Lumbar Disc Degeneration Is an Equally Important Risk Factor as Lumbar Fusion for Causing Adjacent Segment Disc Disease

Lumbar Disc Degeneration Is an Equally Important Risk Factor as Lumbar Fusion for Causing Adjacent Segment Disc Disease Raghu N. Natarajan, Gunnar B.J. Andersson Department of Orthopedic Surgery, Rush University Medical Center, 1611 West Harrison Street, Chicago 60612-3833, Illinois Received 6 January 2016; accepted 29 April 2016 Published online 20 May 2016 in Wiley Online Library (wileyonlinelibrary.com). DOI 10.1002/jor.23283 ABSTRACT: Treatment of degenerative spinal disorders by fusion produces abnormal mechanical conditions at mobile segments above or below the site of spinal disorders and is clinically referred to as adjacent segments disc disease (ASDD) or transition syndrome in the case of a previous surgical treatment. The aim of the current study is to understand with the help of poro-elastic finite element models how single or two level degeneration of lower lumbar levels influences motions at adjacent levels and compare the findings to motions produced by single or two level fusions when the adjacent disk has varying degree of degeneration. Validated grade-specific finite element models including varying grades of disc degeneration at lower lumbar levels with and without fusion were developed and used to determine motions at all levels of the lumbar spine due to applied moment loads. Results showed that adjacent disc motions do depend on severity of disc degeneration, number of disc degenerated or fused, and level at which degeneration or fusion occurred. Furthermore, single level degeneration and single level fusion produced similar amount of adjacent disc motions. The pattern of increase in adjacent segment motions due to disc degeneration and increase in motions at segment adjacent to fusion was similar. Based on the current study, it can be concluded that disc degeneration should also be considered as a risk factor in addition to fusion for generating adjacent disc degeneration. ß 2016 Orthopaedic Research Society. Published by Wiley Periodicals, Inc. J Orthop Res 35:123 130, 2017. Keywords: adjacent disc disease; finite element model; degenerated discs; fusion; risk factors The treatment of degenerative spinal diseases is normally focused on reduction of pain, using both nonoperative and surgical methods. The surgical methods include decompression, fusion, motion preservation, and combinations of these. Subsequent to these treatments, pathological development at mobile segments above or below the site of the intervention is clinically referred to as adjacent segments disc disease (ASDD) or transition syndrome in the case of a previous surgical treatment. We do not know at this time for certain if the ASDD is caused by the neighboring fusion or if they represent the natural progression of the lumbar degenerative processes. Several in vitro experiments using both animal and human cadaveric spine models have shown increased mobility of the adjacent segment as a result of a fusion, a possible etiology of adjacent segment degeneration. 1 3 Bastian et al., 4 Quinnell et al., 2 and Weinhoffer et al. 5 have compared adjacent segment biomechanics after single lumbar segment fusion using cadavers. The effect of short versus long segment fusion was studied by Chow et al. 6 and Nagata et al. 7 Adjacent segment motions were compared between circumferential fusion and anterior fusion by Esses et al. 8 and Lee et al. 1 Finite element studies of normal and fused spine models have also reported an increase in intradiscal pressure in the disc adjacent to a fused level as well as higher stress in the adjacent disc. 5,6,9 11 Most of the literature on spine biomechanical studies of fusion using finite element models are focused on a single variable that affects the adjacent segment motion in a Correspondence to: Raghu N. Natarajan (T: 312-942-5367; F: 312-942-2101; E-mail: raghu_natarajan@rush.edu) # 2016 Orthopaedic Research Society. Published by Wiley Periodicals, Inc. normal spine such as different fusion techniques or motion preservation system. Radiographic analyses of fusion patients have shown an increase in mobility of the adjacent mobile segments. 12 15 A recent MRI evaluation after an interbody fusion with pedicle screw fixation for spondylolisthesis found that more than 70% of the patients developed degeneration at adjacent segments which were normal before surgery. 16 Lumbar spinal segmental motions of 10 patients with degenerated discs at L4/L5 and L5/S1 were analyzed during functional weight-bearing activities using imaging techniques by Passias et al. 17 The results showed increased motions at segments adjacent to degenerated discs. Combined MRI and dual fluoroscopic imaging techniques were used by Wang et al. 18 on 10 subjects with low back pain between L4 and S1 and showed an increase in tensile and shear deformations as well as significantly greater area of large disc deformation at segment cephalic to the degenerated segments during end ranges of motion. The development and severity of ASDD in disc degeneration is broadly believed to depend on a number of risk factors. A 14-year study of patients with either single- or bi-level fusion showed a lower risk of developing ASDD in patients with L5/S1 fusion as compared to patients with L4/5 fusion. 19 A 5-year follow-up study showed that 27% of patients with single level fusion and 45% of three level fusion showed clinical ASDD. 20 The condition of the adjacent disc has also been implicated in ASDD based on the assumption that an already degenerated disc is more likely to deteriorate. 21,22 It is known that a fused disc as well as a degenerated disc is always stiffer than a normal disc and under load produces additional motions at the disc adjacent to the fused/degenerated disc than those 123

124 NATARAJAN AND ANDERSSON occur normally. Thus, adjacent segment disc disease (ASDD) may be produced either by degenerated discs or fusion or both. No study exists that compare the change in biomechanics due to fusion at the adjacent disc that has varying degree of degeneration with change in motions at the adjacent segments due to degeneration at the index level. ASDD due to fusion and/or disc degeneration is broadly believed to depend on a number of variables such as: The severity of degeneration at segment adjacent to fused disc; the number of levels that are degenerated and/or fused; and the location of the degenerative disc or fused disc. The aim of the current study is to explore using finite element models how a fusion at L4/L5 and/or at L5/S1 results in abnormal motions at adjacent segments when they have varying degrees of degeneration which may contribute to the progression of ASDD in a lumbar spine. Furthermore, we explore if degenerative discs by themselves can also produce adjacent segment motions that are comparable in magnitude to those motions produced by fusion. METHODS To study the biomechanical effect of disc degeneration or fusion in a lumbar spine with degenerated adjacent disc, finite element models of different severity of degeneration in the lower lumbar levels were first created. Clinically, the severity of disc degeneration is divided into five grades from 1 normal to 5 severely degenerated (Thompsons Grading system). Our group has developed a procedure to create models of lower lumbar discs with varying grades of degeneration and validated the disc motions by comparing with cadaver test results. 23 26 The procedure adopted is briefly explained here. A previously validated generic poro elastic finite element model of a lumbar spine with normal discs ( II) at all levels 26 was modified to create discs of varying degeneration at different levels. At each level, the geometry of the disc components were modified to correspond to required grade of degeneration. This was achieved by decreasing the disc height and nucleus area of the corresponding normal disc. The disc height corresponding to II was reduced by 15%, 33%, and 70% to represent s III V disc degeneration condition, respectively. Nucleus area for s II and III degeneration case was kept the same, but was reduced by 67% to represent s IV and V degeneration. 23,26 Solid and porous material properties of the annulus and nucleus for different grades of degeneration at different disc levels considered in the current study are shown in Table 1. 23,26 The non-linear stress strain curves of the annular fibers and surrounding ligaments were offset to represent the laxity in the ligaments due to disc height change. 