Exploring mean kurtosis in MR diffusion kurtosis imaging for early detection of lumbar spine degeneration: a systematic review

Introduction

Degenerative diseases are widespread, affecting a significant portion of the population. Lumbar spine degeneration is a prevalent condition that significantly impacts adults, causing chronic low back pain (CLBP) and reduced mobility.¹ The World Health Organization (WHO) has classified CLBP as one of the major causes of disability globally, contributing substantially to both health care and economic burdens.² Degeneration changes in the intervertebral disc space impact the structural integrity, leading to the bulging of the nucleus pulposus (NP) and the annulus fibrosus (AF) tearing. These changes lead to spinal stenosis and reduce the stability of the spine.³ Consequently, this results in decreased intervertebral disc height and spinal abnormalities, impairing spinal functionality and restricting motion. Degenerative changes in the spine are diagnosed at more advanced stages, when they become clinically apparent, limiting the effectiveness of the treatment and negatively patient outcomes. Early diagnosis is essential, as it allows for more effective interventions and facilitates the development of personalized, targeted therapies, thereby improving clinical outcomes and enhancing the overall quality of life.^4,5

Magnetic resonance imaging (MRI) is currently the gold standard in diagnosing degenerative spine diseases, where T1 and T2 weighted sequences are commonly used in detecting structural abnormalities.⁶ Conventional imaging sequences are limited to macroscopic diagnostic changes, and therefore unable to detect subtle microstructural changes. This limitation has driven the development of more advanced MRI applications such as Diffusion Weighted Imaging (DWI) and its application such as Diffusion tensor Imaging (DTI) and DKI.^7,8

DKI estimates kurtosis-tensor-specific parameters including MK, axial kurtosis (AK), radial kurtosis (RK), mean diffusivity (MD), axial diffusivity (AD), and radial diffusivity (RD). DKI parameter maps, such as MK, are directly derived from the dMRI data and often have implausible values without any additional processing. Empirical findings from biological tissues and fundamental physical principles indicate that MK should maintain a positive value and remain below 3.^9–11 Many visually implausible voxels exhibit MK values that fall outside the typical range of 0–3. However, many of these implausible voxels remain within the empirically defined range, yet still appear unusually dark or bright compared to their surrounding.^12,13 MK is derived from DKI, an advanced MRI technique that extends Diffusion Tensor Imaging (DTI) by accounting for non-Gaussian water diffusion in biological tissues. Unlike DTI, which assumes unrestricted diffusion, MK captures microstructural complexity by measuring the degree of diffusion heterogeneity caused by collagen fibers, cellular membranes, and extracellular matrix components. In the intervertebral disc, degeneration leads to proteoglycan loss, increased collagen cross-linking, and fiber disorganization, causing higher MK values. Similarly, in the annulus fibrosus (AF), disruptions in fiber integrity alter diffusion patterns. These properties make MK a sensitive biomarker for early degenerative changes, detecting microstructural alterations before they become apparent on conventional MRI.^14–16

In addition to its applications in spinal imaging, DKI has demonstrated broad diagnostic utility across various fields, including neuroimaging, oncology (liver, breast, renal, and prostate cancers), musculoskeletal, and abdominal imaging. Its ability to detect early microstructural changes has made it a crucial tool for accurate diagnosis, prognostication, and treatment planning.^17–22

DKI and its parameter MK are key techniques for detecting early microstructural changes in the spine. This review aims to evaluate MK’s role in detecting the early diagnosis of degenerative spine disease and assess its effectiveness in improving patient outcomes.

Methods

Results

Study selection and quality assessment

The systematic search initially identified 7290 potential studies, of which only 3 met the inclusion criteria for this review ( Figure 1). All included studies were published between 2010 and 2024. The QUADAS tool quality assessment revealed scores ranging from 11 to 14. Items 9, 10, and 11 of the QUADAS tool, about the clarity of the reference standard execution, the translation effects of the reference standard without index test results, and the translation of index test results without reference standard information, respectively, were identified as areas of uncertainty in many of the included studies.

Figure 1. PRISMA Flowchart of Study Selection A flowchart depicting the systematic review process for evaluating the role of MK in lumbar spine degeneration.

The initial database search identified 7290 studies, of which 3 were included in the final review following eligibility criteria.

Discussion

The systematic review was conducted following PRISMA guidelines in which a total of 7260 articles were screened and three studies were included in the review based on their relevance.

