Change in sagittal profile after implantation of anchored interbody cage in the surgical procedure for degenerative cervical spine disease

Background. The aim of this study was to verify the relationship between changes in the segmental sagittal profile (SSP) and changes in the global sagittal profile (GSP) after anterior cervical discectomy with anchored cage implantation (ACDF). Study design. Prospective study with 2-year follow-up. Methods. This study includes 104 patients after 1-level or 2-level ACDF operated between the May 2013 and March 2016. SSP was evaluated by Cobb angle measurement of operated motion segment (CobbS) and GSP was evaluated by Cobb angle measurement in C2-C7 segments (CobbG). Both SSP and GSP were measured preand postoperatively within a 24 months follow-up period. The influence of factors such as age, gender, number of treated segments and osteoporosis was evaluated using t-tests. The correlation between SSP and GSP changes was assessed by Pearson's correlation coefficient. Results. In the early postoperative period after 1-level ACDF, there was a significantly greater increase in CobbS compared to that of the 2-level ACDF (P=0.0149). Male patients experienced a significant decrease of CobbG during the first 6 months after surgery as well as patients with osteoporosis within 12 months after ACDF. After ACDF the SSP change weakly correlated with the GSP change. Conclusion. SSP change after 1or 2-level ACDF correlates mostly weakly with GSP change. Male gender and osteoporosis were identified as risk factors for global lordotisation following ACDF.


INTRODUCTION
Anterior cervical discectomy and fusion (ACDF) is one of the most common surgical procedures in the treatment of the degenerative cervical spine disease. The aim is to decompress the spinal canal and spinal nerves located in the degenerated motion segment. Implantation of a cervical interbody cage allows for restoration of the intervertebral space height, induction of interbody fusion and lordotisation of the operated motion segment. However, the intraoperative lordotisation of the motion segment may subsequently regress. The effect of the segmental sagittal profile correction (SSP) on the development of the global sagittal profile (GSP) of the cervical spine is variable and influenced by various factors.

MATERIALS AND METHODS
Our prospective study includes 104 patients with degenerative cervical spine disease. All of these patients underwent a 1-or 2-level ACDF with an implantation of Zero Profile Variable Angle ® cage (Zero-P VA ® , DePuy Synthes, Switzerland, Fig. 1.). ACDF was indicated based on clinical and graphical findings after 6 weeks of un-satisfactory conservative treatment. In all cases, motion segments with graphical signs of osteochondrosis were indicated for treatment. The exclusion criteria for this study were defined as: presence of myelopathy, pregnancy, inflammatory or malignant disease of the cervical spine, injuries of the cervical spine and general contraindications of elective surgery. According to age they were divided into two groups -patients younger than 55 years of age and patients aged 55 years and more. Similarly, our cohort was divided according to sex, number of operated cervical motion segments and osteoporosis (which was confirmed by means of bone densitometry). All patients underwent a cervical spine X-ray examination in anterior and lateral projections before surgery, in early postoperative period (up to 48 h after ACDF), at 6 weeks, 3 months, 6 months, 12 months and 24 months after ACDF. The SSP was evaluated by means of measurement of the Cobb angle between the upper body surface of proximal and the lower body surface of the distal vertebra of the motion segment (CobbS). The GSP was similarly evaluated using the measurement of the Cobb angle between the lower vertebral body of C2 and the lower vertebral body of C7 (CobbG) or the most distally visualized vertebra above the C7, but always in the same manner for a particular patient (Fig. 2.). Early postoperative changes in sagittal profile were then evaluated as the difference between CobbS (CobbG) values in the preoperative and early postoperative period (up to 48 h). Differences in SSP and GSP during the further postoperative period were evaluated as the difference between CobbS (CobbG) values in the early postoperative period and CobbS (CobbG) values over 6 weeks, 3 months, 6 months, 12 months and 24 months after ACDF. Statistical significance of factors such as age, gender, number of operated motion segments and osteoporosis was evaluated using Student's unpaired t-test. The impact of the cage subsidence (CS) on CobbS value in term of 24 months after ACDF was evaluated using the same test. Correlation between SPP and GSP changes was assessed by means of the Pearson's correlation coefficient (r). The Cohen scheme was used to interpret Pearson correlation coefficient -absolute values up to 0.1 represented a trivial correlation, values between 0.1 -0.3 represented a weak correlation, values 0.3 -0.5 represented a moderate correlation and values above 0.5 represented a strong correlation 1 .

RESULTS
We performed ACDF with the implantation of the anchored Zero-P VA ® cage in an overall of 159 motion segments. In 49 patients it was a 1-level ACDF and in 55 it was a 2-level ACDF. The level C5/6 was treated most frequently (Table 1). Table 2 depicts the division of the cohort according to age, gender, number of treated mo-  tion segments and presence of densitometrically verified osteoporosis.
The an average preoperative value of CobbS was 1.45°, representing a lordotic SSP. Age, gender and osteoporosis had no significant effect on the SSP according to preoperative findings. In the early postoperative period, the average CobbS was increased by 319.3% (4.63°). In females, patients younger than 55 years and non-osteoporotic patients the acquired segmental lordotisation was bigger, however the differences were not statistically significant (P=0.05). Patients after 1-level ACDF presented with a significantly higher increase of the CobbS value than the CobbS increase in patients after the 2-level ACDF (P=0.0149). 6 weeks after the ACDF an average decrease of CobbS by 16.6% was observed, 3 months postoperatively the decrease reached 27.6% with a further gentle correction -a total decrease of 27.3% in 6 months postoperatively. The average value of CobbS decreased by The average preoperative value of CobbG was 7.5° with more lordotic values being observed in male patients, however, this difference was not statistically significant. Patients aged 55 years and more presented with a significantly higher average value of the CobbG than younger patients (P=0.0178). Osteoporotic patients also presented with a significantly higher preoperative CobbG than patients without this particular comorbidity (P=0.0011). In the early postoperative period, the average value of CobbG was increased by 34.8%. In contrast to female patients, in the male group we observed a significant early decrease of the CobbG (P=0.0014). Age, osteoporosis and number of operated segments did not affect the early postoperative GSP (P=0.05). Six weeks after the ACDF, the average CobbG was increased by 8.8%, 3 months postoperatively the early obtained average value of CobbG decreased by 2.8%. The average CobbG was later increased -by 21.6% after 6 postoperative months and by 8.6% after 12 postoperative months. 24 months after the ACDF the average CobbG was larger by 1.8% only compared to early postoperative findings, but increased by 25.7% compared to preoperative findings. After a 1-level ACDF the global lordotisation was subsequently lost. In contrast, the global lordotisation increased in the further postoperative period after the 2-level ACDF compared with the early postoperative values, however these differences were not significant (P=0.05). The female patients present with an increase in average postoperative CobbG, while an opposite trend was recorded in male patients. These differences were significant at 6 weeks, 3 months and 6 months after the ACDF (P<0.0298), but not significant in the following period (P=0.05). In osteoporotic patients, the early rec- ognised global lordotisation was significantly decreased at 12 months after the ACDF (P=0.0354), the differences were, however, not significant during the other imaging controls (P=0.05). Patient age did not significantly affect the dynamics of the average postoperative CobbG in 6 weeks to 24 months after the ACDF (P=0.05).
In the early postoperative period both global and segmental lordotisation was increased. However, in the further postoperative X-ray controls the segmental lordotisation was observed to be decreasing while the GSP did not follow this trend (Fig. 3.). 24 months after the ACDF the average value of the CobbS was lower than the preoperative value. In contrast the average CobbG was still increased. The correlation between the values of the CobbS and CobbG was moderate at 3 and 6 months postoperatively and weak in 6 weeks, 12 and 24 months after the ACDF (Table 3).