26 Conventionally, finite element models of a lumbar spine are validated by comparing the model-predicted motions with cadaver test results without including grades of disc degeneration into the model analyses. Cadaver spine specimens do contain one or more degenerated discs, as do the lumbar spines of the patients typically operated on. To include this fact in the current study, previously validated generic poro elastic finite element model of a lumbar spine with normal discs at all levels 26 was modified, as explained above, such that the degenerative grade of the discs at all levels correspond to those observed in the cadaver specimens used for biomechanical tests 27 thus creating 12 specimen specific models that correspond to 12 cadaver specimens. For consistency, the same loading was used in the models and in the in vitro experiments: 7.5 Nm moments and a 400 N follower compressive pre-load applied in over 50 s. Motions (mean and standard deviation) at each level obtained from the 12 lumbar spine models with degenerative discs were compared with the corresponding cadaver results 27 to validate the specimen-specific finite element model. This method of creating varying grades of degenerated discs was adopted to simulate spines with pre-existing disc pathology or disc fusion to understand the development and progression of ASDD and explained below. Models to Study the Effect of Progressive Degeneration at Segments L4/L5 and/or L5/S1 on Its Adjacent Segment Motions Sixteen different models representing four different grades of disc degeneration (Thompsons s II V) either at L4/L5 and/or at L5/S1 level were created in the same manner as explained above. A hybrid approach was used for analyzing all the above models. The rationale for using a hybridtesting protocol was that subjects with degenerated discs would attempt to move their spine in the same manner as they did with normal discs during daily activity. So, the baseline range of motion (motion of L1 with respect to S1) remained the same both in the normal spine model as well as in the spine model with degenerated discs. Initially, in addition to a compressive pre-load of 400 N, a moment of 7.5 Nm in the three principal planes was independently applied (load control) in 50 s to the generic lumbar spine model consisting of normal disc at all levels. The motions of L1 with respect to S1 in all the three principal directions were obtained from these analyses and applied to the lumbar spine models with degenerated disc(s). The results of motions at each of the lumbar levels obtained from models containing degenerated disc(s) were compared with the corresponding values obtained from a model of lumbar spine with normal discs. These comparisons provided an understanding on how biomechanical motions at the segments adjacent to degenerated discs depend on (i) severity of disc degeneration at the index level; (ii) number of degenerative discs; and (iii) level at which degeneration occurred. Models to Study the Effect of Fusion at L4/L5 and/or L5/S1 at Adjacent Segment Motions That Has Varying s of Degeneration The current study was performed to understand how fusion of L4/5 and/or L5/S1 in combination with severity of degeneration at the adjacent mobile segment accelerates the progression of ASDD in a lumbar spine. Twelve different models representing fusion at either L4/L5 or L5/S1 or at L4/L5 and L5/S1 along with four different grades of disc degeneration (Thompsons s II V) at the segment adjacent to the fused segment(s) (either at L5/S1 or L4/L5 or at L3/L4) were created as explained before. Furthermore, two additional models representing degenerated discs with grade 3 or 4 at the L3/L4 spinal level were developed by modifying a lumbar spine model with normal discs at all levels. A hybrid approach was again used for analyzing all the above models. The results of motions at each of the lumbar levels obtained from models containing fused segment(s) along with degenerated disc adjacent to fused segment(s) were compared with the corresponding

FUSION, DEGENERATION ADJACENT DISC DISEASE 125 Table 1. Annulus and Nucleus Material Properties Assumed for Different s of Degenerated Discs L3 L4 Annulus Nucleus C1,C2(E06) m E m Pa m 4 N 1 S 1 Pa m 4 N 1 S 1 2 0.28 1.34E 15 8.00Eþ05 1.70E 15 3 0.68 9.84E 16 9.60Eþ05 1.30E 15 4 0.68 8.80E 16 9.60Eþ05 1.27E 15 5 0.136 8.80E 16 9.60Eþ05 1.27E 15 L4 L5 Annulus Nucleus C1,C2(E06) m E m Pa m 4 N 1 S 1 EPa m 4 N 1 S 1 2 0.21 1.01E 15 6.00Eþ05 1.28E 15 3 0.51 7.38E 16 7.20Eþ05 9.78E 16 4 0.51 6.70E 16 7.20Eþ05 9.54E 16 5 0.102 6.70E 16 7.20Eþ05 9.54E 16 L5 S1 Annulus Nucleus C1,C2(E06) m E m Pa m 4 N 1 S 1 Pa m 4 N 1 S 1 2 0.6 1.34E 15 2.40Eþ06 1.70E 15 3 0.72 9.84E 16 