Does Thoracic Kyphosis Have Any Importance in Selective Versus Nonselective Fusion Preference in Patients With Lenke Type 5C Adolescent Idiopathic Scoliosis? A Contribution to the Guidance of the Lenke Classi cation

AIM
This study aimed to investigate the importance of thoracic kyphosis (TK) for treatment preference in patients with Lenke Type 5C adolescent idiopathic scoliosis by comparing radiological outcomes of the patients who underwent selective fusion (SF) or nonselective fusion (NSF).


MATERIAL AND METHODS
Twenty-nine patients with Lenke Type 5C AIS were included and then divided into two groups as per the fusion procedure used in the surgical treatment. SF group including 16 patients (14 female patients; mean age = 15.56 yr; age range, 14-18) with normal TK and NSF group including 13 patients (nine female patients; mean age = 15.54 yr, age range, 13-18) with thoracic hyperkyphosis. Thoracolumbar/lumbar (TL/L) Cobb, thoracic (T) Cobb, TK and lumbar lordosis (LL), pelvic incidence (PI), pelvic tilt (PT), and sacral slope (SS) were measured on standing spine radiographs preoperatively and at the final follow-up. The correction rates (CRs) of Cobb angles and the difference in each other radiological parameters were calculated.


RESULTS
No significant differences were observed in the mean CRs of T/L Cobb and T Cobb angles, PI, SS, and PT (p = 0.313, p = 0.444, p = 0.51, p = 0.472, and p = 0.14, respectively). However, significant differences were observed in the mean TK angle, which was -2.13° ± 13.52° (range, 29-27°) in SF group and 28.46° ± 15.05° (range, -4°-°47°) in NSF group (p = 0.001), and LL angle was 0.88° ± 14.23° (range, -21°-32°) in SF group and 11.54° ± 17.79° (range, -31°-34°) in NSF group (p = 0.016).


CONCLUSION
In patients in whom Lenke\'s sagittal modifier is N, SF can be performed efficiently. NSF can be preferred for those with (+) Lenke\'s sagittal modifiers as it provides better TK control.


Abstract
Background Although Lenke classi cation analyses the sagittal plane as (+), N, and (-), it does not consider it in the choice of treatment, and it has limitations with overall thoracic kyphosis (TK). To investigate the importance of TK for treatment preference in patients with Lenke 5C adolescent idiopathic scoliosis (AIS) by comparing radiological outcomes of the patients who underwent selective fusion (SF) or nonselective fusion (NSF).

Conclusion
In patients in whom Lenke's sagittal modi er is N, SF can be performed e ciently. NSF should be preferred in those with Lenke's sagittal modi ers (+) as TK can be better controlled with NSF.

Level of Evidence: 3 Background
The main goals of corrective surgery in adolescent idiopathic scoliosis (AIS) are to obtain an optimally corrected and well-balanced spine, to prevent curve progression, and to provide maximum functionality of the spine with minimal fused motion segments [1,2]. Considering these goals, the concept of selective fusion (SF), in which the structural curve is fused while sparing the nonstructural curves to preserve the mobility of the spine, has gained popularity among spine surgeons for the past few decades [2][3][4][5][6][7] [9]. Unlike King-Moe classi cation, the Lenke classi cation, which evaluates not only the coronal plane but also the sagittal plane, is the most widely used classi cation today ( According to the Lenke classi cation, type 5 represents the third most prevalent curve type of AIS, and this curve type is characterized by a single structural thoracolumbar/lumbar (TL/L) curve with non-structural or compensatory thoracic (T) curve ( Table 1). The Lenke classi cation suggests a selective TL/L fusion for all of the type 5 curves [8]. On the other hand, remarkable number of experienced scoliosis surgeons prefer considerably non-selective fusion (NSF) for the type 5 curves [11,12]. This often leads to the questioning of the treatment recommendations of the Lenke Classi cation [7,10]. Although Lenke classi cation analyses the sagittal plane as (+), N, and (-), it does not consider it in the choice of treatment, and it has limitations with overall TK. This situation can be a dilemma for the spine surgeon who refers to the Lenke Classi cation in the choice of treatment, especially in the treatment of AIS patients with advanced sagittal plane deformity.
The aim of this study is to investigate the importance of thoracic kyphosis (TK) for the treatment choice in Lenke type 5C patients by comparing the radiological outcomes of the patients who underwent SF or NSF with respect to this criterion. We hypothesize that NSF provides better correction rates of the TK and lumbar lordosis (LL) and has negligible superiority on restoring pelvic parameters in cases with (+) Lenke's sagittal modi er whereas SF improves T/TL Cobb and T Cobb angles as effective as NSF in cases with Lenke's sagittal modi er is N.

