4) | 75.0 (10.0, 53.2–84.9) |
| FBCI (%) | 93.8 (24.9, 50.0–126.7) | 113.3 (21.8, 67.6–171.4) | 111.2 (30.8, 80.0–175.9) |
PT proximal thoracic curve, MT main thoracic curve, TL/L thoracolumbar/lumbar curve, PA posteroanterior, FBR fulcrum-bending radiograph, FBCI fulcrum-bending correction index
Fig. 1.
Correlation of the fulcrum-bending flexibility and the operative correction rate of a PT, b MT, and c TL/L curves
A statistically significant difference was found between preoperative and immediate postoperative coronal alignment to that of the PT, MT, and TL/L curves (p < 0.001). No statistical significant difference was found between immediate and 2-year follow-up postoperative Cobb angle (PT p = 0.273, MT p = 0.073, TL/L p = 0.212). Although there was some curve correction loss and the FBCI decreased on the 2-year follow-up, the difference between FBCI of immediate postop and FBCI of 2-year follow-up was not statistically significant in the PT (p = 0.364), MT (p = 0.084), and TL/L curves (p = 0.308). Left supine side-bending PT curves were significantly greater than fulcrum-bending PT curves (p < 0.05). Moreover, supine side-bending MT curves were less than fulcrum-bending MT curves (Fig. 2b, c), and there was no significant difference between side-bending TL/L curves and fulcrum-bending TL/L curves (p < 0.05). Although left supine side-bending radiographs revealed greater curve correction than that of the FBR in PT curves, the correlation between the FBR and the postoperative Cobb angle (r = 0.94, r2 = 0.88) was higher than that of left side-bending and the postoperative Cobb angle (r = 0.89, r2 = 0.79; p < 0.05).
Fig. 2.
a Standing coronal radiograph of a 12-year-old female with AIS and a main thoracic curve of 45° from T4 to T11. b Right supine side-bending radiograph of the patient suggested correction of the thoracic curve to 34°, and c the fulcrum-bending radiograph suggested correction of the thoracic curve to 10°. The fulcrum-bending flexibility was 77.8%. d, e Postoperative posterior–anterior and lateral radiograph illustrated curve correction to 10° in the coronal plane. The FBCI was 100%
No intra- or postoperative instrumentation-related complications were noted. On 2-year follow-up, fusion was achieved in all patients. Although there were five patients having shoulder imbalance greater than grade 1 preoperatively, no shoulder imbalance greater than grade 1 was noted after surgery. One patient reported minimal back pain, which resolved following 2 weeks of physiotherapy. A superficial wound infection occurred in two patients, both of which resolved by antibiotics and wound washout.
Discussion
Preoperative evaluation of curve flexibility in AIS patients is important for surgical planning and assessment of clinical and radiographic outcomes. Various methods have been reported to assess the curve flexibility, such as the traction, push-prone, and supine side-bending radiographs [13, 14]. Until now, supine side-bending has been considered the standard method for curve assessment and surgical decision-making [9, 15]. However, some authors have suggested that the correlation of preoperative supine side-bending radiographs and postoperative curve correction is suboptimal [16–18], especially as pedicle screw utilization has become increasingly popular. An additional disadvantage of supine side-bending imaging is that it is not standardized. Images are instead dependent on patient participation as well as the instruction and technique of the person obtaining the radiograph. For this reason, it is not ideally suited for all patients, such as patients with reduced mental capacity.