2.80Eþ06 1.30E 15 4 0.72 8.90E 16 2.80Eþ06 1.24E 15 5 0.3 8.90E 16 2.80Eþ06 1.24E 15 C1 and C2, hyper-elastic Mooney Rivilin material constants used in defining annulus material properties; E, elastic modulus of the tissue; m, permeability of the tissue. Material properties for grades 2 5 annulus and nucleus were adopted from the literature (Qasim et al. 23 ; Ruberte et al. 26 ). values obtained from a model of the lumbar spine with degenerated discs located at the same level as in the fused spine model. This method of comparison was adopted so that the biomechanical effect of fusion alone in a spine with degenerated adjacent segment can be studied. These comparisons provided understanding on how biomechanical motions at the segments adjacent to fused segment depend on (i) severity of disc degeneration at the level adjacent to fused segment; (ii) number of fused discs; and (iii) level at which the disc was fused. RESULTS The compressive pre-load of 400 N combined with a moment of 7.5 Nm in the three principal planes applied independently to a spine with grade 2 discs at all levels produced motions of L1 with respect to S1 of 29.9 in flexion, 22.2 in extension, 13.8 in L þ R torsion, and 41.3 in L þ R lateral bending. These were the motions that were applied to the lumbar spine models with degenerated discs or models that contain both fused discs as well as degenerated discs so that the base line range of motion remained the same both in the normal spine model and in the spine models with degenerated and/or fused discs. Validation of the Specific Finite Element Model Mean and standard deviation of the segmental motions at each level obtained from the 12 specimen specific finite element models were compared with the segmental motions obtained from the cadaver specimens and showed an excellent agreement (Fig. 1). This validated the current finite element model approach in representing the biomechanical effect of different stages of disc degeneration in a lumbar spine. Validation of Primary Motions Obtained From the FE Models With In Vivo Measured Motions Segmental in vivo vertebral motion during functional human lumbar spine activities measured in a normal spine 28 and in a spine with degenerated discs at L4/S1 levels 17 were compared with the motions obtained from FE models. Instead of comparing the actual segmental motions obtained from in vivo study with FE model results, contributions from each segment to the total lumbar spine motion in both in vivo study and FE model results were compared (Figs. 2 and 3). This approach for validating the FE model result was adopted because the FE model does not include the

126 NATARAJAN AND ANDERSSON Figure 1. Comparison of specimen-specific model results with cadaver results. The model results in all the three principal directions at all the lumbar disc levels compared very well with the corresponding cadaver test results. effect of muscles and the maximum in vivo motions the lumbar spine experienced were not the same as calculated in FE models. Change in Adjacent Segment Biomechanics Due to Progressive Degeneration of Segments L4/L5 and/or L5/S1 In models with single level degeneration, larger increase in adjacent segments motion for all loading modes was seen when L4/L5 segment was degenerated (varying between 5% and 38%) as compared to increase in adjacent segment motions with L5/S1 degeneration (varying between 3% and 13%) (Table 2). In all loading modes, the largest increase in adjacent segment motions was seen when the degenerated segment at L4/L5 or L5/S1 had a grade V degeneration. When both L4/L5 and L5/S1 segments were degenerated, maximum increase in adjacent segment motions (varying between 25% and 45%), as Figure 3. Range of primary motions of the degenerated lumbar discs at different levels. Comparison of model evaluated motion carried by each of the degenerated segments as a percentage of total lower lumbar spine motion agreed well with the corresponding value obtained in an in vivo study that is available in the literature. expected, was larger than those produced by single segment degeneration for all loading modes (Table 3). Maximum increase in adjacent segment motions occurred with grade V in both L4/L5 and L5/S1 segments except under torsion. Also, for a specific grade of degeneration at either L4/L5 or L5/S1 level, the adjacent segment motions (L3/L4) increased monotonically as the grade of degeneration increased in the other disc (L5/S1 or L4/L5, respectively). For example, in flexion, with grade V degeneration at L4/L5 as the degenerative grade of L5/S1 disc increased from s II to V, the adjacent segment motions increased from 19% to 42%. Motion at disc L3/L4 adjacent to degenerated L4/S1 segments ( V in both L4/L5 and L5/S1 in FE model) was calculated as a percentage of L2/L5 segmental motions from FE model results. The model results of motion at L3/L4 were compared with corresponding percentage motion at L3/L4 calculated from results based on a published in vivo study. 17 The results compared well in both cases when L4/S1 segments were either normal or degenerated for all the three loading cases and presented in Figure 4. Table 2. Percent Change in Adjacent Segment Motions Due to Degenerated Disc Either at L4/L5 or L5/S1 Figure 2. Range of primary motions of the normal lumbar discs at different levels. Comparison of model evaluated motion carried by each of the normal segments as a percentage of total lower lumbar spine motion agreed well with the corresponding value obtained in an in vivo study that is available in the literature. Flexion Extension R þ L Torsion L þ R Lat. Bending Degenerative grade at L4/L5 III 10 17 5 19 IV 16 34 12 24 V 19 38 7 27 Degenerative grade at L5/S1 III 3 0 3 3 IV 9 5 9 7 V 13 9 9 9

FUSION, DEGENERATION ADJACENT DISC DISEASE 127 Table 3. Maximum Percent Change in Adjacent Segment Motions (%) Due to Degenerated Disc Both at L4/L5 and L5/S1 Degenerative at L5/S1 Flexion Degenerative at L4/L5 II III IV V II 0 10 16 19 III 3 13 21 24 IV 10 21 28 32 V 15 27 34 42 Extension Degenerative at L4/L5 Degenerative at L5/S1 II III IV V II 0 17 34 38 III 7 32 36 40 IV 27 36 38 42 V 32 44 44 35 Degenerative at L5/S1 R þ L Torsion Degenerative at L4/L5 II III IV V II 0 5 12 7 III 3 9 16 10 IV 9 23 22 16 V 10 16 23 14 Degenerative at L5/S1 R þ L Lat. Bending Degenerative at L4/L5 II III IV V II 0 19 26 29 III 4 23 32 36 IV 8 28 38 42 V 11 32 42 45 Change in Adjacent Segment Biomechanics Due to Fusion at L4/L5 and/or L5/S1 Segment(s) Fusion at L4/L5 in most cases produced larger increase (22 50%) in adjacent segment motions for all grades of degeneration at L5/S1 (Table 4) as compared to the corresponding increase (12 35%) due to fusion at L5/S1 for all grades of degeneration at L4/L5. With L4/L5 fusion, the maximum percentage increase in motions at adjacent segment L5/S1 occurred with grade V degeneration at L5/S1 (varying between 28% and 50%). On the other hand, with L5/S1 fusion, the largest increase in L4/L5 segment motion was observed with grade III or IV degeneration at L4/L5 Figure 4. Comparison of FE model with in vivo motions at L3/L4 as a percentage of total L2/L5 motion when L4/S1 is either normal or degenerated. Comparison of model evaluated motion carried by L3/L4 segment adjacent to degenerated L4/S1 segment as a percentage of total lower lumbar spine motion agreed well with the corresponding value obtained in an in vivo study that is available in the literature. (varying between 24% and 49%). As anticipated, the change in adjacent segment motions (L3/L4) was larger (60% and 200%) with two level fusion (L4/S1) as compared to single level fusion (Table 4). Degenerative grade IV at the adjacent segment L3/L4 produced largest increase in adjacent segment motions (varying between 98% and 137%). Comparison of Adjacent Segment Biomechanics Due to Fusion With Adjacent Segment Motions Due to Disc Degeneration Single level degenerated disc at L4/L5 produced largest increase in adjacent segment motions of 15 40% that were comparable to those produced by one level fusion at L4/L5 (30 50%) (Fig. 5a). Similarly, single level degenerated disc at L5/S1 produced largest increase in adjacent segment motions of 10 15% that were comparable to those produced by one level fusion at L5/S1 (25 50%). Two level disc degeneration produced a maximum increase in adjacent segment motions of 25 45% which was lower than those produced by two level fusion (100 200%). Comparison of the largest increase in adjacent segment motions due to one level fusion or degeneration showed that larger adjacent segment motion always occurred when L4/L5 was fused or degenerated as compared to fusion or degeneration at L5/S1. Similarly, comparison of the largest increase in adjacent segment motions due to one level fusion or degeneration with two level fusion or degeneration, showed that two level fusion or degeneration produced largest change in adjacent segment motions. To additionally compare the effect of fusion or disc degeneration on adjacent disc biomechanics, the mean

