Clustering of Knee Joint Types Affected by the Degenerative-Dystrophic Process

Introduction

Degenerative joint diseases are the most frequent pathological conditions among musculoskeletal system disorders. The most common form of the degenerative-dystrophic process is osteoarthritis, which affects more than 500 million people worldwide, accounting for over 7% of the global population [1–6]. Degenerative joint diseases lead to long-term loss of work capacity and remain the primary cause of disability among working-age individuals. In recent decades, there has been a steady increase in the prevalence of osteoarthritis. Between 1990 and 2019, the number of cases of degenerative joint diseases worldwide increased by 48% [6]. According to estimates by the Centers for Disease Control and Prevention (CDC, 2006), this figure is expected to rise by 50% globally by 2040 [5].

It is known that osteoarthritis affects 9.6% of men and 18% of women over the age of 60 [3]. More than 40% of people over the age of 65 suffer from manifestations of degenerative-dystrophic joint disorders, and more than 80% of people over the age of 75 show signs of the disease [1, 7].

In the structure of degenerative joint diseases, the proportion of primary gonarthrosis reaches 83% [2, 4, 5, 8]. Severe symptoms of knee osteoarthritis are observed in 10–38% of elderly individuals [6].

Degenerative joint diseases are a heterogeneous group of pathological conditions, the etiology and pathogenesis of which remain poorly understood. Metabolic, biomechanical, and genetic factors, as well as age, obesity, and injuries, are considered leading proven pathogenetic risk factors [1, 3, 6, 9, 10]. The heterogeneity of etiological factors and clinical manifestations of degenerative-dystrophic diseases complicates the understanding of pathogenesis, classification, and development of targeted treatment strategies [6, 11, 12]. Various classifications have been proposed, depending on the primary pathophysiological mechanisms, biochemical and genomic profiles, clinically significant individual risk factors, disease stage, lesion location and degree of inflammation [6, 10].

Indeed, in the study by F.R.T. Nelson (2018), four phenotypes of knee osteoarthritis were identified based on meta­bolic factors (dyslipidemia, hypertension, obesity, and diabetes), mechanical characteristics (joint morphology, alignment, previous trauma, medical history), associated musculoskeletal disorders (adjacent joint and spinal involvement, neuromuscular diseases, osteoporosis), structural joint elements (synovial membrane, meniscus, articular cartilage, patellar fat pad, bone sclerosis, and bone cysts), and pain localization [13]. In the study by Nelson A.E. et al. (2022), six biclusters were presented, which the authors considered as potential phenotypes of knee osteoarthritis [4]. In the work of Z. Huang et al. (2024), using unsupervised machine learning methods, the researchers identified four phenotypes of clinical progression of knee osteoarthritis, taking into account differences in symptoms of stiffness and pain, functional limitations, and patients’ dietary patterns [3]. Approaches to phenotype clustering for patients with degenerative knee joint diseases were also presented by D. Demanse et al. (2023) and Z. Huang et al. (2023) [9, 12].

Despite the large number of studies conducted on this topic, the segmentation of the heterogeneous population of patients with degenerative knee joint diseases and the determination of predictors of the formation of osteoarthritis phenotypes remain unsolved tasks [9, 14].

In addition to the heterogeneity of clinical symptoms, a high variability in the morphology of the femur and tibia has been demonstrated in patients with degenerative-dystrophic diseases [15, 16]. Significant differences in knee joint morphology have been scientifically confirmed, taking into account ethnic and gender characteristics, as well as height and body weight indicators [17–21].

The heterogeneity of research results emphasizes the need for the development of a simple and universal classification system for the morphological variants of knee joint in case of degenerative-dystrophic diseases.

One of the modern systems for assessing axial ratios of knee joint structures is the coronal plane alignment of the knee (CPAK) classification. CPAK is a simple and comprehensive tool for predicting prearthritic features of the knee joint morphology, which can be used for both healthy individuals and patients with osteoarthritis [19, 22]. A better understanding of knee joint phenotype variability enhances personalized approaches to knee arthroplasty and improves treatment effectiveness for patients in a heterogeneous population [13, 23].

However, it should be noted that performing orthoroentgenograms, which is a prerequisite for using CPAK, is not always feasible. Moreover, the classification does not consider morphological parameters defined in the sagittal plane, individual patient factors, and is unsuitable in cases of segmental extra-articular knee deformities [24].

We consider the development of a classification that takes into account the anatomical parameters of the femur and tibia as determined in both the frontal and sagittal planes, as well as the indicators of the structure and position of the patella, to be a promising direction for scientific research.

