Imaging features of the aging spine

Introduction

Morphological changes of the spine resulting from aging are natural and progressive processes [1-3]. Aging is also a risk factor for the development of osteoarthritis of the spine, during which degeneration of intervertebral discs, vertebral bodies, and intervertebral joints are often its first manifestations [4,5]. Therefore, the exact distinction between the physiological aging of the spine and its degeneration in the course of degenerative disease is not always straightforward. Clinicians need to pinpoint the culprits responsible for pain, suffering, and diminished quality of life, because findings from imaging studies might often trigger medical and surgical interventions [6].
Knowledge of the incidence and morphology of individual markers of spine aging in various age groups and within selected structures may aid in the differential diagnosis between symptomatic disease and asymptomatic changes related to aging alone. Therefore, this narrative review aims to provide an overview of the features related to the aging of the spine and distinguish them from pathological degenerative changes, based on the literature. Therefore, we aimed to answer the following questions: What is the prevalence of imaging findings associated with degenerative spine conditions in asymptomatic individuals? Which of these findings are strongly associated with older age despite an asymptomatic clinical picture? What is the typical image of the aging spine on magnetic resonance imaging (MRI)?

Material and methods

We reviewed articles located using the PubMed, EMBASE, and Google Scholar databases. The databases were searched for the following keywords: “aging”, “asymptomatic”, “magnetic resonance imaging”, “spine”, “intervertebral disc”, “aging/pathology”, “degeneration”. Inclusion criteria were as follows: published papers of all study designs, including the lumbar region of the spine, in Polish or English, with no years of restriction. Articles describing pathological or symptomatic spine disorders were excluded.