Methods
This retrospective study was conducted on patients who were diagnosed and operated for Lenke type 5C AIS by a single spine surgeon between 1998 and 2009 in a single tertiary referral center. Inclusion criteria for the study were: (1) a diagnosis of Lenke type 5C AIS with thoracic normal or hyperkyphosis; (2) surgical treatment with posterior pedicle screw instrumentation; (3) no history of previous spinal surgery; (4) complete sets of preoperative and nal follow-up standing full-length anteroposterior (AP) and lateral radiographs of the spine; and (5) being willing to participate in the study. Exclusion criteria were: (1) a diagnosis of Lenke type 5C AIS with thoracic hypokyphosis; (2) a history of previous spinal surgery (hybrid or anterior pedicle screw instrumentation, or corrective osteotomies); (3) lost to follow-up; (4) inadequate radiographic imaging; (5) concomitant neuromuscular or congenital disorders; and (6) being unwilling to participate in the study.
A total of 40 patients were evaluated based on the above eligibility criteria. After excluding 8 patients, remaining 32 with Lenke type 5C AIS who met the inclusion criteria were included in the study and invited to a nal follow-up examination for radiographic assessment. Patients included in the study were categorized into two groups based on the fusion procedure used in the surgical treatment and thoracic kyphosis angle: SF group including patients with normal thoracic kyphosis and NSF group including patients with thoracic hyper-kyphosis ( Fig. 1). Normal thoracic kyphosis is de ned as 10° to 40° of thoracic convexity, and thoracic hyper-kyphosis as 40° or higher [8].

Surgical Technique
All surgical procedures were performed by the same senior surgeon. All patients underwent general anesthesia and placed in a prone position on a surgical table. After a posterior midline incision was made, subperiosteal paraspinal muscles were dissected to expose the posterior elements of the spinal fusion levels. The pedicle screws were inserted by a free-hand technique and checked with intraoperative uoroscopy [13]. First, a lordotic shape titanium rod was placed at the convex side of curvature to obtain lordosis and correct the coronal deformity. The concave rod was given less lordosis than the rst rod.
Curve correction was achieved using the rod-rotation maneuver with convex rod, followed by slight convex compression and concave distraction. After decortication of the posterior elements and facet excision autogenous and allogenic bone grafts were used for fusion.
All radiographic measurements were performed on standing anteroposterior and lateral radiographs of the entire spine, by a single attending surgeon, who did not participate in the treatment of the patients, using a validated software (Surgimap™, Nemaris Inc, New York), preoperative and at the nal follow-up. TK angle was measured from the superior endplate of T5 vertebra to the inferior endplate of T12 vertebra.
LL angle was measured from the superior endplate of L1 vertebra to the superior endplate of S1 vertebra.
The correction rate of each radiographic parameter in both groups was calculated as following equation: (preoperative X angle -nal follow-up X angle)/preoperative X angle in the standing lm × 100.

Statistical Analysis
IBM SPSS Statistics software, version 22.0 (IBM Corp., Armonk, New York, NY, USA), was used for statistical analysis. A p < 0.05 was considered statistically signi cant. Normality tests were conducted using the Kolmogorov-Smirnov test. Between-group comparisons were performed using the Student t-test for parametric variables, Mann-Whitney U test for non-parametric variables, and Fisher's exact test for categorical variables. Statistical signi cance was set at p < 0.05.