128 NATARAJAN AND ANDERSSON Table 4. Percent Change in Adjacent Segment Motions Due to Fusion at L4/L5 and/or L5/S1 Segments Percent change in adjacent segment (L5S1) motions due to fusion at L4/L5 Degenerative grade at L5/S1 Flexion Extension R þ L Torsion L þ R Lat. Bending II 33 25 27 42 III 36 22 27 45 IV 36 25 29 47 V 35 28 29 50 Percent change in adjacent segment (L4L5) motions due to fusion at L5/S1 Degenerative grade at L4/L5 Flexion Extension R þ L Torsion L þ R Lat. Bending II 24 23 18 13 III 32 49 22 32 IV 34 27 24 22 V 32 25 14 21 Percent change in adjacent segment (L3L4) motions due to fusion at L4/L5 and L5/S1 Degenerative grade at L3/L4 Flexion Extension R þ L Torsion L þ R Lat. Bending II 83 113 57 62 III 126 198 63 90 IV 137 111 98 106 effect (produced by all grades of degeneration considered in this study) of disc degeneration or fusion on adjacent segment motion for each loading mode was calculated (Fig. 5b). The increase in mean adjacent segment motions due to L4/L5 degeneration was about 50% of the corresponding increase produced by fusion at L4/L5. Similarly, increase in mean adjacent segment motions due to L5/S1 degeneration was about 30% of the corresponding increase produced by fusion at L5/S1 in all loading modes. Increase in mean adjacent segment motions due to two level degeneration was about 15% of the corresponding increase produced by fusion at L4/S1 in all loading modes. DISCUSSIONS The method adopted in the FE modeling to represent degenerated discs in a lumbar spine model was validated by comparing the FE model results with cadaver results. Further the primary motions obtained in both normal lumbar spine and spine with degenerated lumbar discs obtained from FE models were compared with segmental rotations obtained in the three principal directions from an in vivo study and available in the literature. The results compared well at all lower lumbar levels in all loading modes. The current study showed that degeneration or fusion at L4/L5 caused larger increase in adjacent segment motions than those produced by L5/S1 degeneration or fusion leading to the conclusion that chances of developing ASDD due to either fusion or disc degeneration are greater when L4/L5 level degenerates or is fused. The conclusion that the magnitude of adjacent segment motions do depend on the level at which fusion occur is very similar to those conclusion reached by a 14-year study of patients that showed a lower risk of developing ASDD in patients with L5/S1 fusion as compared to patients with L4/L5 fusion. 19 The results also showed that as the severity of disc degeneration at lower lumbar level discs increased, the motions at segments adjacent to degenerated disc also increased leading to the conclusion that increase in adjacent segment motions do depend on the severity of disc degeneration. Two level disc degeneration or two level fusion was found to produce larger increase in adjacent segment motions than produced by single level degeneration or fusion. This conclusion that the adjacent segment motions depend on number of fused segments compare well with those reached after a 5-year follow-up study that showed larger percentage of patients with multilevel fusion developed ASDD as compared to patients with single level fusion. 20 The conclusion that presence of degenerated discs at L4/S1 levels produce larger increase in motions at L3/L4 than that in normal discs compares well with the conclusions from an in vivo study. 17 Also, the comparison of L3/L4 motions from FE model with in vivo motions as a percentage of L2/L5 motion when two level degeneration (L4/S1) was considered showed an excellent agreement in all loading modes, again validating the current FE model. With single level fusion either at L4/L5 or at L5/S1 or two level fusion at L4/S1 levels showed increase in adjacent segment motions with increase in degenerative grade at that adjacent segment leading to the conclusion that the motions at the segment adjacent to the fusion do depend on the severity of degeneration at the adjacent segment. This conclusion that an already degenerated disc at the segment adjacent to the fused level is more

