TL;DR
This study presents the largest radiographic analysis to date of lower limb coronal morphology in a Chinese population without knee osteoarthritis (KOA). Using advanced AI for precise measurements, it reveals significant differences in limb alignment between left and right sides, genders, and age groups, providing crucial normative data for KOA prediction and treatment.
What Problem Does This Paper Address?
Knee osteoarthritis (KOA) is a major cause of pain and disability, and its development is strongly linked to the mechanical alignment of the lower limbs. Understanding the "normal" or non-pathological variations in this alignment is critical for predicting who might develop KOA, diagnosing it early, and planning effective surgical interventions like total knee arthroplasty (TKA). While previous studies have examined these parameters, there's a significant gap in large-scale, detailed analyses specifically for Chinese populations, who may have distinct morphological characteristics. Furthermore, traditional measurement methods are time-consuming and prone to variability. This paper aims to fill this gap by providing comprehensive, AI-driven radiographic analysis of lower limb coronal morphology in a large cohort of Chinese individuals without KOA, exploring variations across laterality, gender, and age.
Methods
The researchers conducted a cross-sectional retrospective analysis of long-leg anteroposterior radiographs (LLRs) from 407 Chinese individuals (814 lower limbs) without KOA. A key innovation in this study is the use of an in-house developed artificial intelligence (AI) software for the meticulous assessment of several coronal morphological parameters. These parameters include:
- Hip-knee-ankle angle (HKAA): Overall alignment of the lower limb.
- Weight-bearing line ratio (WBLR): The position of the mechanical axis relative to the tibial plateau.
- Joint line convergence angle (JLCA): The angle of the knee joint line in the coronal plane.
- Mechanical lateral-proximal-femoral angle (mLPFA): Femoral alignment in the proximal region.
- Mechanical lateral-distal-femoral angle (mLDFA): Femoral alignment in the distal region.
- Mechanical medial-proximal-tibial angle (mMPTA): Tibial alignment in the proximal region.
- Mechanical lateral-distal-tibial angle (mLDTA): Tibial alignment in the distal region.
The AI software offers advantages in terms of accuracy and efficiency compared to manual measurements. The study then statistically analyzed these parameters to identify differences between left and right limbs, males and females, and younger (<50 years) versus older (≥50 years) age groups.
Datasets
The study utilized a dataset of 814 lower limb LLRs from 407 Chinese patients without KOA, collected between December 2020 and October 2022. This dataset represents a new, large-scale collection of radiographic data specifically for the Chinese non-KOA population, enabling the establishment of population-specific normative values.
Key Insights and Ideas
- AI-Driven Precision: The use of AI for radiographic measurement is a significant methodological advancement, promising greater accuracy and efficiency for large-scale studies.
- Subtle Asymmetries Matter: Even in a healthy population, there are statistically significant differences in limb alignment between the left and right sides, particularly in the joint line convergence angle (JLCA). This suggests that perfect bilateral symmetry might not be the norm and has implications for using the contralateral limb as a perfect template for surgical correction.
- Gendered Morphology: Significant gender-based differences exist in most coronal parameters, with males tending to have more varus alignment (lower HKAA, JLCA, mMPTA, mLDTA, and WBLR) compared to females. This highlights the need for gender-specific reference values.
- Age-Related Changes are Specific: While most parameters remain stable with age, the mechanical lateral-proximal-femoral angle (mLPFA) consistently increases with age. This suggests that specific aspects of hip and proximal femur morphology change over time.
- Clinical Relevance for TKA: The findings provide crucial normative data for Chinese populations, which can help clinicians better identify deviations from normal alignment that might predispose individuals to KOA or inform surgical planning for TKA.
Key Results
- Left-Right Differences: Significant differences were found in HKAA (left vs. right: 178.2° vs. 178.6°) and JLCA (left vs. right: 1.8° vs. 1.4°), with the left side being more varus.
- Gender Differences: Males generally exhibited more varus alignment than females across several parameters, including HKAA (177.9° vs. 179.0°), JLCA (1.5° vs. 1.8°), mMPTA (87.1° vs. 88.1°), and WBLR (38.7% vs. 43.6%).
- Age Differences: Only mLPFA showed a significant increase with age (younger vs. older: 90.1° vs. 93.4°).
- Subgroup Analysis: In younger individuals, males had smaller HKAA, JLCA, mMPTA, and WBLR, and larger mLDFA and mLPFA than females. In older individuals, males had lower HKAA, JLCA, mMPTA, mLDTA, and WBLR compared to females.
Limitations and Future Work
The authors acknowledge several limitations:
- Patient Cohort: Participants were recruited from hospital databases, not necessarily representing a perfectly healthy volunteer population.
- Rotation Influence: While efforts were made to exclude excessive rotation, its subtle influence on measurements cannot be entirely eliminated.
- Confounding Conditions: Patients with spinal deformities, neuromuscular disorders, or different loading patterns on their lower limbs were not explicitly excluded, which could affect coronal morphology.
- Bilateral Loading: Differences in loading between the two lower limbs can influence JLCA and HKAA, and this was not controlled for.
Future work could focus on longitudinal studies to track changes over time, investigate the impact of these morphological variations on KOA progression more directly, and further explore the intriguing gender-specific age-related trends observed in mLDFA. Establishing Chinese-specific standards for lower limb abnormality based on gender is also recommended.
Summary updated