Structuring the variability of morphological features of the knee joint in cases of degenerative-dystrophic diseases will allow for the improvement of existing methods of corrective osteotomy of the femur and tibia, as well as techniques of unicompartmental and total replacement, taking into account the principles of personalized surgery.

Aim: to characterize the types of knee joints affected by the degenerative-dystrophic process by clustering of morphological parameters determined in the frontal plane.

Object and methods of research

The results of the examination of 100 patients with degenerative knee joint diseases were analyzed. The average age of the patients was 63.56±8.10 years.

Among the morphological parameters, the following were evaluated: indicators in the frontal plane, including the position of the anatomical axes of the femur, tibia, and the overall lower limb axis; intra-articular spatial relationships; morphometric measurements of the distal femur and proximal tibia; auxiliary criteria for assessing degenerative-dystrophic changes in the knee joint; and indicators in the sagittal plane, such as the morphology of the distal femur and proximal tibia, as well as the structure and positioning of the patella. This article presents the characteristics of the indicators of the anatomical axes of the femur, tibia, and the lower limb axis (FTA, aLDFA, aMPTA) in groups formed based on the morphological features of the knee joint structure in cases of degenerative-dystrophic diseases.

The anatomical lateral distal femoral angle (aLDFA) was defined as the angle between the line connecting the most distal parts of both femoral condyles and the anatomical axis of the femur, measured laterally. The angle between the line tangent to the proximal joint surface of the tibia and the anatomical axis of the tibia, measured from the medial side, was defined as the anatomical medial proximal tibial angle (aMPTA). The joint line convergence angle (JLCA) is the angle formed between the orientation lines of the joint surfaces of the distal femur and proximal tibia, the characteristics of which will be provided in future studies. The angle between the anatomical axes of the femur and tibia was defined as the femorotibial angle (FTA) [25].

Short weight-bearing anteroposterior radiographs of the knee were used for analysis.

Additionally, the predictive value of the defined parameters was analyzed regarding the identification of established morphotypes of knee joints in degenerative-dystrophic diseases.

The statistical analysis was performed using Statistica 13 software. Descriptive statistical methods were applied to evaluate and analyze quantitative data. For comparing indicators between independent groups, the non-parametric Kruskal — Wallis test was used, and the functional relationship between variables was assessed using Kendall’s τ rank correlation coefficient. The predictive value of the parameters in determining the morphological variant of knee joint structure affected by the degenerative-dystrophic process was evaluated using a logistic regression model, with the calculation of the odds ratio (OR) and 95% confidence intervals (CI). Statistical significance was set at p≤0.05.

Results and discussion

Based on the clustering of the determined morphometric indicators of the examined patients, 4 groups were formed, corresponding to 4 morphological variants of the knee joint in cases of degenerative-dystrophic diseases (Fig. 1). Morphotype I was identified in 21 (21.00%) patients, morphoty­pe II in 38 (38.00%) patients, morphotype III was observed in 29 (29.00%) individuals, and morphotype IV in 12 (12.00%) patients.

Figure 1. Dendrogram of patient distribution into groups considering the morphological features of the knee joint

To characterize the differences between the identified morphological variants of the knee joints, we described the FTA and the position of the anatomical axes of the femur and tibia (Table 1).

Table 1. Characteristics of the Position of the Anatomical Axes of the Femur, Tibia, and Lower Limb Axis Considering the Knee Joint Morphotype

The average FTA value in the examined patients was 177.09±5.57°. The mean FTA in patients with morphotype I was 169.76±1.48°, indicating a varus deviation of the lower limb axis. A varus deviation of the axis was also observed in patients with morphotype II, whose average FTA was 175.58±1.32°. In patients with morphotype III, the mean FTA was 180.14±1.49°, corresponding to a neutral position of the lower limb axis. In individuals with morphotype IV, the average value of the studied angle was 187.33±3.75°, indicating a valgus deviation of the axis. The comparison of the average FTA among the formed groups showed a statistically significant difference in the obtained values (p<0.0001) (Fig. 2).

Figure 2. Characteristics of the FTA values, considering the knee joint morphotype in degenerative-dystrophic diseases

The mean aLDFA value in the studied group of patients was 84.04±2.88°. The mean aLDFA was 84.57±3.26° in individuals with morphotype I, 85.18±2.38° in those with morphotype II, and 83.24±2.27° in patients with morphotype III. The lowest angle values were observed in patients with morphotype IV, at 81.42±2.97°. Analysis of the aLDFA values revealed a statistically significant difference among the groups (p=0.0009) (Fig. 3a).