Discussion
In the surgical treatment of Lenke type 5 AIS, SF of the structural TL/L curve has been considered to be the leading treatment method [5,8,14]. Many studies have reported satisfactory radiological and clinical outcomes, and also spontaneous T curve correction with SF. However, Lenke type 5 AIS is unique, and it differentiates from other types as the T kyphosis cannot be controlled by only fusing the structural TL curve. Although Lenke classi cation identi es the T kyphosis as (+), N, and (-), it made no recommendation regarding kyphosis. Contrary to the recommendations of the Lenke Classi cation, some spine surgeons have been reported to perform NSF in 27% of patients with Lenke type 5 AIS [11]. The main reason for performing NSF has been to control the T coronal plane deformity [11]. However, it has been stated that maintaining sagittal balance is crucial for favorable radiological and clinical outcomes, and it should not be neglected in AIS [12,15]. T5-T12 T kyphosis and T1-T4 sagittal alignment were determined as the criterion to be considered in achieving sagittal balance [12,16]. In a study by Connolly et al., sagittal plane parameters have been stated to be more substantial in the long-term health of the spine [17]. Consistent with this statement, in another study by Takayama et al. [18], it was reported that patients with low functional scores were the ones whose sagittal balances could not have been restored.
Considering that the importance of the sagittal plane in the treatment of AIS has been supported by current publications [6], another reason to extending the fusion to T spine may be to control and restore the T kyphosis [6]. Accordingly, it was planned to investigate effects of T kyphosis on the treatment choice and radiological outcomes in Lenke type 5C patients in the current study. Moreover, the importance of the correction of TL kyphosis and T hypokyphosis was also emphasized in a study by Suk et al. [16] In our study, hypokyphotic patients were not included in the study due to underpowering, thus only normokyphotic and hyperkyphotic patients were compared.
In a study by Lark et al., 58 Lenke Type 5 patients underwent SF or NSF, and a signi cant difference was reported in both postoperative TL/L Cobb and T Cobb angles in the matched groups. (The mean TL/L Cobb angle was 19° ± 6° in SF group and it was 13° ± 6° in NSF group, p < 0.001; mean T Cobb was 22° ± 9° in SF and it was 12° ± 6° in NSF, p < 0.001). In our study, there was no signi cant difference between the groups in terms of postoperative TL/L Cobb and T Cobb angles. (The mean TL/L Cobb angle was 10.24° ± 9.97° in SF and 10,93° ± 8.12° in NSF group [p = 0.718]; the mean T Cobb angle was 5.35°± 4.52°i n SF group and 6,80° ± 5.34° in NSF group [p = 0.198]). In that study, TK increased in the group SF and it decreased in the group NSF postoperatively. In parallel, TK increased in the group SF (5,2%) and decreased in the group NSF (56,4%) in our study (p < 0.001). Contrary to reported postoperative hypokphosis (mean: 18° ± 6°) by Lark et al. in the group NSF, the patients in both groups in our study had normokyphosis postoperatively. The reason for this difference may be that patients who underwent NSF were hyperkyphotic ones in our study, whereas they were normokyphotic ones in that study. Another reason for that may be the longer time required for the normalization of the sagittal pro le, as stated in a meta-analysis by Pasha et al. [19]. Unlike that study, the other spinopelvic parameters were also evaluated in our study. Coherent with our ndings, in the preoperative evaluation of Lenke type 5 patients, the mean values of spinopelvic parameters reported by Farshad et al. [20] were as 48° ± 13°, 36° ± 9°, 12°± 7°, 50° ± 12° for the PI, SS, PT, LL; respectively. As reported in the literature, increased sacral slop was present in our patient series as well to probably compensate for increased LL (Table 4) [20][21][22]. In our study, PI did not change similar to previously reported [20][21][22], yet SS decreased while PT increased signi cantly postoperatively (p = 0.434, p = 0.037, p = 0.001) in both groups. However, the observed difference between the groups was not statistically signi cant.
Lonner et al. reported that the frequency of complications has been increasing related to AIS surgery [19].
In another study, it was reported that post junctional kyphosis (PJK) developed in 28% of patients with AIS. In a study in which PJK was reported as 8.5% in Lenke type 5 AIS, hyperkyphosis was de ned as the main risk factor. Also, Wang et al. reported that PJK was frequently seen in short-segment instrumentation [6]. In our study, it has been demonstrated that SF can be performed in patients with Lenke type 5 and T sagittal pro le can be restored better with NSF in patients with TK. Contrary to those reported in the literature, the reason for developing PJK in none of the 32 patients in our series with a mean of 42,3 months follow-up might be due to the consideration of the sagittal plane analysis in the preference of surgical treatment method.
In our retrospective case series, the patients with Lenke type 5 AIS who underwent SF or NSF were compared radiologically. When the Cobb angles are evaluated in both groups, the mean T curves angles was not high, so they can be considered as non-structurally. Whereas SF can be performed for both groups according to the Lenke Classi cation, the surgeon included the T region into the fusion area in those who have high TK based on his own experience. While the sagittal modi er of Lenke remains N in patients undergoing SF, it changed from (+) to N in most of the patients who underwent NSF. These ndings show us that the sagittal plane evaluation of the Lenke Classi cation System may be insu cient to guide the treatment. In this study, in the mid-to-long term follow-up, it has been shown that SF can be performed for the patients with Lenke Type 5 AIS, additionally the sagittal plane is restored better with NSF in patients with TK.
To the best of the authors' knowledge, this study is the rst study to demonstrate that TK, which the Lenke Classi cation does not consider in the treatment recommendation, can be an important determinant in the choice of SF versus NSF in patients with Lenke 5C AIS by evaluating TL/L, T Cobb angles, TK, LL as well as PI, SS, PT. It must be noted that the ndings of this study should be supported by prospective randomized controlled trials involving a larger number of patients.
This study was retrospective in nature and it contains similar de ciencies with other retrospective studies. First, the outcomes of SF and NSF treatments were evaluated only radiologically. The sample size was relatively small. The decision to perform a SF versus NSF was based on TK and there was no control group. However, this potential study bias was minimized by the preoperative similarity in terms of age, gender, follow-up duration, and preop coronal and sagittal plane parameters except for T5-T12 TK of both groups being compared. The patients in the study were homogeneous and they underwent posterior instrumentation and fusion by the same surgeon with the same pedicle screw instrumentation system. Also, a considerable length of follow-up duration a mean of 41,7 months is another strength of the study.

Conclusion
The ndings of this study support that it is important to consider TK in the preference of NSF versus SF in patients with Lenke Type 5 AIS. In cases in which Lenke's sagittal modi er is N, SF can be performed e ciently. NSF should be preferred in those with Lenke's sagittal modi ers (+) since TK can be better controlled with NSF.

Consent to participate
Written informed consent was obtained from all the participants. Preoperative (a and b) and 37 month-follow-up radiographs (c and d) of a 25-year -old female patient with a 54° TL/L Cobb angle, 36° T Cobb angle, and 50° T hyper-kyphosis were corrected by non-selective fusion procedure. TL/L Cobb, T Cobb and T kyphosis were 1°, 9°, and 2°; respectively.