Distal junctional kyphosis (DJK) can be a source of pain, imbalance, and poor cosmesis, and may create increased mechanical stress on adjacent segments, eventually resulting in adjacent segment degenerative disc disease. To date and to our knowledge, no peer reviewed published studies have attempted to quantify and compare the incidence of DJK in patients with adolescent idiopathic scoliosis (AIS) treated surgically with either a selective anterior or posterior instrumented fusion of the thoracic spine. Likewise, specific “risk factors” for the development of DJK in patients undergoing either selective anterior or posterior instrumentation and fusion for AIS have yet to be identified.
The first goal of this article was to determine the incidence of preoperative and postoperative DJK in a multicenter AIS population undergoing selective thoracic instrumented fusion via either an anterior or posterior approach. The second goal was to determine specific “risk factors” within each group for the development of DJK to make recommendations for minimizing the likelihood of developing postoperative DJK.
Methods and Materials
Erect 36-in standing radiographs of 375 patients with thoracic AIS from a multicenter database were reviewed retrospectively, before surgery and at 2-year follow-up. Of those patients, 238 underwent anterior thoracic single smooth-rod instrumentation and fusion, including structural interbody support at levels distal to T9, and 137 were treated with posterior double rod hybrid thoracic instrumentation and fusion. DJK was defined as the angle, ≥10°, between the superior endplate of the lowest instrumented vertebra and the inferior endplate of the adjacent distal vertebra. The radiographic parameters analyzed included regional sagittal and coronal Cobb measurements, levels instrumented, magnitude, C7 sagittal plumbline, and incidence of DJK ≥10°. Mean age at surgery was 14.4 years (range 9.1–20.9) in the anterior group and 14.7 years (range 10.2–20.7) in the posterior group. Mean preoperative curve was 53° (range 40°–85°) for all the patients who underwent the anterior procedure and 54° (range 40°–96°) for all those who underwent the posterior procedure. The demographics and Lenke classification for each of the groups are summarized in Table 1.1
The overall incidence of preoperative DJK ≥10° was 17 of 375 (4.5%). There were 10 patients (4.2%) in the anterior group of 238, including 1 patient with more than 19°, and 7 patients (5.0%) in the posterior group of 137, all less than 20°. The overall incidence of postoperative DJK was 37 of 375 (9.9%). Postoperative DJK developed in 17 patients (7.1%) in the anterior group and 20 patients (14.6%) in the posterior group. All coronal parameters, including instrumented and uninstrumented curve magnitude and flexibility, C7 plumbline, apical vertebral translation, and end instrumented vertebra translation and angulation, were similar between groups.
Of the parameters analyzed, a positive correlation was noted between the preoperative and postoperative T10–L2 sagittal measurement, the preoperative T5–T12 sagittal measurement, and the distal instrumented level relative to the distal Cobb level (Tables 2, 3). When postoperative DJK developed in the posterior group, mean postoperative T10–L2 was +17° kyphosis compared to +2° in the posterior group without DJK (P < 0.001). When postoperative DJK developed in the anterior group, mean postoperative T10–L2 was +12° kyphosis compared to +2° for the anterior group without DJK (P = 0.006). There were no significant differences in the preoperative T10–L2 region in patients with preoperative DJK in either group when compared with the posterior group without DJK. Case No. 1 preoperatively had no DJK, which developed postoperatively with a measurement of +21°, demonstrating the effect of instrumenting posteriorly to only Cobb +0 as opposed to Cobb +1 (Figure 1). The preoperative kyphosis in the anterior group that had DJK develop had a mean of +5° as opposed to 0° in the anterior group without DJK (P = 0.006). When DJK developed in the posterior group, there was a mean increase of +16° in T10–L2 as opposed to a +4° increase in kyphosis for the posterior group without DJK (P = 0.01).
The mean preoperative sagittal T5–T12 measurement of +31° correlated positively with patients in the posterior group who had postoperative DJK develop compared to the mean measurement of +23° for the posterior group without DJK (P = 0.02). The postoperative T5–T12 measurement in the posterior group that developed DJK was +25° compared to +20° for the group without DJK, which were significantly different (P = 0.05). There was no correlation between preoperative or postoperative sagittal T5–T12 measurements in the anterior group and the development of DJK.
All mean preoperative T2–T5 sagittal measurements for each group were in the normal range. In the T2–T5 sagittal region, the anterior group as a whole showed a slight decrease in kyphosis, while the posterior group became slightly more kyphotic. The C7 sagittal plumbline was not significantly different before or after surgery between patients with or without DJK in either the anterior or posterior instrumentation groups.