In an effort to assess the flexibility of the curve with a technique that is simple to perform and does not require active patient participation, Cheung and Luk [1] devised the FBR in order to assess accurate reducibility of structural curves. It has been shown that the FBR is a useful and reliable technique in the determination of curve correction in AIS patients undergoing surgical intervention. Luk and Cheung [1] also found that the FBR was predictive of the final correction obtained with the use of segmental hooks/rods instrumentation (TSRH, Medtronic Sofamor Danek, Minneapolis, MN, USA). Such an instrumentation system was able to take advantage of the spinal flexibility detected by the FBR and seldom corrected to a greater degree than the flexibility allowed. In 1998, Luk et al. [6] introduced the FBCI as part of the postoperative assessment of curve correction in cases of thoracic scoliosis. Since the FBCI took curve flexibility into account, it was considered a better method than the correction rate for comparing curve corrections with different instrumentation system. Cheung et al. [19] prospectively assessed the predictive value of the FBR in AIS curve correction in the context of alternate level pedicle screw fixation of the MT curve. A mean FBCI of 122% was obtained, and they draw the conclusion that pedicle screw constructs have a better ability to correct scoliosis compared to hooks and hybrid constructs. Other authors have also noted the utility of the FBR in the setting of AIS, primarily for the assessment of the MT curve [20, 21].
Klepps et al. [8] prospectively evaluated four different techniques utilized in preoperative assessment of curve flexibility in AIS patients. These techniques included lying supine, supine side-bending radiographs, FBR and push-prone radiographs. The authors noted that the FBR demonstrated a statistically better correction than other methods for MT curves. It was also suggested that there was no difference between side-bending and fulcrum-bending for TL/L curve flexibility evaluation in the same study. The left side-bending showed to be a more effective method for evaluating PT curves than the FBR. However, after comparing the preoperative methods and postoperative results, they draw the conclusion that all preoperative methods including fulcrum-bending fell short of accurately predicting the correction obtained by spinal fusion in MT and TL/L curves. In fact, they barely focused on the values obtained by surgery, failing to analyze the relation between curve flexibility revealed by the FBR and operative correction rate. In addition, there was no evaluation of curve correction on PT curves (as described in the present study). It was unknown whether the mobility suggested by side-bending was correlated to the degree of curve correction by instrumentation. Other shortcomings noted were the variability in both the approach utilized (both anterior and posterior procedures were included) and a lack of detail in the specific instrumentation used.
Our study suggests that although correction rate tends to be different between the PT, MT, and TL/L curves, the FBCI in all regions is not. Further analysis suggested that the operative correction rate has a positive correlation with the curve flexibility revealed by the FBR. There was no significant difference in curve flexibility revealed by fulcrum-bending between the MT and the TL/L curves, but the flexibility in those curves was higher than that of the PT curves. It showed that operative correction rate between the three groups of curves was a function of preoperative curve flexibility. However, the correction capability of pedicle screws in the PT, MT, and TL/L curves was similar under the same condition, which is additionally strengthened by the fact that all cases presented in this report were undergone pedicle screws fixation, and every case was addressed via a posterior approach only.
The finding that supine side-bending radiographs reveal greater mobility than FBR in PT curves was similarly suggested by Klepps et al. [8]. However, we have also found that the correlation between the FBR and postoperative curve correction (r = 0.94) was higher than that of the supine side-bending and postoperative curve correction (r = 0.89). The finding that the FBCI in the PT curve approximates 100% suggests that the pedicle screw/rod constructs had “utilized” all the flexibility revealed by the FBR. However, pedicle screw fixation provided greater curve correction than that noted by the FBR of the MT (FBCI = 117%) and TL/L (FBCI = 115%) curves.
The FBCI is influenced by several factors: the corrective power provided by instrumentation, surgeon intentions applied intraoperatively, and curve flexibility. Luk et al. [7] prospectively compared four kinds of instrumentation systems composed of hooks, including CD-Horizon (CD-H), Moss Miami (MM), TSRH, and ISOLA, showing that when curve flexibility is taken into account, despite differences in material and design of instruments, the ability of each system to correct thoracic scoliosis is the same. In our study, it was found that when curve flexibility is taken into account, the corrective capability of the PT, MT, and TL/L curves by pedicle screw fixation is not different. Luk et al. [7] also noted a negative correlation between the FBR flexibility and that of the postoperative FBCI in stiff curves (which they defined as a fulcrum-bending flexibility <50%), being that as the flexibility of curve decreased the FBCI increased. For flexible curves, instrumentation assumed the majority of the curve’s flexibility based on preoperative FBR assessment and the FBCI approximated 100%. Stiff curves, however, were associated with higher FBCI values. In reviewing the FBCI equation (above), it suggests that the applied correction forces for the more rigid curves were greater than the correction forces generated by the body’s weight during the FBR. It is also possible that surgeons tend to put more effort in correcting the more rigid curves. However, these increased correction efforts may be reduced in PT curves out of concern for causing a postoperative shoulder imbalance. So, in flexible curves, the FBR may act as an accurate predictor in the outcome of the surgery; however, in rigid curves, various factors may play a more significant role in curve correction.
One potential limitation of the current study is that the number of patients with PT and TL/L curves is smaller than the number of patients with MT curves. However, the major strength of this study is a prospective, self control study; all cases presented in this report were approached similarly. Specifically, pedicle screws were the only method of fixation in every case, and every case was addressed via a posterior approach only. Be that as it may, a larger, prospective study would be needed to further validate our findings.
Conclusion
In AIS patients, the FBR has the capacity for assisting the assessment of the PT, MT, and TL/L curves, which can help to predict operative curve correction upon the flexibility of curves. When curve flexibility is taken into account by FBR, the ability of pedicle screws to correct PT, MT, and TL/L curves is the same. Although supine left side-bending radiographs suggested greater preoperative curve flexibility than the FBR of the PT curves, the flexibility suggested by the FBR more closely approximates the postoperative outcome. Larger, prospective studies would be helpful in further verification of our findings.
Conflict of interest
The device(s)/drug(s) are FDA-approved or approved by a corresponding national agency for this indication. No sources of funding were received in relation to this work. The authors have no financial or competing interests in relation to this work. However, both Dr. Albert and Dr. Vaccaro are consultants and receive royalties on spinal products of Depuy Spine.
References
- 1.Cheung KM, Luk KD. Prediction of correction of scoliosis with use of the fulcrum bending radiograph. J Bone Joint Surg Am. 1997;79:1144–1150. doi: 10.2106/00004623-199708000-00005. [DOI] [PubMed] [Google Scholar]
- 2.Lenke LG, Bridwell KH, Baldus C, Blanke K. Preventing decompensation in King type II curves treated with Cotrel-Dubousset instrumentation. Strict guidelines for selective thoracic fusion. Spine. 1992;17:S274–S281. doi: 10.1097/00007632-199208001-00011. [DOI] [PubMed] [Google Scholar]
- 3.Large DF, Doig WG, Dickens DR, Torode IP, Cole WG. Surgical treatment of double major scoliosis. Improvement of the lumbar curve after fusion of the thoracic curve. J Bone Joint Surg Br. 1991;73:121–124. doi: 10.1302/0301-620X.73B1.1991744. [DOI] [PubMed] [Google Scholar]
- 4.Kleinman RG, Csongradi JJ, Rinksy LA, Bleck EE (1982) The radiographic assessment of spinal flexibility in scoliosis: a study of the efficacy of the prone push film. Clin Orthop Relat Res (162):47–53 [PubMed]
- 5.Luk KD, Don AS, Chong CS, Wong YW, Cheung KM. Selection of fusion levels in adolescent idiopathic scoliosis using fulcrum bending prediction: a prospective study. Spine. 2008;33:2192–2198. doi: 10.1097/BRS.0b013e31817bd86a. [DOI] [PubMed] [Google Scholar]
- 6.Luk KD, Cheung KM, Lu DS, Leong JC. Assessment of scoliosis correction in relation to flexibility using the fulcrum bending correction index. Spine. 1998;23:2303–2307. doi: 10.1097/00007632-199811010-00011. [DOI] [PubMed] [Google Scholar]
- 7.Luk KD, Lu DS, Cheung KM, Wong YW. A prospective comparison of the coronal deformity correction in thoracic scoliosis using four different instrumentations and the fulcrum-bending radiograph. Spine. 2004;29:560–563. doi: 10.1097/01.BRS.0000106494.14707.B2. [DOI] [PubMed] [Google Scholar]
- 8.Klepps SJ, Lenke LG, Bridwell KH, Bassett GS, Whorton J. Prospective comparison of flexibility radiographs in adolescent idiopathic scoliosis. Spine. 2001;26:E74–E79. doi: 10.1097/00007632-200103010-00002. [DOI] [PubMed] [Google Scholar]
- 9.Lenke LG, Betz RR, Harms J, Bridwell KH, Clements DH, Lowe TG, Blanke K. Adolescent idiopathic scoliosis: a new classification to determine extent of spinal arthrodesis. J Bone Joint Surg Am. 2001;83-A:1169–1181. [PubMed] [Google Scholar]
- 10.Lenke LG, Betz RR, Clements D, Merola A, Haher T, Lowe T, Newton P, Bridwell KH, Blanke K. Curve prevalence of a new classification of operative adolescent idiopathic scoliosis: does classification correlate with treatment? Spine. 2002;27:604–611. doi: 10.1097/00007632-200203150-00008. [DOI] [PubMed] [Google Scholar]
- 11.Cobb JR (1948) Outline for the study of scoliosis. In: Instructional course lectures for the American Academy of Orthopaedic Surgeons, vol 5. JW Edwards, Ann Arbor, pp 261–275
- 12.Franzblau A. A primer of statistics for non-statisticians. New York: Harcourt, Brace & World; 1958. [Google Scholar]
- 13.Vaughan JJ, Winter RB, Lonstein JE. Comparison of the use of supine bending and traction radiographs in the selection of the fusion area in adolescent idiopathic scoliosis. Spine. 1996;21:2469–2473. doi: 10.1097/00007632-199611010-00012. [DOI] [PubMed] [Google Scholar]
- 14.Vedantam R, Lenke LG, Bridwell KH, Linville DL. Comparison of push-prone and lateral-bending radiographs for predicting postoperative coronal alignment in thoracolumbar and lumbar scoliotic curves. Spine. 2000;25:76–81. doi: 10.1097/00007632-200001010-00014. [DOI] [PubMed] [Google Scholar]
- 15.Lenke LG, Kuklo TR, Ondra S, Polly DW., Jr Rationale behind the current state-of-the-art treatment of scoliosis (in the pedicle screw era) Spine. 2008;33:1051–1054. doi: 10.1097/BRS.0b013e31816f2865. [DOI] [PubMed] [Google Scholar]
- 16.Aronsson DD, Stokes IA, Ronchetti PJ, Richards BS. Surgical correction of vertebral axial rotation in adolescent idiopathic scoliosis: prediction by lateral bending films. J Spinal Disord. 1996;9:214–219. doi: 10.1097/00002517-199606000-00006. [DOI] [PubMed] [Google Scholar]
- 17.McCall RE, Bronson W. Criteria for selective fusion in idiopathic scoliosis using Cotrel-Dubousset instrumentation. J Pediatr Orthop. 1992;12:475–479. doi: 10.1097/01241398-199207000-00011. [DOI] [PubMed] [Google Scholar]
- 18.Winter RB, Lonstein JE, Denis F. How much correction is enough? Spine. 2007;32:2641–2643. doi: 10.1097/BRS.0b013e31815a5207. [DOI] [PubMed] [Google Scholar]
- 19.Cheung KM, Natarajan D, Samartzis D, Wong YW, Cheung WY, Luk KD. Predictability of the fulcrum bending radiograph in scoliosis correction with alternate-level pedicle screw fixation. J Bone Joint Surg Am. 2010;92:169–176. doi: 10.2106/JBJS.H.01831. [DOI] [PubMed] [Google Scholar]
- 20.Chan CY, Kwan MK, Saravanan S, Saw LB, Deepak AS. The assessment of immediate post-operative scoliosis correction using pedicle screw system by utilising the fulcrum bending technique. Med J Malays. 2007;62:33–35. [PubMed] [Google Scholar]
- 21.Hay D, Izatt MT, Adam CJ, Labrom RD, Askin GN. The use of fulcrum bending radiographs in anterior thoracic scoliosis correction: a consecutive series of 90 patients. Spine. 2008;33:999–1005. doi: 10.1097/BRS.0b013e31816c8343. [DOI] [PubMed] [Google Scholar]