Notably, in individuals with morphotypes I and II, who exhibited varus deviation of the lower limb axis, no cases were observed with aLDFA values ≤78°. In contrast, such values were found in 2 (6.90%) individuals with morphotype III and 4 (33.33%) patients from morphotype IV, with a statistically significant difference in frequency (p=0.0002).

aLDFA values corresponding to the reference normal range were observed in 44 (44.00%) patients in the group. A range of aLDFA values 79–83° was found in the majority of individuals with morphotypes III — 19 patients (65.51%) and IV — 6 patients (50.00%), which were associated with neutral and valgus positions of the lower limb axis, respectively. In the case of varus deviation of the axis, similar values were observed in 10 (26.32%) individuals with morphotype II and 9 (42.86%) patients with morphotype I. The difference in the frequency of aLDFA values 79–83° among the formed groups was statistically significant (p=0.02).

In 50 (50.00%) of the examined patients, aLDFA values ≥84° were observed. Elevated values of this parameter were recorded in the vast majority of patients with varus deviation of the lower limb axis, specifically in 12 (57.14%) individuals with morphotype I and 28 (73.68%) patients with morphotype II. In patients with neutral and valgus positions of the axis, aLDFA values ≥84° were recorded in 8 (27.59%) and 2 (16.67%) cases, respectively. A comparison of the obtained data, considering the established morphotypes, demonstrated a statistically significant difference in the indicators (p=0.0002).

Furthermore, it was established that an increase in the FTA is significantly associated with a decrease in the aLDFA values in individuals with degenerative knee diseases (τ=–0.29, p=0.00001), indicating an inward tilt of the anatomical axis of the femur and a shift of the axis tangent to the femoral condyles downward and medially.

The mean aMPTA value in the examined patients was 84.24±4.06°. In individuals with morphotype I, the average aMPTA was 79.57±2.50°, while in patients with morphotype II, it was 83.53±2.31°. For those with morphotype III, the mean aMPTA was 85.83±2.16°, and the highest values were found in patients with morphotype IV, at 90.83±3.10°. The differences among these values were statistically significant (p<0.00001) (Fig. 3b).

aMPTA values 85–90°, corresponding to the reference normal range, were observed in 38 (38.00%) patients. Most of these individuals belonged to morphotype III (21 patients, 72.41%), followed by morphotype IV (4 patients, 33.33%), and morphotype II (13 patients, 34.21%). No cases with aMPTA values within the reference normal range were observed in the morphotype I group. A comparison of the frequency of this parameter among the groups revealed a statistically significant difference (p<0.00001).

In the majority of the examined patients — 52 (52.00%), aMPTA values ≤84° were found. This included all patients with morphotype I — 21 (100.00%) and most patients with morphotype II — 25 (65.79%). Among individuals with morphotype III, who had a neutral axis position, these values were noted in 6 (20.69%) cases. No patients with morphotype IV exhibited such values. The comparison of aMPTA frequencies across the different knee joint morphotypes showed a statistically significant difference (p<0.00001).

aMPTA values ≥91° were observed in 10 (10.00%) of the examined patients, predominantly in those with morphotype IV (8 (66.67%) and in 2 (6.90%) patients with morphotype III. These values were not seen in patients with varus deviation of the lower limb axis. The difference in the frequency of aMPTA values ≥91° across the knee joint morphoty­pes was statistically significant (p<0.00001).

Furthermore, a significant increase in aMPTA values was observed with an increase in FTA (τ=+0.67, p<0.00001). This indicates that in cases of varus deviation of the lower limb axis, the anatomical axis of the tibia deviated inward, while in cases of valgus deviation, the tibial axis shifted outward.

The knee joint morphotype variants identified in patients with degenerative-dystrophic diseases are illustrated in Fig. 4–7.

Figure 4. X-ray of the right knee joint in the anteroposterior view of patient Z., 66 years old. Diagnosis: primary osteoarthritis of the right knee joint, stage III. Knee joint morphotype I. The measured angles are shown in the image

Figure 5. X-ray of the left knee joint in the anteroposterior view of patient M., 49 years old. Diagnosis: primary osteoarthritis of the left knee joint, stage III. Knee joint morphotype II. The measured angles are shown in the image

Figure 6. X-ray of the right knee joint in the anteroposterior view of patient K., 60 years old. Diagnosis: primary osteoarthritis of the right knee joint, stage II. Knee joint morphotype III. The measured angles are shown in the image

Figure 7 X-ray of the right knee joint in the anteroposterior view of patient K., 57 years old. Diagnosis: primary osteoarthritis of the right knee joint, stage III. Knee joint morphotype IV. The measured angles are shown in the image

The prognostic value of the positional parameters of the anatomical axes of the femur, tibia, and the axis of the lower limb in determining the knee joint morphotype in degenerative-dystrophic diseases was further evaluated, and the results are presented in Table 2.