DISCUSSION
In our study, a single type of an anchored zero profile interbody cage was used, specifically Zero-P VA ® . This implant is fixed by inserting two divergently directed screws introduced into the adjacent vertebral bodies. The cage design assumes an elimination of disadvantages of the conventional cervical plates and maintenance of their benefits 2,3 . Stated advantages of anchored interbody cages include a reduction of postoperative adhesions, reduction of postoperative dysphagia, smaller surgical approach, shorter duration of surgery and a prevention of adjacent segment ossification when compared to those after the use of cages fixed by the conventional plate [3][4][5][6][7][8][9][10] .
We observed an early lordotisation of the treated motion segment, which was more significant after 1-level ACDF. During the further postoperative period, a reduction of the surgically-acquired segmental lordotisation occurred. The change was however not affected by factors such as gender, age, osteoporosis or number of treated segments. The design of the interbody cage used in our study (Zero-P VA ® ) is similar to that of a Zero-P ® implant. A number of studies describe a loss of surgicallyacquired segmental lordotisation after the implantation of Zero-P ® cage 3,[9][10][11][12][13][14] . Fixation of Zero-P VA ® cage (similar to the Zero-P ® cage) is provided by using divergent screws and causes a certain degree of segmental distraction in the ventral portion of the treated motion segment, especially during the early postoperative period after ACDF. A greater lordotising effect therefore occurs compared to the one obtained by a cage fixation with the conventional plate. Also, when the conventional plate is used, a compression of the ventral portion of intervertebral space occurs, causing a reduction of the segmental lordotic profile 3,13,14 .
During the further postoperative period the surgically-acquired segmental lordotisation tends to reduce due to the effect of axial loading and also due to a lower biomechanical stability of anchored cages, especially in anteflexion and retroflexion 2 . When a conventional cervical plate is used for stabilisation, the physiological mechanical load-ing acts against the plate's tension 3 . Several studies have reported a significantly lower segmental lordotisation after Zero-P ® cage implantation compared to that obtained by using a cage fixed by a conventional plate 9,10 . Likewise Zhang et al., we have not confirmed any effect of osteoporosis on the SSP development after the implantation of the anchored cage 15 . Anchored interbody cages fixed with divergently inserted screws are therefore safe to use even in the osteoporotic regions. A more significant segmental lordotisation obtained after 1-level ACDF was characteristic for the whole cohort of our patients. We believe that this is a result of a tendency of the operating surgeon to obtain a greater lordotisating effect when operating in a single motion segment. However, in order to acquire a segmental lordotisation it is important to emphasize the need of a correct placement of the Zero-P ® cage -its anterior titanium plate should always follow the anterior vertebral line 13 . During the first 6 postoperative months the global lordotisation was more prominent in female patients after both 1-and 2-level ACDF with anchored cage. Osteoporosis acted as a risk factor for the loss of the surgically-acquired global lordotisation only in term of 12 months after ACDF. Moreover an average value of CoobG/CoobS was significantly higher in patients with this comorbidity even prior to the surgery. Osteoporosis has also no significant impact on GSP development after the implantation of anchored cage 15 . Numerous studies have found no differences in GSP development when comparing the effects of implanted Zero-P ® cage and cage fixed by the conventional cervical plate 3,14,16,17 . Son et al. however refer a more significant global lordotisation after the implantation of the anchored cage in comparison to a cage fixed by the conventional plate 18 . CS had no significant impact on the clinical outcome after ACDF 19,20 . CS occurred most often in ventral portion of motion segment 19,20. In our study, CS in the ventral portion of motion segment had no significant impact on SSP in term of 24 months after ACDF. However, this result was influenced by the low incidence of CS in our patient cohort (8.2%).
Segmental lordotisation after 1-and 2-level implantation of the anchored interbody cage was accompanied by an increase in global lordosis in the early postoperative period. A gradual loss of segmental lordotisation occurred during the further postoperative period. The GSP did not follow this trend and an overall increase in global lordosis was recorded. The change in the SSP in one or two cervical motion segments only weakly correlated with the GSP change. The 1-or 2-level ACDF has therefore a limited impact on the GSP development after the surgery 3,11,19,21 .

CONCLUSION
The change of SSP after the 1-or 2-level ACDF weakly correlates with the GSP change. A significantly greater increase in CobbS was observed during the early postoperative period after 1-level ACDF compared to that after the 2-level ACDF. Osteoporosis, gender, CS and age have no significant impact on the SSP development after the surgery. Risk factors associated with the loss of the surgically-acquired global lordotisation are male gender and osteoporosis.