FUSION, DEGENERATION ADJACENT DISC DISEASE 129 Figure 5. (a) Largest change in adjacent segment motions due to either fusion or degeneration. Degenerated disc at L4/L5 produced similar amount of largest adjacent segment motions as induced by the fusion at L4/L5. Degenerated disc at L5/S1 produced nearly half of the amount of largest adjacent segment motions as induced by the fusion at L5/S1. Degenerated disc at L4/S1 produced nearly half of the amount of largest adjacent segment motions as induced by the fusion at L4/S1. (b) Effect of disc degeneration or fusion on mean adjacent segment motion for each loading mode. Degenerated disc at L4/L5 produced similar amount of mean adjacent segment motions as induced by the fusion at L4/L5. Degenerated disc at L5/S1 produced nearly half of the mean amount of adjacent segment motions as induced by the fusion at L5/S1. Degenerated disc at L4/S1 produced nearly half of the mean amount of adjacent segment motions as induced by the fusion at L4/S1. likely to deteriorate was also reached in several in vivo studies. 20,21,26 28 Similarly, in the case of two level degeneration, the increase in adjacent disc motions do depend on level of degeneration at each of these two levels. The effects of disc degeneration on adjacent disc motions and the increase in motions at segment adjacent to fused segment(s) were found to be very similar. The findings that fusion at L4/L5 is a greater risk than fusion at L5/S1 was similar to the conclusion that degeneration at L4/L5 produces larger adjacent segment motions as compared to motion produced by degeneration at L5/S1. Two level disc degeneration produced larger adjacent segment motions than produced by single level degeneration is similar to that two level fusion produced larger increase in adjacent segment motions than induced by single level fusion. Increase in adjacent segment motions do depend on severity of disc degeneration at the index level(s). This is similar to that the adjacent segment motions do depend on the severity of degeneration at the disc adjacent to fusion. The mean increase in adjacent segment motions due to disc degeneration was of similar magnitude as of those increase produced by fusion in all loading modes. There are some limitations on the current study. It is well known that the FE model study cannot simulate a living system exactly such as, to name a few, include the normally observed phenomenon of no distinct demarcation between nucleus and annulus in a degenerated lumbar disc, change in the other components of the motion segment such as endplates, facet joints, and vertebra that simultaneously occur along with disc degeneration. A systematic study of the effect of fusion and degeneration on adjacent disc biomechanics which is difficult but not impossible to do by other means (such as cadaver study or in vivo study) can be accomplished by this model analyses. Muscles, which are not included in the current model, do affect the kinematics of the spine and is a limitation of the current study. Movement of patients with back pain and patients with fusion may be limited because of physiological conditions. In the current study, the motion of L1 with respect to S1 is made same in normal as well as in degenerated/fusion cases (Hybrid method of analyses). This is another limitation to the current study. In conclusion, the current study showed that the increase in adjacent segment motions (i) depend on the level at which degeneration or fusion occur; (ii) depend on severity of degeneration at the index level; (iii) depend on number of levels fused or degenerated; and (iv) depend on the severity of degeneration at the disc adjacent to fused disc. Also, the effects of disc degeneration on adjacent disc motions and the increase in motions at segment adjacent to fused segment(s) were found to be very similar. The current study thus suggest that in addition to fusion being one of the factors that cause ASDD, degeneration of the discs can also cause increase in adjacent segment motions that could be a biomechanical factor relating to progression of disc degeneration. AUTHORS CONTRIBUTIONS Both the authors contributed to the article. Both the authors have read and approved the manuscript. REFERENCES 1. Lee CK, Langrana NA. 1984. Lumbosacral spinal fusion. A biomechanical study. Spine 9:574 581. 2. Quinnell RC, Stockdale HR. 1981. Some experimental observations of the influence of a single lumbar floating fusion on the remaining lumbar spine. Spine 6:263 267.