Table 2. Prognostic value of the positional parameters of the anatomical axes of the femur and tibia in determining the knee joint morphotype

It was found that aLDFA values ≤78° in patients with gonarthrosis were significantly associated with higher chances of developing knee joint morphotype IV (OR=21.50, CI 3.32–139.31, p=0.0009).

Higher chances of developing morphotype III (OR=3.50, CI 1.39–8.76, p=0.006) were recorded in patients with degenerative-dystrophic knee joint diseases with aLDFA values of 79–83°. Moreover, it was proven that aLDFA values corresponding to the normative reference range significantly reduce the chances of developing morphotype II (OR=0.29, CI 0.12–0.72, p=0.005).

Significantly higher chances of developing knee joint morphotype II in patients with degenerative-dystrophic diseases were proven with aLDFA values ≥84° (OR=5.09, CI 2.07–12.54, p=0.0002). Conversely, increased aLDFA values were associated with significantly lower chances of developing morphotypes III (OR=0.26, CI 0.10–0.68, p=0.004) and IV (OR=0.17, CI 0.03–0.82, p=0.01).

When analyzing the prognostic value of the aMPTA parameter, it was shown that aMPTA values ≤84° significantly increase the chances of developing morphotype II (OR=2.49, CI 1.07–5.82, p=0.03) and decrease the risk of morphotype III (OR=0.14, CI 0.05–0.40, p=0.00004) in knee joints affected by degenerative-dystrophic diseases.

In patients with aMPTA values of 85–90°, which correspond to the reference normative values, there were higher chances of developing knee joint morphotype III in degenerative-dystrophic diseases (OR=8.34, CI 3.09–22.49, p=0.000006).

The presence of aMPTA values ≥91° in patients with gonarthrosis was associated with significantly higher chances of developing knee joint morphotype IV (OR=86.00, CI 13.27–557.33, p<0.00000001).

Conclusions

By clustering morphometric parameters determined in the frontal plane, we identified four morphological variants of the knee joint in degenerative diseases.

When comparing the positional indicators of the anatomi­cal axes of the femur, tibia, and lower limb as a whole, statistically significant differences were found among the groups formed based on the morphological structure of the knee joint in degenerative-dystrophic diseases: FTA (р<0.00001), aLDFA (p=0.0009), and aMPTA (p<0.00001).

It was proven that an increase in the FTA is significantly associated with a decrease in the aLDFA values in individuals with degenerative-dystrophic diseases of the knee joint (τ=–0.29, p=0.00001). This reflects an inward inclination of the anatomical axis of the femur and a downward and medial displacement of the axis tangent to the femoral condyles. Additionally, a significant increase in the aMPTA values was observed with a higher FTA (τ=+0.67, p<0.00001), indicating a medial deviation of the anatomical axis of the tibia in the presence of a varus deviation of the lower limb axis, and a lateral deviation in cases of valgus.

No significant influence of the femur, tibia, and lower limb axis position indicators on the prediction of morphotype I of the knee joint in degenerative-dystrophic diseases was established. However, there were significantly higher chances of developing morphotype II in the presence of aLDFA ≥84° (OR=5.09, CI 2.07–12.54, p=0.0002) and aMPTA ≤84° (OR=2.49, CI 1.07–5.82, p=0.03). Patients with degenerative-dystrophic knee diseases had higher chances of developing morphotype III with an aLDFA of 79–83° (OR=3.50, CI 1.39–8.76, p=0.006) and an aMPTA of 85–90° (OR=8.34, CI 3.09–22.49, p=0.000006), which aligned with the normative reference values. A significantly higher likelihood of develo­ping knee joint morphotype IV was observed in patients with osteoarthritis who had aLDFA values ≤78° (OR=21.50, CI 3.32–139.31, p=0.0009) and aMPTA values ≥91° (OR=86.00, CI 13.27–557.33, p<0.00000001).

References

Надійшла до редакції/Received: 24.03.2025
Прийнято до друку/Accepted: 27.03.2025