The findings of the review emphasize the role of MK as a biomarker for detecting early degenerative changes in the lumbar intervertebral disc. MK, derived from DKI, extends beyond DTI by focusing on the non-Gaussian diffusion properties of water molecules, offering a deeper understanding of tissue complexity and microstructural integrity. Lumbar degeneration is marked by progressive loss of disc hydration, a reduction in proteoglycan content, and collagen fiber degradation. MK sensitively captures these early-stage macrostructural changes, quantifying the degree of water diffusion restriction, which is influenced by the heterogeneity and complexity of tissue microstructure. In a healthy intervertebral disc, the NP is highly hydrated, allowing for unrestricted diffusion. However, in degenerative discs, the loss of proteoglycans and increased collagen fiber cross-linking lead to higher diffusion heterogeneity, which in turn increases MK values.^28–30

The systematic review findings suggested the variability of the MK in the NP and AF in the healthy and degenerative lumbar spine. Zeng et al.,²⁵ reported a significant increase in the MK for both NP and AF in the degenerative spine. They suggested that the MK could be a useful biomarker for the early microstructural changes in the spine. Similarly, Li et al.,²⁶ stated that there were limited changes in MK values of the Healthy population selected in their study. Despite this, the degenerative population in the NP region of the discs showed minimal change. In contrast, Takano et al.,²⁷ reflected in their research on higher baseline MK values, with lesser in degenerative disease. These inconsistencies likely reflect the differences in the parameters, study populations, and methodological approach, highlighting the need to standardize the imaging protocol to enhance the compatibility of the findings across the studies.

The review further analyzed the sensitivity and specificity which indicate that the diagnostic performance of the DKI varied across the studies where Zeng et al.,²⁵ demonstrated moderate sensitivity and specificity. In contrast, Li et al.,²⁶ demonstrated higher sensitivity and lower specificity, suggesting that DKI may be highly sensitive to early degenerative changes but may face challenges in specificity.

MK demonstrated it ability to detect the early microstructural changes in the spine, while T2- T2-weighted MRI remains the gold standard for assessing disc hydration but it cannot detect the microstructural changes.³¹ Similarly, T2 mapping and relaxometry techniques provide the ability to detect proteoglycan loss but it is limited to assess the complexity of the fiber tract integrity in the AF.^32,33

Although this review primarily focused on lumbar degeneration, the applicability of MK extends to other spinal regions. For instance, studies on cervical spondylotic myelopathy (CSM) have shown significantly lower MK values in stenotic segments, suggesting that MK is not only sensitive to disc degeneration but also to gray and white matter disruptions caused by chronic compression. Notably, MK values in the gray matter correlate with Japanese Orthopaedic Association (JOA) scores, reinforcing its potential role as a clinical indicator of neurological impairment.^34–36

The clinical implications of MK in spinal pathology are significant. First, MK can facilitate early diagnosis, potentially allowing for preventive interventions before severe structural damage occurs. Second, MK may serve as a biomarker for monitoring disease progression and guiding treatment decisions in conservative vs. surgical management. Lastly, its role in detecting microstructural damage in neural tissues suggests that MK may be useful in assessing post-surgical recovery or evaluating regenerative therapies such as platelet-rich plasma (PRP) or stem cell injections. This review emphasizes the importance of early detection and management of degenerative spine in line with Sustainable Development Goal 3 (good health and well-being). There is concern about this condition globally, especially among aging populations, which gradually reduces quality of life and increases health care costs. DKI plays a vital role in early diagnosis, of the degenerative spine and hence, thereby reducing healthcare costs, improving patient outcomes, promoting global health equity, and improving individual quality of life.

Conclusion

The systematic review helps in identifying the diagnostic importance of the advanced MRI applications such as DKI and its parameters MK which played a vital role in the early diagnosis of the degenerative spine and subtle microstructural change that occurs in the intervertebral disc space of the lumbar spine. The review focused on the effectiveness of the MK in the detection of early degenerative in the degenerative lumbar spine. Integrating MK into routine diagnostic workflows could significantly enhance the accuracy of detecting lumbar spine degeneration, enabling timely interventions. This advancement has the potential to reduce healthcare costs, improve patient outcomes, and contribute to global health objectives aimed at promoting better quality of life through early and effective management of degenerative spine conditions.

Ethics and consent

Ethical approval and consent were not required.