Purpose: While varus inclination of proximal tibia and bone marrow lesion (BML) are important factor for understanding knee osteoarthritis (OA) etiology, the relationship between these wo factors remain unclear. This study aimed to investigate the effect of proximal tibial inclination and serum bone metabolic markers on bone marrow lesion (BML) in the Japanese general population with early knee osteoarthritis (EKOA). Methods: A total of 441 female volunteers who participated in the Iwaki Health Promotion Project in 2017 were enrolled. Participants without radiographic abnormalities were divided into normal and EKOA groups according to Luyten’s classification criteria for EKOA (Figure 1). In this study, proximal tibial inclination was assessed with reference to growth plate (GP). The GP-tibial plateau (TP) angle was defined as the angle between the GP and TP lines (a), whereas the GP-medial tibial plateau (MTP) angle was defined as the angle between the GP and MTP lines (b) The medial proximal tibial angle (MPTA), GP-TP angle, and GP-MTP angle were measured on standing anteroposterior radiographs of the knees. BML severity on T2-weighted fat-suppressed magnetic resonance imaging (MRI) was scored using the Whole-Organ MRI Score method. Serum levels of N-telopeptide of type I collagen (NTx), tartrate-resistant acid phosphatase-5b (TRACP-5b), bone-specific alkaline phosphatase, procollagen type I N-terminal propeptide, pentosidine, and homocysteine were assessed. Serum concentrations of bone metabolic markers and radiographic measurements were compared using the Wilcoxon rank sum test. To investigate the relationship among BML, proximal tibial inclination, and serum bone metabolic markers, multiple linear regression analyses were conducted with the total BML score as the dependent variable and with age, body mass index, MPTA, GP-MTP angle, GP-TP angle, and serum bone metabolic markers as independent variables in patients with and without EKOA. To estimate the predictive cut-off level of proximal tibial inclination for detecting BML, receiver operating characteristic (ROC) analysis was performed for the EKOA group. Finally, in order to evaluate the cut-off values, the participants were divided into three groups according to the cut-off values estimated from the ROC analysis—the neutral group, tibia vara (extra-articular deformity) group, and genu varum (intra-articular deformity) group. Serum bone metabolic marker levels were compared among these three groups using one-way analysis of variance and Tukey’s test. Of these 309 participants, 48 participants (16%) had EKOA. With reference to the GP line, the GP-TP angle was inclined at 3.6±1.1° and the GP-MTP angle was 9.9±2.6°.Tabled 1Clinical characteristics of the participantsGP not observedGP observedp valueTotalNon-OAEKOASample number13130926148nsAge (years)52.7±14.154.5±9.554.4±9.755.0±8.8nsHeight (cm)156.1±6.0157.4±6.0157.5±6.0156.8±6.1nsBMI (kg/m2)22.4±3.622.0±3.121.9±3.122.3±3.2nsMPTA (°)86.6±1.286.1±5.986.4±1.586.6±1.4nsGP-TP (°)3.6±1.13.6±1.23.6±0.9nsGP-MTP (°)9.9±2.69.9±2.69.8±2.6nsNTx (nM)15.0±5.114.9±3.815.0±3.814.5±3.7nsTRACP-5b (mU/dL)415±190422.4±185431±14393.5±27nsBAP (μg/L)13.8±5.013.0±8.913.1±4.99.8±4.3nsPINP (μg/L)51.0±19.647.4±18.447.8±1.545.9±2.6nsPentosidine (pmol/mL)27.4±9.630.5±24.329.3±16.837.1±47.2nsHomocysteine (nmol/mL)9.0±3.18.5±2.78.5±0.28.3±0.4ns Open table in a new tab Serum bone metabolic marker levels did not significantly differ between the non-OA and EKOA groups. The MPTA was negatively correlated with the GP-TP and GP-MTP angles (r=−0.253, p<0.001 and r=−0.369, p<0.001, respectively). The BML was detected in 64 participants (31%) out of 205 participants who were performed MRI (). The MPTA was smaller in participants with EKOA who had BML than in those who did not (p=0.020), and the GP-MTP angle was larger in participants with BML than in those without BML (p=0.018).Tabled 1Radiographic measurements between participants with and without BMLNon-OAp valueEKOAp valueBML (-)BML (+)BML (-)BML (+)MPTA86.4±1.486.3±1.8ns86.8±1.386.0±1.20.020GP-TP3.5±1.33.5±1.1ns3.5±1.03.9±0.7nsGP-MTP10.0±2.59.9±2.9ns9.4±2.711.1±2.00.018 Open table in a new tab Linear regression analysis revealed that age was positively associated with BML severity in overall participants. In the non-OA group, serum TRACP-5b level was positively associated with BML (β=0.236, p=0.047). Proximal tibial inclination (MPTA and GP-MTP angle) was associated with BML only in the EKOA group (β=−0.281, p=0.011 and β=0.262, p=0.046, respectively).Tabled 1Relationship among bone marrow lesion, tibial inclination, and serum bone metabolic markersParameterNon-OApEKOApββAge0.1340.0090.503<0.0001BMI−0.014ns−0.072nsMPTA−0.056ns−0.2890.011GP-MTP−0.053ns0.2620.046GP-TP0.082ns0.137nsTotal PINP0.052ns0.060nsTRACP-5b0.2360.0470.076nsBAP−0.004ns−0.039nsNTx−0.032ns−0.074nsPentosidine0.024ns−0.136nsHomocysteine−0.011ns−0.136ns Open table in a new tab The ROC analysis indicated that the cutoff values for detecting BML were 86.7° for the MPTA (AUC: 0.703, p=0.030) and 9.7° for the GP-MTP angle (AUC: 0.706, p=0.028). Based on these cutoff values, tibial bone morphology was divided into the following three groups: neutral (MPTA ≥86.7°). (a), tibia vara (extra-articular deformity: MPTA <86.7° and GP-MTP angle <9.7°) (b), and genu varum (intra-articular deformity: MPTA <86.7° and GP-MTP angle ≥9.7°) groups (c). In the neutral group, 55 (46%) participants out of 120 participants had GP-MTP angle ≥9.7°. The serum NTx level was significantly lower in the genu varum group than in the tibia vara group (p=0.023). The serum TRACP-5b level was significantly lower in the genu varum group than in the neutral group (p=0.012).Tabled 1Serum bone metabolic marker levels among proximal tibial inclination typesNeutral (MPTA ≥86.7°)Inclined (MPTA <86.7°)p valueTibia vara (MTP-GP <9.7°)Genu varum (MTP-GP ≥9.7°)Sample number12075114MPTA87.6±0.785.5±0.985.1±1.2<0.001GP-MTP9.1±2.47.7±1.512.2±1.7<0.001NTx14.9±3.515.9±7.913.8±3.90.023BAP13.1±0.412.5±0.511.7±0.4nsTRACP-5b421±193409±202351±1680.012PINP46.8±18.249.4±27.245.1±19.7nsPentosidine32.6±2.230.5±28.228.2±2.2nsHomocysteine8.4±2.58.6±3.08.3±2.5ns Open table in a new tab Conclusions: The GP was a reasonable reference for assessing TP inclination. The presence of BML was associated with greater TP inclination and higher bone formation activity, even in participants without radiographic abnormalities. Residual medial tibial inclination may potentially result in knee pain and symptoms in EKOA.
Read full abstract