130 NATARAJAN AND ANDERSSON 3. Yang SW, Langrana NA, Lee CK. 1986. Biomechanics of lumbosacral spinal fusion in combined compression-torsion loads. Spine 11:937 941. 4. Bastian L, Lange U, Knop C, et al. 2001. Evaluation of the mobility of adjacent segments after posterior thoracolumbar fixation: a biomechanical study. Eur Spine J 10:295 300. 5. Weinhoffer SL, Guyer RD, Herbert M, et al. 1995. Intradiscal pressure measurements above an instrumented fusion. A cadaveric study. Spine 20:526 531. 6. Chow DH, Luk KD, Evans JH, et al. 1996. Effects of short anterior lumbar interbody fusion on biomechanics of neighboring unfused segments. Spine 21:549 555. 7. Nagata H, Schendel MJ, Transfeldt EE, et al. 1993. The effects of immobilization of long segments of the spine on the adjacent and distal facet force and lumbosacral motion. Spine 18:2471 2479. 8. Esses SI, Doherty BJ, Crawford MJ, et al. 1996. Kinematics evaluation of lumbar fusion techniques. Spine 21:676 684. 9. Chen CS, Cheng CK, Liu CL. 2002. A biomechanical comparison of posterolateral fusion and posterior fusion in the lumbar spine. J Spinal Disorders Tech 15:53 63. 10. Chen CS, Cheng CK, Liu CL, et al. 2001. Stress analysis of the disc adjacent to interbody fusion in lumbar spine. Med Eng Phys 23:483 491. 11. Cegonino J, Calvo-Echenique A, Palomar P. 2015. Influence of different fusion techniques in lumbar spine over adjacent segments: a 3D finite element study. J Orthop Res 33:993 1000. 12. Axelsson P, Johnsson R, Stromqvist B. 1997. The spondylolytic vertebra and its adjacent segment. Mobility measured before and after posterolateral fusion. Spine 22:414 417. 13. Frymoyer JW, Hanley EN, Howe J, et al. 1979. A comparison of radiographic findings in fusion and nonfusion patients ten or more years following lumbar disc surgery. Spine 4:435 440. 14. Lehmann TR, Spratt KF, Tozzi JE, et al. 1987. Long-term follow up of lower lumbar fusion patients. Spine 12:97 104. 15. Stokes IA, Wilder DG, Frymoyer JW, et al. 1981. 1980 Volvo award in clinical sciences. Assessment of patients with lowback pain by biplaner radiographic measurement of intervertebral motion. Spine 6:233 240. 16. Kim KH, Lee SH, Shim CS, et al. 2010. Adjacent segment disease after interbody fusion and pedicle screw fixations for isolated L4-L5 spondylolisthesis: a minimum five-year follow up. Spine 35:625 634. 17. Passiac PG, Wang S, Kozanek M, et al. 2011. Segmental lumbar rotation in patients with discogenic low back pain during functional weight-bearing activities. J Bone Joint Surg 93-A:29 37. 18. Wang S, Xia Q, Passias P, et al. 2011. How does lumbar degenerative disc disease affect the disc deformation at the cephalic levels in vivo. Spine 36:E574 E581. 19. Disch AC, Schmoelz W, Matziolis G, et al. 2008. Higher risk of adjacent segment degeneration after floating fusions: long-term outcome after low lumbar fusions. J Spinal Disord 21:79 85. 20. Cheh G, Bridwell KH, Lenke GL, et al. 2007. Adjacent segment disease following lumbar/thoracolumbar fusion with pedicle screw instrumentation. Spine 32:2253 2257. 21. Hsu K, Zucherman J, White A. 1993. The long term effect of lumbar spine fusion: deterioration of adjacent motion segments. In: Yonenobu K, One K, Takemitsu Y, editors. Lumbar fusion and stabilization. Tokyo: Springer. p54 64. 22. Naki S, Yoshizawa H, Kobayashi S. 1999. Long-term followup study of posterior lumbar interbody fusion. J Spinal Disord 12:293 299. 23. Qasim M, Natarajan RN, An HS, et al. 2014. Damage accumulation under cyclic loading in the lumbar disc shifts from inner annulus lamellae to peripheral with increasing disc degeneration: a finite element study. J Biomech 47:24 31. 24. Natarajan RN, Williams JR, Andersson GBJ. 2006. Modeling changes in intervertebral disc mechanics with degeneration. J Bone Joint Surg Am 88A:36 40. 25. Natarajan RN, Williams JR, Andersson GBJ. 2004. Recent advances in analytical modeling of lumbar disc degeneration. Spine 29:2733 2741. 26. Ruberte LM, Natarajan RN, Andersson GBJ. 2009. Influence of single level lumbar degenerative disc disease on the behavior of the adjacent segments a finite element model study. J Biomech 42:341 348. 27. Renner SM, Natarajan RN, Patwardhan AG, et al. 2007. Novel model to analyze the effect of a large compressive follower pre-load on range of motions in a lumbar spine. J Biomech 40:1326 1332. 28. Li G, Wang S, Passias P, et al. 2009. Segmental in vivo vertebral motion during functional human lumbar spine activities. Eur Spine J 18:1013 1021.