Research Article: Normal aging in human lumbar discs: An ultrastructural comparison

Date Published: June 20, 2019

Publisher: Public Library of Science

Author(s): Ricardo B. V. Fontes, Josemberg S. Baptista, Said R. Rabbani, Vincent C. Traynelis, Edson A. Liberti, Thomas Abraham.


The normal aging of the extracellular matrix and collagen content of the human lumbar intervertebral disc (IVD) remains relatively unknown despite vast amounts of basic science research, partly because of the use of inadequate surrogates for a truly normal, human IVD. Our objective in this study was to describe and compare the morphology and ultrastructure of lumbar IVDs in 2 groups of young (G1—<35 years) and elderly (G2—>65 years). Thirty L4-5 and L5-S1 discs per group were obtained during autopsies of presumably-asymptomatic individuals and analyzed with magnetic resonance imaging (MRI), a morphological grading scale, light microscopy, scanning electron microscopy (SEM) and immunohistochemistry (IHC) for collagen types I, II, III, IV, V, VI, IX and X. As expected, a mild to moderate degree of degeneration was present in G1 discs and significantly more advanced in G2. The extracellular matrix of G2 discs was significantly more compact with an increase of cartilaginous features such as large chondrocyte clusters. Elastic fibers were abundant in G1 specimens and their presence correlated more with age than with degeneration grade, being very rare in G2. SEM demonstrated persistence of basic structural characteristics such as denser lamellae with Sharpey-type insertions into the endplates despite advanced age or degeneration grades. Immunohistochemistry revealed type II collagen to be the most abundant type followed by collagen IV. All collagen types were detected in every disc sector except for type X collagen. Statistical analysis demonstrated a general decrease in collagen expression from G1 to G2 with an annular- and another nuclear-specific pattern. These results suggest modifications of IVD morphology do not differ between the anterior or posterior annulus but are more advanced or happen earlier in the posterior areas of the disc. This study finally describes the process of extracellular matrix modification during disc degeneration in an unselected, general population and demonstrates it is similar to the same process in the cervical spine as published previously.

Partial Text

Pathologic conditions of the spine have been described ever since medical knowledge was placed in written form. The management of fractures and dislocations is featured in the Edwin Smith papyrus (c. 1600 BC) and Hippocratic writings (c. 400 BC) [1,2]. Low back pain due to degenerative conditions, however, would only feature prominently in medical texts during the Industrial Age, possibly due to a combination of an increase in life expectancy, technical advances in surgery and even the emergence of litigation for work- and accident-related health issues[3]. Today the lifetime prevalence of low back pain (LBP) approaches 80% and it is not only one of the most common general medical complaints but is the biggest cause of years lived with a disability [4–6]. Accordingly, a great body of literature has been produced since the mid-19th century focusing on the anatomy and degeneration of the intervertebral disc (IVD). Its basic structure has been described since at least 1858 and a very important series of three papers on modifications induced by aging was published in 1945[7,8].

This work was performed with human whole L4-5 and L5-S1 discs obtained during unselected autopsies as described in our previous study with cervical discs [13]. The larger size of lumbar discs and the usual presence of an identifiable nucleus pulposus (NP) allowed them to be divided into 3 sectors for analysis: anterior (aAF), a middle fragment (NP) and posterior (pAF). This study was granted IRB approval at ICB-USP. Briefly, thirty L4-S1 vertebral blocks were collected from unselected autopsies of recently-deceased (<6 hours) cadavers at the SVOC-USP. Next of kin provided written consent and were interviewed to exclude individuals with known history of neck or back pain, neoplasms or rheumatological conditions[14]. Specimen age and data is provided in Table 1. Group 1 (G1) was comprised of 15 young cadavers (<35 years) and group 2 (G2) included 15 specimens from cadavers aged 65 or older. Throughout the study, L4-5 and L5-S1 discs were analyzed jointly, thus resulting in 30 discs/age group. Specimens were assigned random identifiers and masked to researchers. There is a wealth of basic science data on lumbar disc degeneration produced since Schmorl and Junghanns’ pioneering study in 1932 but how much of it is directly and perfectly applicable to human disc degeneration is unknown. For example, discs from quadruped species, particularly canines, have a propensity to undergo calcification during normal degeneration, while the rat tail suspension model has been extensively demonstrated to differ from normal human disc aging from both morphological and molecular standpoints[10,11,19]. Lumbar discs from patients undergoing deformity correction surgery have different mechanical and matrix properties than those undergoing surgery for degenerative disc disease, including different collagen and calcium content and are inadequate surrogates for normal or even pathological degeneration[9,20]. Collagen content in discs from other animal species exhibit variations in collagen composition ranging from 50 to 600% of the equivalent human content, despite prior interpretations of this collagen content being “similar”[21]. In vitro and animal models are thus very useful for testing therapeutic possibilities in a pre-clinical setting but are inherently poor choices to understand ECM modifications during normal disc aging, while discs from human patients being treated for other pathologies should be reserved for the study of those particular conditions[22].   Source: