View metadata, citation and similar papers at core.ac.uk brought to you by CORE provided by Bern Open Repository and Information System (BORIS) Eur Spine J (2006) 15: 965–971 DOI 10.1007/s00586-005-0986-3 ORIGINAL ARTICLE Bronek M. Boszczyk An immunohistochemical study of the tissue Alexandra A. Boszczyk Wolfdietrich Boos bridging adult spondylolytic defects—the Andreas Korge H. Michael Mayer presence and significance Reinhard Putz of fibrocartilaginous entheses Michael Benjamin Stefan Milz Abstract Introduction Spondylolytic also detected around bony spurs that Received: 29 December 2004 Revised: 9 March 2005 spondylolisthesis is an osseous dis- extended from the enthesis into the Accepted: 10 June 2005 continuity of the vertebral arch that lysis-zone. The entheses also labelled Published online: 7 September 2005 predominantly affects the fifth lum- for types I, III and VI collagens, Ó Springer-Verlag 2005 bar vertebra. Biomechanical factors chondroitin four and six sulfate, ker- are closely related to the condition. atan and dermatan sulfate, link pro- An immunohistochemical investiga- tein, versican and tenascin. tion of lysis-zone tissue obtained Conclusions: Although the gap filled from patients with isthmic spondyl- by the lysis tissue is a pathological olisthesis was performed to determine feature, the tissue itself has hallmarks the molecular composition of the ly- of a normal ligament—i.e. fibrocar- sis-zone tissue and enable interpreta- tilaginous entheses at either end of an tion of the mechanical demands to ordered collagenous fibre structure. B. M. Boszczyk Æ W. Boos which the tissue is subject. Methods: The fibrocartilage is believed to dis- Neurosurgery, Berufsgenossenschaftliche During surgery, the tissue filling the sipate stress concentration at the Unfallklinik Murnau, Murnau, Germany spondylytic defects was removed hard/soft tissue boundary. The B. M. Boszczyk (&) from 13 patients. Twelve spondylo- widespread occurrence of molecules Orthopaedic Surgery, listheses were at the L5/S1 level with typical of cartilage in the attachment Inselspital, University of Berne, Bern, Switzerland slippage being less than Meyerding of the lysis tissue, suggests that com- E-mail: [email protected] grade II. Samples were methanol pressive and shear forces are present fixed, decalcified and cryosectioned. to which the enthesis is adapted, in A. A. Boszczyk Æ R. Putz Æ S. Milz Anatomische Anstalt, Sections were labelled with a panel of addition to the expected tensile forces Ludwig-Maximilians-Universita¨t, monoclonal antibodies directed across the spondylolysis. Such a Mu¨nchen, Germany against collagens, glycosaminogly- combination of tensile, shear and A. Korge Æ H. M. Mayer cans and proteoglycans. Results: The compressive forces must operate Spine Center, Orthopa¨dische Klinik Mu¨n- lysis-zone tissue had an ordered col- whenever there is any opening or chen Harlaching, Mu¨nchen, Germany lagenous structure with distinct fi- closing of the spondylolytic gap. M. Benjamin brocartilaginous entheses at both School of Biosciences, ends. Typically, these had zones of Keywords Spondylolytic University of Wales, Cardiff, UK calcified and uncalcified fibrocarti- spondylolisthesis Æ Spondylolysis Æ S. Milz lage labelling strongly for type II Fibrocartilage Æ Aggrecan Æ Type II AO-Research Institute, Davos, Switzerland collagen and aggrecan. Labelling was collagen Introduction isthmus—the pars interarticularis—predominantly occur- ring in the fifth lumbar vertebra. Kilian [27] is credited Spondylolytic spondylolisthesis (SSL) is defined as an with the first recognition of spondylolisthesis as anterior osseous discontinuity of the vertebral arch at the slippage of the last lumbar vertebra against the sacrum 966 [12, 40]. The first theoretical and subsequent biomechan- imaging, was present in all cases. Persistent or exacer- ical identification of a discontinuity of the pars interar- bated back pain had been present for at least 3 months ticularis as a compulsory feature of spondylolisthesis, in in all patients and most reported a history of recurring addition to disruption of the intervertebral disc, was made low back pain for several years along with mild symp- by Robert [43]. Several epidemiological studies have since toms of L5 nerve root compression. Informed consent revealed the incidence in Caucasian populations to reach was obtained from all patients for the entire procedure 4–6% [35, 52], which rise as high as 26% in secluded Es- and the Declaration of Helsinki (http://www.fda.gov/oc/ kimo populations [50]. The classical form of SSL does health/helsinki89.html) was strictly followed. The however not occur in non-ambulatory individuals [44], spondylolytic region was identified during surgery and linking it not only to a genetic predisposition [18, 62, 58] the lysis-zone tissue and a small part of the adjacent but also to biomechanical factors associated with spino- bone was removed as completely as possible without pelvic balance and repetitive loading [4, 15, 22, 25, 34, 59]. endangering the neighbouring nerve root. In six patients, While several conventional histological investigations of the tissue could be collected from both sides and in se- the isthmic lesion have been performed in children [39, 49, ven, from one side only. Thus a total of 19 specimens 64] and adults [52, 58], the functional significance of the were examined. In 12 of these, both entheses were in- tissue bridging the bony defect (subsequently referred to cluded, but only one enthesis was available in the others. as the ‘lysis-zone tissue’) is unclear. As this tissue forms a Tissue samples were fixed for 24 h in 90% methanol at fibrous link between two regions of bone, it acts as a lig- 4°C. The tissue was decalcified in 5% EDTA, infiltrated ament. Normal ligaments have specialised bony attach- with a 5% sucrose solution in PBS for 12 h and cryo- ment sites or ‘entheses’, which may be fibrous or sectioned at 12 lm on a HMV500 Microm cryostat. fibrocartilaginous, depending on the mechanical loads to Sections were stained with toluidine blue and labelled which they are subject [2, 3]. Although both types of en- with a panel of monoclonal antibodies directed against theses transfer tensile load to bone, fibrocartilaginous collagens (types I, II, III and VI), glycosaminoglycans attachments are also subject to compression and shear [2, (chondroitin four and six sulfates, keratan and dermatan 3]. This is reflected by differences in the composition of the sulfates) and proteoglycans (aggrecan, link protein, extracellular matrix (ECM) and in particular by the versican and tenascin). Details of the antibodies are gi- presence of aggrecan and type II collagen at fibrocarti- ven in Table 1, together with pre-treatment procedures. laginous entheses [2, 3]. These molecules are typical of The activity of endogenous peroxidase was blocked with articular cartilage —which is a tissue noted for its ability 0.3% hydrogen peroxide in methanol and any non-spe- to distribute compressive forces across synovial joints cific binding of the secondary antibody was reduced by [55]. In any ligament where ‘insertional angle’ changes incubating the sections with horse serum. Antibody accompany joint movement, that ligament will inevitably binding was detected with a Vectastain ABC ‘Elite’ be subject to compressive and shearing forces at its en- avidin/biotin/peroxidase kit (Vector Laboratories, Bur- theses. As the tissue bridging a spondylolytic defect can be lingame, CA, USA) and control sections were made by viewed as a very short ligament, any relative movement of omitting the primary antibody. All control sections were bone on either side of the spondylolysis should magnify unlabelled insertional angle change. We have thus sought to deter- mine the structure of the lysis-zone tissue through an immunohistochemical study of its ECM and, in particu- Results lar, whether its entheses are fibrous or fibrocartilaginous, hereby-enabling interpretation of the mechanical de- The position of the lysis-zone tissue in a typical spondyl- mands to which the tissue is subject. olytic defect at L5/S1 is shown in Fig. 1a. In all specimens, the tissue was highly organised with structurally ordered collagen bundles and distinct fibrocartilaginous entheses Materials and methods at both its pedicle and vertebral arch ends (Figs. 1b, 2a). Typically, these entheses had zones of calcified and un- The lysis-zone tissue filling the spondylytic defects of 13 calcified fibrocartilage that were separated from each individuals (age range 36–60 years, mean 45 years, both other by a tidemark (Fig. 2a). The cells in the uncalcified sexes) undergoing spondylodesis for low-grade isthmic fibrocartilage were typically arranged in longitudinal spondylolisthesis was removed during surgery. All pa- rows (Fig. 2a). tients had bilateral spondylolysis and all of the spond- In all specimens, the ECM of the enthesis fibrocar- ylolisthesis were at the L5/S1 level except one, which was tilage labelled strongly for type II collagen and aggre- at L4/5. In no case was slippage greater than Meyerding can (Fig. 2b, c). Labelling was also detected around grade II (i.e. greater than 50% [61]). Moderate to bony spurs that extended from the enthesis into the advanced disc degeneration, with obvious loss of disc lysis-zone in some specimens (Fig. 2d) and in regions of height or loss of hydration signal on magnetic resonance chondroid bone (Fig. 2e). At the latter sites, the fi- 967 Table 1 List of monoclonal antibodies used, together with their dilutions, pretreatments and sources Antigen(s) recognized: Antibody Dilution Enzyme pretreatment Source References Collagen I Col 1 1:2000 Hyal (1.5 IU/ml) and ChABC (0.25 IU/ml) Sigma None Collagen II CIICI 1:5 Hyal (1.5 IU/ml) and ChABC (0.25 IU/ml) DSHB 21 Collagen III III-53 (4H12) 1:500 Hyal (1.5 IU/ml) and ChABC (0.25 IU/ml) ICN None Collagen VI 5C6 1:6 None DSHB 20 Chondroitin-4-sulfate 2B6 1:1500 ChAC (0.25 IU/ml) B. Caterson 10 Chondroitin 4 and dermatan sulfates 2B6 1:1500 ChABC (0.25 IU/ml) B.