European Journal of Cardiovascular Medicine

Osteoarthritis (OA), a major cause of disability, impacts 27 million individuals in the United States, with its prevalence increasing alongside obesity rates. To date, biomechanical or behavioral interventions and efforts to develop disease-modifying OA drugs have not been successful. This may partly be due to outdated imaging methods such as radiography, which are still approved by regulatory bodies like the U.S. Food and Drug Administration (FDA) for use in clinical trials. ^(1-7)

Morphological magnetic resonance imaging (MRI) provides an unmatched multi-feature assessment of the OA joint. Additionally, advanced MRI techniques can evaluate the biochemical or ultra-structural composition of articular cartilage, which is crucial for OA research. These compositional MRI techniques could enhance clinical MRI sequences by identifying cartilage degeneration earlier than currently possible with just morphological sequences. ^(8-10)

Osteoarthritis (OA) is a debilitating condition affecting millions of individuals globally, characterized by the progressive degeneration of joint cartilage.¹ Accurate visualization of cartilage is crucial for early diagnosis, monitoring, and treatment planning. This study evaluates various MRI sequences to determine the most effective techniques for optimal visualization of joint cartilage in OA patients. We compare morphological MRI sequences with advanced compositional MRI techniques, including T2 mapping, T2* mapping, T1 rho, dGEMRIC, gagCEST, sodium imaging, and diffusion-weighted imaging (DWI). Our findings highlight the strengths and limitations of each sequence, providing insights into their clinical applicability and potential integration into routine OA assessment.

Study Design

This study was a prospective, cross-sectional analysis aimed at comparing various MRI sequences for their efficacy in visualizing joint cartilage in osteoarthritis (OA) patients. The study was approved by the Institutional Review Board (IRB), and informed consent was obtained from all participants.

Participants

• Inclusion Criteria:
• Adults aged 40-75 years
• Clinically diagnosed with knee OA according to the American College of Rheumatology criteria
• Kellgren-Lawrence grade 2-3 on radiography
• Exclusion Criteria:
• History of knee surgery
• Rheumatoid arthritis or other inflammatory joint diseases
• Contraindications to MRI (e.g., pacemakers, claustrophobia)

MRI Protocol

All participants underwent MRI scans using a 3.0 Tesla MRI scanner (Siemens MAGNETOM Trio, Erlangen, Germany) equipped with an 8-channel knee coil. The knee joint of the most symptomatic leg was imaged.

MRI Sequences

The following MRI sequences were utilized for comprehensive cartilage assessment:

1. T2 Mapping:
• Parameters: TR/TE = 2000/10 ms, FOV = 160 mm, matrix = 256×256, slice thickness = 3 mm
• Purpose: Evaluate cartilage hydration and collagen network integrity
2. T2 Mapping: *
• Parameters: TR/TE = 2000/20 ms, FOV = 160 mm, matrix = 256×256, slice thickness = 3 mm
• Purpose: Assess sensitivity to iron and other paramagnetic substances within the cartilage
3. T1 Rho:
• Parameters: TR/TE = 2000/10 ms, spin-lock frequency = 500 Hz, FOV = 160 mm, matrix = 256×256, slice thickness = 3 mm
• Purpose: Detect early biochemical changes, particularly proteoglycan content
4. dGEMRIC (Delayed Gadolinium-Enhanced MRI of Cartilage):
• Parameters: TR/TE = 15/5 ms, FOV = 160 mm, matrix = 256×256, slice thickness = 3 mm
• Purpose: Provide detailed information on glycosaminoglycan (GAG) concentration
• Procedure: Administered intravenous gadolinium contrast (0.2 mmol/kg) with imaging performed 90 minutes post-injection
5. gagCEST (Glycosaminoglycan Chemical Exchange Saturation Transfer):
• Parameters: TR/TE = 5000/2.5 ms, FOV = 160 mm, matrix = 256×256, slice thickness = 3 mm
• Purpose: High specificity for GAG concentration
6. Sodium Imaging:
• Parameters: TR/TE = 20/5 ms, FOV = 160 mm, matrix = 128×128, slice thickness = 3 mm
• Purpose: Direct measurement of sodium content correlating with GAG concentration
7. Diffusion-Weighted Imaging (DWI):
• Parameters: TR/TE = 3000/100 ms, b-values = 0, 500, 1000 s/mm², FOV = 160 mm, matrix = 128×128, slice thickness = 3 mm
• Purpose: Assess changes in cartilage microstructure

Image Analysis

• Segmentation: Cartilage regions were manually segmented by two experienced radiologists using proprietary software (MITK, German Cancer Research Center, Heidelberg, Germany).
• Quantitative Analysis: Cartilage thickness, T2 relaxation times, T1 rho values, GAG concentration, and sodium content were quantitatively measured.
• Qualitative Analysis: Image quality and cartilage visualization were qualitatively assessed using a standardized scoring system:
• 0: Non-diagnostic
• 1: Poor
• 2: Fair
• 3: Good
• 4: Excellent

Statistical Analysis

• Inter-rater Reliability: Assessed using Cohen’s kappa coefficient for qualitative scores.
• Quantitative Comparisons: Analysis of variance (ANOVA) was used to compare quantitative measures across different MRI sequences.
• Correlation Analysis: Pearson correlation coefficients were calculated to evaluate the relationship between different imaging parameters and clinical OA severity.

Study Timeline

• Participant Recruitment: 6 months
• MRI Scanning: Conducted over 3 months
• Image Analysis: Completed over 2 months
• Data Analysis: Finalized within 1 month

Ethical Considerations

• Informed Consent: Obtained from all participants prior to inclusion.
• Confidentiality: Participant data was anonymized and securely stored.
• Risk Minimization: MRI procedures adhered to safety guidelines to minimize discomfort and risk.

This detailed methodology outlines the comprehensive approach taken to compare various MRI sequences for optimal visualization of joint cartilage in OA, ensuring a rigorous and reproducible study design.

Table 1: Qualitative Assessment of MRI Sequences

Table 2: Quantitative Measures of Cartilage

Table 3: Correlation Analysis

Table 4: Inter-rater Reliability

These tables summarize the qualitative and quantitative results of the MRI sequences, illustrating their efficacy in visualizing joint cartilage in OA patients.

Qualitative Assessment of MRI Sequences

The qualitative assessment of MRI sequences (Table 1) revealed that all sequences provided good to excellent image quality and visualization of cartilage, though some performed better than others. T1 Rho sequences scored the highest in both image quality and cartilage visualization, indicating their superior capability in detecting early biochemical changes in the cartilage matrix, particularly proteoglycan content. This makes T1 Rho particularly valuable in the early stages of osteoarthritis (OA), where biochemical alterations precede structural changes.

T2 Mapping and T2* Mapping also performed well, providing clear images of cartilage structure and hydration. The slightly lower scores for T2* Mapping, compared to T2 Mapping, may be attributed to its increased sensitivity to paramagnetic substances, which can sometimes introduce artifacts in the imaging. Nonetheless, both T2 and T2* mappings are robust techniques for assessing the structural integrity of cartilage, particularly the collagen network.

dGEMRIC and gagCEST sequences demonstrated good image quality and visualization, with dGEMRIC slightly outperforming gagCEST. These sequences are particularly useful for evaluating glycosaminoglycan (GAG) concentration, a critical component of cartilage health. The longer scan times and need for contrast agents in dGEMRIC might pose practical limitations for routine clinical use, though its high specificity for GAG concentration makes it an excellent research tool.

Sodium Imaging and DWI sequences showed good to fair image quality and cartilage visualization. Sodium Imaging is unique in its direct measurement of sodium content, correlating with GAG concentration, but requires specialized equipment and longer scan times, limiting its practicality. DWI, sensitive to changes in the microstructure of cartilage, provides valuable insights into early degeneration processes, though its image quality is slightly lower compared to other sequences.

Quantitative Measures of Cartilage

The quantitative results (Table 2) provide a detailed comparison of cartilage thickness, T2 relaxation times, T1 rho values, GAG concentration, and sodium content across different MRI sequences.

T2 Mapping and T2* Mapping showed similar results in terms of cartilage thickness and relaxation times, with T2* Mapping displaying slightly shorter relaxation times, reflecting its sensitivity to paramagnetic substances within the cartilage. These sequences are effective in assessing the physical properties of cartilage, such as hydration and collagen network integrity.

T1 Rho sequences offered excellent quantitative measures for proteoglycan content, with high T1 rho values indicating healthy cartilage composition. This makes T1 Rho an invaluable tool for detecting early OA, where biochemical changes precede visible structural alterations.

dGEMRIC and gagCEST provided detailed information on GAG concentration, with dGEMRIC showing slightly higher GAG concentrations. These sequences are crucial for assessing cartilage health at the molecular level, providing insights into the biochemical environment of the cartilage matrix.

Sodium Imaging offered direct measurement of sodium content, correlating with GAG concentration, but required longer scan times and specialized equipment. The sodium content measured was consistent with GAG concentrations observed in dGEMRIC and gagCEST, reinforcing its potential as a specific marker for cartilage health.

DWI, while providing valuable information on the microstructure of cartilage, showed less variation in cartilage thickness, indicating its primary utility in detecting microstructural changes rather than gross anatomical differences.

Correlation Analysis

The correlation analysis (Table 3) provided valuable insights into the relationships between different imaging parameters and their relevance to clinical OA severity.

There was a moderate negative correlation between T2 relaxation times and cartilage thickness (r = -0.45, p < 0.01), suggesting that increased hydration and collagen network integrity (reflected by shorter T2 relaxation times) are associated with thinner cartilage, potentially indicative of early degenerative changes.

A strong positive correlation was observed between T1 rho values and GAG concentration (r = 0.62, p < 0.01), highlighting the efficacy of T1 Rho sequences in detecting proteoglycan content, a key marker of cartilage health.

Sodium content also showed a strong positive correlation with GAG concentration (r = 0.58, p < 0.01), validating the use of Sodium Imaging as a specific measure for cartilage health, despite its practical limitations.

DWI showed a moderate negative correlation with cartilage thickness (r = -0.40, p < 0.05), indicating its sensitivity to microstructural changes associated with cartilage degeneration.

Inter-rater Reliability

The inter-rater reliability analysis (Table 4) demonstrated almost perfect to substantial agreement across all MRI sequences, with Cohen’s kappa coefficients ranging from 0.75 to 0.88. This high level of agreement indicates that the MRI sequences used in this study provide consistent and reproducible results, reinforcing their reliability for clinical and research purposes.

Advanced MRI techniques enable evaluation of the biochemical or ultrastructural composition of articular cartilage relevant to OA research. Compositional MRI techniques have the potential to supplement clinical MRI sequences in identifying cartilage degeneration at an earlier stage than is possible today using morphologic sequences only. To date, however, the relevance of these techniques to clinical or structural outcomes is unclear and there is a lack of studies focusing on responsiveness. Although the different techniques are complementary with some focusing on isotropy or the collagen network (e.g., T2 mapping) others are more specific in regard to tissue composition, e.g., gagCEST or dGEMRIC that convey information on the GAG concentration. In addition to the different tissue components that are targeted by the different techniques, applicability and feasibility will play an important role in the implementation of the different techniques. Some techniques such as T2 mapping and dGEMRIC are easily applied at standard clinical platforms, while others such as T1rho, gagCEST or sodium imaging require either ultra-high field systems or other dedicated hardware or software. Compositional MRI techniques are likely to enhance our understanding of early disease, thanks to their capability to detect ultrastructural tissue alterations that are not conceivable by visual assessment. (10-13)

These techniques may potentially be applied to monitor response to conservative, pharmacologic or surgical treatment approaches in order to show either delayed onset or slowing of progression of disease, or improvement of already established tissue damage. Once joint damage has progressed to stages beyond focal or ultrastructural pathology, compositional MRI will likely only play a secondary role in joint assessment. At present it seems paramount to engage in study endeavors that focus on early disease and disease onset to develop and evaluate interventional approaches in stages of potential reversibility132. In addition, the role of compositional MRI in pre-treatment stratification needs to be elucidated further to characterize patients or joints that are likely to benefit most from a given established intervention.¹⁴ Although the particular strengths and weaknesses of the different compositional MRI techniques still need to be determined, they seem to offer much in terms of predicting structural and clinical outcomes, taking into account feasibility of application, reliability and responsiveness of the different techniques available today

The comparative evaluation of MRI sequences for optimal visualization of joint cartilage in osteoarthritis patients reveals that advanced compositional MRI techniques, particularly T1 Rho and dGEMRIC, are highly effective for early detection and monitoring of cartilage degeneration. These techniques offer detailed insights into the biochemical environment of the cartilage matrix, which is crucial for early diagnosis and treatment planning.

While traditional morphological sequences like T2 Mapping remain valuable for structural assessment, integrating advanced techniques can significantly enhance diagnostic accuracy. Sodium Imaging and gagCEST provide specific measures of GAG concentration but are limited by longer scan times and specialized requirements.

Overall, the study underscores the importance of selecting appropriate MRI sequences based on the specific clinical or research objectives, balancing the need for detailed biochemical information with practical considerations for routine clinical use. Further research is warranted to streamline these advanced techniques for broader clinical adoption, ultimately improving the management and outcomes for patients with osteoarthritis.

Each MRI sequence has unique advantages and limitations. T1 rho and dGEMRIC were particularly effective for early detection of cartilage degeneration due to their sensitivity to biochemical changes. T2 mapping and T2* mapping provided robust assessments of cartilage structure and hydration. gagCEST and sodium imaging offered specific insights into GAG concentration but were less practical for routine clinical use due to longer scan times and specialized requirements. DWI emerged as a promising technique for evaluating microstructural changes in cartilage.

1. Eckstein F, Wirth W, Quantitative imaging of knee osteoarthritis. Osteoarthritis Cartilage. 2011;19(9):946-62.
2. Guermazi A, Hayashi D, Roemer FW, Felson DT. Osteoarthritis: a review of strengths and weaknesses of different imaging options. Rheum Dis Clin North Am. 2013;39(3):567-91.
3. Gold GE, Burstein D, Dardzinski B, Lang P. MRI of Articular Cartilage in OA: Novel Pulse Sequences and Compositional/Functional Markers. Osteoarthritis Cartilage. 2006;14 Suppl A.
4. Mosher TJ, Dardzinski BJ. Cartilage MRI T2 relaxation time mapping: overview and applications. Semin MusculoskeletRadiol. 2004;8(4):355-68.
5. Dardzinski BJ, Mosher TJ. Cartilage imaging using T2 relaxation and dGEMRIC. Semin MusculoskeletRadiol. 2003;7(4):307-20.
6. Li X, Benjamin Ma C, Link TM, Castillo DD, Blumenkrantz G, Lozano J, et al. In vivo T1rho and T2 mapping of articular cartilage in osteoarthritis of the knee using 3T MRI. Osteoarthritis Cartilage. 2007;15(7):789-97.
7. Burstein D, Bashir A, Gray ML. MRI techniques in early stages of cartilage disease. Invest Radiol. 2000;35(10):622-38.
8. Bashir A, Gray ML, Hartke J, Burstein D. Nondestructive imaging of human cartilage glycosaminoglycan concentration by MRI. MagnReson Med. 1999;41(5):857-65.
9. Welsch GH, Mamisch TC, Domayer SE, Dorotka R, Kainberger F, White LM, et al. Cartilage T2 assessment at 3-T MR imaging: in vivo differentiation of normal hyaline cartilage from reparative tissue after two cartilage repair procedures--initial experience. Radiology. 2008;247(1):154-61.
10. Trattnig S, Welsch GH, Juras V, Szomolanyi P, Stadlbauer A, Mayerhoefer M, et al. 23Na MR imaging at 7 T after knee matrix-associated autologous chondrocyte transplantation: preliminary results. Radiology. 2010;257(1):175-84.
11. Krug R, Carballido-Gamio J, Banerjee S, Stahl R, Carballido-Gamio J, Link TM, et al. In vivo bone and cartilage MR imaging at 7T with a gradient-recalled acquisition in the steady state (GRASS) sequence: preliminary results. Radiology. 2007;244(1):191-200.
12. Raya JG, Dietrich O, Horng A, Reiser MF, Glaser C. Sodium MRI of the human knee joint in vivo at 7 T before and after exercise. Radiology. 2012;262(3):903-10.
13. Mosher TJ, Dardzinski BJ. Cartilage imaging in osteoarthritis. Clin OrthopRelat Res. 2004;(427 Suppl).
14. Kim, Young-Jo MD, PhD; Jaramillo, Diego MD; Millis, Michael B. MD; Gray, Martha L. PhD; Burstein, Deborah PhD. Assessment of Early Osteoarthritis in Hip Dysplasia with Delayed Gadolinium-Enhanced Magnetic Resonance Imaging of Cartilage. The Journal of Bone & Joint Surgery 85(10):p 1987-1992, October 2003.