Discussion

Intervertebral disc

The intervertebral disc complex includes nucleus pulposus, annulus fibrosus, and adjacent cartilaginous endplate of the vertebral bodies [4,7,8]. The centrally located nucleus consists of 70-90% water. The remaining part consists of proteoglycans, type II collagen fibres, and loosely distributed chondrocytes located mainly in the endplate cartilage area [8]. At a young age, the nucleus makes up most of the intervertebral disc, while the annulus is narrow and well demarcated. Changes associated with the aging of the body are mainly due to a decrease in the hydration of the disc, especially the nucleus pulposus [7,8] – the number of viable chondrocytes decreases, which translates into a decrease in the production of water-binding proteoglycans. At the same time, the relative collagen content increases. In MRI studies, this is manifested by a decrease in signal intensity in T2-weighted images, blurring of the boundary between the nucleus pulposus and the annulus fibrosus, and a decrease in disc height [9,10]. The above changes, characteristic of the progressive age-related degeneration of intervertebral discs, were graded by Pfirr­mann on a five-grade classification, which is presented in Table 1 and Figures 1-5.
Foltz et al. suggest that the Pfirrmann classification might be subjective, thus failing to adequately capture primary degenerative changes. They propose quantitative MRI techniques using T2-intensity signal mapping among intervertebral discs, which allow evaluation of the initial stages of degeneration [11]. Moreover, some authors described diurnal variations in the intervertebral disc T2 signal and bulging that may lead to misinterpretation of intervertebral disc degeneration (IDD) [12,13]. Danielson and Willen reported a change in dural cross-sectional area while applying an axial load to the spine [14]. In another study, Pfirrmann et al. also described age-related changes in the shape and volume of the intervertebral discs on lumbar spine MRI in 70 asymptomatic volunteers with no back pain. They measured several parameters of intervertebral discs, including height, volume, and convexity index. Their study demonstrated that despite the lack of degenerative changes, the volume and height of discs decrease with age, and their shape becomes less convex [15]. Yang et al. analysed MRIs of 65 asymptomatic volunteers to evaluate the applicability of the distance between the nucleus’ centre weighted by the signal intensity and the geometric centre as a parameter of nucleus pulposus homo­geneity [16]. In their study, in 85.8% of studied intervertebral discs the weighted centre was located poste­rior to the geometric centre, and the distance between these centres on the longitudinal axis was significantly shorter in homogenous than in heterogenous discs.
Asymptomatic intervertebral disc degenerative changes have been widely reported before [2,17-33]. Their occurrence increases with lower assessed lumbar level and is most frequent at L5/S1 [23-25,27,30,32,34]. Older age groups are associated with a higher prevalence of these abnormalities [2,23-25,27,29-32]; however, many studies have shown that IDD occurs in all age groups, even in ado­lescents or young adults, despite the absence of low back pain (LBP) [19,21,28,34-36]. Correlation between the appearance of IDD in imaging studies and back pain is not explicit [26,37]. Carragee et al. performed lumbar spine MR in 200 participants with no previous history of low back pain and monitored them in a 5-year observational study. In the case of a new severe low back pain (LBP) episode, subjects were assessed for new imaging tests. Out of 51 re-evaluated subjects, 43 (84%) had either unchanged MR or showed regression of baseline changes [38]. MR studies of symptomatic patients with IDD appearance should be evaluated considering other abnormalities that may cause pain, such as Modic changes [39], Schmorl’s nodes [40], or spinal stenosis [41,42]. Brinjikji et al. reviewed 33 studies assessing age-related changes of the spine in MR among asymptomatic populations. Their results are presented in Table 2. Degenerative changes of the intervertebral discs, decrease in their height, reduction of the intensity of the T2 signal, bulging, and the degeneration of intervertebral joints were documented in almost 90% of asymptoma­tic patients over 60 years old, which suggests that such changes, especially those that were diagnosed accidentally, might be considered as a natural part of the aging process and do not require therapeutic intervention. However, protrusion of the intervertebral disc and annular fissures did not show a significant increase in the incidence with the increase of the patient’s age, which may suggest that they are not part of the natural aging process [1].
The prevalence of degenerative features may also vary between different populations. Rajeswaran et al. studied lumbar spine MRI studies of 98 asymptomatic junior elite tennis players in the mean age of 18 years [43]. They found that 89.7% of them had facet joint arthropathy, 62.2% had disc degeneration, and 30.6% had disc herniation. However, in most cases, these pathological findings were mild (84.5%, 76.2%, and 86.1%, respectively). Ranson et al. described 61% cases of IDD among 36 asymptomatic professional fast bowlers in cricket, and one-third of them were classified as Pfirrmann IV or V. Moreover, 81% of them had stress reaction or stress fracture of pars interarticularis. However, other authors reported no significant increase in degenerative changes in middle-aged male athletes [44] or young female dancers [45]. Likewise, according to Weinreb et al., pregnancy was not associated with an increased IDD rate [46]. Savage et al. studied 149 men of 5 different occupations (including car production workers, ambulance men, and office staff) and found no correlation between lumbar spine MRI appearance and employment group or their low back pain history. Gwak et al. studied 10 patients with asymptomatic intervertebral disc degeneration and 10 healthy asymptomatic participants, and evaluated core muscle cross-sectional area (CSA) in MRI scans and their function using strength sensor, various endurance tests, and stability test. Subsequently, they compared those results between asymptomatic participants with and without intervertebral disc degeneration in magnetic resonance imaging [47]. They found no significant difference in CSA and muscle function between both groups, which might help in future differentiation between accidental findings concerning intervertebral discs from pathologies that cause patients syndromes; however, further studies should be performed.

Osteophytes

Vertebral osteophyte development is believed to be a common feature of various conditions, including aging, intervertebral disc degeneration, and obesity [48-51]. Many studies have focused on describing osteophyte formation patterns among spine regions related to various factors, including age, sex, BMI, race, and others. Several age-estimation systems evaluating vertebral osteophytes have been proposed [48,52,53]. Nathan studied the prevalence of vertebral osteophytes in the spine among 400 skeletons of people who died at between 11 and 105 years of age. The osteophytes of the anterior and lateral edges of the vertebral bodies occurred with a high frequency. In cadavers over 40 years, anterior and lateral osteophytes were found in 100% of cases. Posterior osteophytes, however, occurred in the minority of cases. Thus, it may indicate that osteophytes of the anterior and lateral edges of the vertebral bodies are formed as a result of the natural aging process. In contrast, osteophytes of the posterior edges of the vertebra should not be interpreted by a radiologist as a normal finding [51].
Kacar et al. retrospectively reviewed computed tomography images of the thoracolumbar spine of 564 adult individuals (279 males and 285 females; between 20 and 86 years of age) to assess the effect of age and sex on the severity and localization of osteophyte formation. They reported an increase in osteophyte score between 40 and 70 years of age in both sexes, while in the lower segments of thoracic spine, males showed a significantly higher frequency of osteophytes than females.