The distal instrumented level played a significant role both in the resolution of preoperative DJK and the development of DJK after surgery. In the patients with preoperative DJK that resolved after surgery, the distal instrumented levels were to Cobb +1 in the anterior group and Cobb +2 in the posterior group. In the anterior group that had postoperative DJK develop, instrumentation of the Cobb level was not significantly different than that of the anterior group without postoperative DJK (P = 0.97). However, the posterior group that had postoperative DJK develop was instrumented to Cobb +0 compared to Cobb +1 in the posterior group without postoperative DJK (P = 0.0001). Case No. 2 shows the development of postoperative DJK following anterior instrumentation extending only to Cobb −1, which was 2 levels short of the group in which DJK resolved (Figure 2).
With the development of segmental instrumentation systems, the sagittal plane has become an important area of focus in the surgical treatment of AIS. Bernhardt and Bridwell2 established regional and segmental sagittal values for the normal spine. DJK has been reported following posterior instrumentation and fusion in patients with AIS.3–5 DJK can be a source of pain, imbalance, and poor cosmesis, and may create increased mechanical stress on adjacent segments, eventually resulting in adjacent segment degenerative disc disease. Richards et al 4 noted DJK in 16 of 53 patients with AIS following posterior fusion with Cotrel-Dubousset instrumentation. However, they were unable to identify any risk factors for the development of DJK in their patients. McCance et al 5 in another series of 67 patients with AIS compared patients treated with posterior Harrington instrumentation to those treated with segmental instrumentation and noted DJK in only 1 patient. To our knowledge, there have not been any peer-reviewed articles dealing with DJK following anterior instrumentation of thoracic AIS curves.
We found the incidence of postoperative DJK to be twice as common following posterior instrumentation and fusion (14.6%) compared to (7.1%) in the instrumentation group. When preoperative DJK was present in either group, extending the instrumentation an additional level distally to include the kyphotic segment appeared to be important in avoiding persistent DJK after surgery. In the posterior group, it appeared important to maintain or achieve a normal sagittal profile over the main thoracic region (T5–T12) and thoracolumbar junction (T10–L2). Another “risk factor” for the development of DJK in the posterior group was ending the distal instrumentation at the Cobb level rather than 1 or 2 levels beyond the Cobb level.
In the anterior group, “risk factors” for the development of DJK included, as in the posterior group, not restoring or maintaining a normal sagittal T10–L2 region after surgery. This was probably less of a problem in the anterior group because of the use of structural interbody support by the authors to avoid kyphosis in the thoracolumbar region. The sagittal T5–T12 region was within the normal range before and after surgery, and was not a “risk factor” for the development of DJK when anterior instrumentation was used. An unexpected finding in the series was the statistically significant postoperative increase in T2–T5 kyphosis in the posterior group and decrease in the anterior group. These differences were not associated with DJK and were believed to be a compensatory mechanism for maintaining global sagittal balance, which did not change after surgery in either group.
The incidence of preoperative DJK for the entire group was 4.5%. The incidence of postoperative DJK was 7.1% for the anterior group and 14.6% for the posterior group. It appears that careful attention to the sagittal profile of the spine, achieving normal main thoracic and thoracolumbar regions, including the junctional levels when preoperative DJK exists, and avoiding shorter than normal distal instrumentation levels will decrease the likelihood of postoperative DJK.
- When DJK is present preoperatively, the lower end of posterior instrumentation should include the junctional level to avoid postoperative DJK.
- Preoperative thoracolumbar kyphosis is a risk factor for development of postoperative DJK.
- The incidence of postoperative DJK following posterior instrumentation was 14.7% and following anterior instrumentation was 7.5%.
1. Lenke LG, Harms J, Bridwell KH, et al. Adolescent idiopathic scoliosis: A new classification to determine extent of spinal arthrodesis. J Bone Joint Surg
2. Bernhardt M, Bridwell KH. Segmental analysis of the sagittal plane alignment of the normal thoracic and lumbar spines and thoracolumbar junction. Spine
3. Massey TB, Winter RB, et al. Selection of fusion levels with special reference to coronal and sagittal balance in right thoracic idiopathic scoliosis using Cotrel-Dubousset instrumentation. Paper presented at: SRS 25th Annual Meeting; 1990; Honolulu, HI.
4. Richards BS, Birch JG, Herring JA, et al. Frontal plane and sagittal plane balance following Cotrel-Dubousset instrumentation for idiopathic scoliosis. Spine
5. McCance SE, Denis F, Lonstein DF, et al. Coronal and sagittal balance in surgically treated adolescent idiopathic scoliosis with the King II curve pattern. A review of 67 consecutive cases having selective thoracic arthrodesis. Spine