Bone marrow

The bone marrow is a cell-rich connective tissue located inside the bone marrow cavity. It consists of 2 components – hematopoietic (red) and fatty (yellow) marrow. MRI is a very sensitive method of bone marrow imaging due to its accurate distinction between the fat content of the yellow marrow and the haematopoietic content of red marrow [54,55]. The distinction between bone marrow types is best seen on T1-weighted and fat-suppressed sequences [55]. The yellow marrow has a high signal intensity similar to that of subcutaneous fat, while the red marrow has an intermediate signal that is less intense than subcutaneous fat, but more intense than the intervertebral disc or muscle tissue. At birth, nearly all bone marrow is haematopoietically active and consists exclusively of red bone marrow [55]. As the body matures, the red bone marrow gradually converts to yellow bone marrow [55,56]. Bone marrow conversion typically follows Neumann’s law, which means that it starts from the distal phalanges and then runs towards the proximal bones and to the axial skeleton. A mature pattern of bone marrow distribution is achieved at about 25 years of age [55,56]. At this age, the haematopoietic marrow is located in the axial skeleton, sternum, ribs, and a lesser amount in proximal parts of the humerus and femur. Above this age, there is a slow, gradual further transformation of the red marrow into yellow marrow [55].
From adolescence to old age, men show a gradual, steady increase in bone marrow fat content. For women, this increase is linear up to around 55 years of age [57]. Then, bone marrow conversion accelerates in women between 55 and 65 years old [57,58]. There is a rapid decrease in the concentrations of oestrogens in the perimenopausal period, which results in a change in the pattern of fat distribution in the body, increasing the amount of visceral fat and fat contained in the bone marrow [57]. The trend of the relatively high share of yellow bone marrow in men compared to women is reversed. Over the age of 60 years, the bone marrow fat content is about 10% higher in women [57]. Few studies have described common anatomic sites of red and yellow bone marrow in the spine [54,59,60]. Ricci et al. studied MRI images of the axial skeleton in patients without known bone marrow abnormality, who ranged in age from 6 months to over 70 years, including 70 examinations each of the skull, cervical spine, thoracic spine, lumbar spine, pelvis, and proximal femur.
They described 4 typical patterns of bone marrow distribution in the vertebral bodies:

Pattern 1 – high signal-intensity fatty marrow is seen confined to linear areas along the basivertebral veins (Figure 6);
Pattern 2 – band-like and triangular areas of fatty marrow are located peripherally (Figure 7);
Pattern 3a – multiple small areas of high intensity of fatty marrow (Figure 8);
Pattern 3b – multiple, relatively large, well-marginated areas of fatty marrow (Figure 9).
The frequency of occurrence of bone marrow distribution patterns was different depending on the age of the patient – in people under 30 years old, the most common was pattern 1, the frequency of patterns 2 and 3 increased with age [54].

Conclusions

MRI remains a useful tool in evaluating the aging features of the spine. Many abnormalities occur widely in older asymptomatic patients. Early signs of spine aging include intervertebral disc signal changes, disc height loss, and bulging. Osteophytes of anterior and lateral edges of vertebral bodies frequently develop in people over 40 years old, whereas those of posterior edges are seen in a minority of cases, which may indicate an abnormal condition. Bone marrow fat content increases gradually with age, with acceleration in women between 55 and 65 years old. Also, yellow and red marrow distribution follows characteristic patterns depending on the patient’s age. Knowledge of the prevalence of these changes is essential in the appropriate evaluation of spine imaging studies.

Conflict of interest

The authors report no conflict of interest.

References