Conceptualization, J.Z., Z.Q. and A.Z.; data curation, S.L. and L.F.; formal analysis, J.C., Q.L., B.Y. and L.W.; investigation, L.F. and X.Y.; methodology, J.C., Q.L., L.W., Z.X. and A.Z.; resources, J.Z.; software, S.L. and Z.X.; Supervision, Z.X., J.Z., Z.Q. and A.Z.; validation, Z.Q. and A.Z.; visualization, S.L.; writing—original draft, J.C. and Q.L.; writing—review and editing, L.W., Z.X., Z.Q. and A.Z. All authors have read and agreed to the published version of the manuscript.

Figure 1. Measurement of the lateral trochlea inclination (LTI) and trochlear depth. (A) The line through the sulcus of the medial epicondyle and the prominence of the lateral epicondyle (SEA) and a line subtended from the surface of the lateral trochlea (LTF) are shown. The red dotted line is parallel to the SEA, and LTI is the angle between the SEA and LTF. (B) The highest points of the lateral/medial trochlear facets and the deepest point of the trochlea were identified, showing their vertical distance to the posterior condylar reference line (PCRL), L, M, and G, respectively. Trochlea depth is calculated according to the following formula: [(L + M)/2 − G].

Figure 1. Measurement of the lateral trochlea inclination (LTI) and trochlear depth. (A) The line through the sulcus of the medial epicondyle and the prominence of the lateral epicondyle (SEA) and a line subtended from the surface of the lateral trochlea (LTF) are shown. The red dotted line is parallel to the SEA, and LTI is the angle between the SEA and LTF. (B) The highest points of the lateral/medial trochlear facets and the deepest point of the trochlea were identified, showing their vertical distance to the posterior condylar reference line (PCRL), L, M, and G, respectively. Trochlea depth is calculated according to the following formula: [(L + M)/2 − G].

Figure 2. Tibial tubercle-Roman arch (TT-RA) distance (A) The posterior condylar reference line (PCRL), a line parallel to the PCRL and tangent to the Roman arch (the red dotted line), and a line that passes through the tangent point (the black dot) and perpendicular to the PCRL (RAL) are shown. (B) Axial CT image with the completed contact of the tibial tendon, showing the center of tibial tuberosity (TT). (C) The line perpendicular to the PCRL and parallel to RAL is drawn through the TT (TTL), and the distance between the RAL and TTL is defined as the TT-RA distance.

Figure 2. Tibial tubercle-Roman arch (TT-RA) distance (A) The posterior condylar reference line (PCRL), a line parallel to the PCRL and tangent to the Roman arch (the red dotted line), and a line that passes through the tangent point (the black dot) and perpendicular to the PCRL (RAL) are shown. (B) Axial CT image with the completed contact of the tibial tendon, showing the center of tibial tuberosity (TT). (C) The line perpendicular to the PCRL and parallel to RAL is drawn through the TT (TTL), and the distance between the RAL and TTL is defined as the TT-RA distance.

Figure 3. Four segments of femur are selected (A) The femoral head–neck axis is drawn through the center of the femoral head and neck (Line a). (B) The femur–lesser–trochanter line is drawn through the center of the femur and the midpoint of the lesser trochanter (Line b). (C) Line c is tangent to the posterior aspect of the femur on the slice just above the attachment of the gastrocnemius. (D) A line through the sulcus of the medial epicondyle and the prominence of the lateral epicondyle (SEA) is shown. The angles formed between Line a and Line b, Line b and Line c, Line c and SEA, and Line a and SEA are regarded as the neck torsion, mid torsion, distal torsion, and total femoral torsion, respectively.

Figure 3. Four segments of femur are selected (A) The femoral head–neck axis is drawn through the center of the femoral head and neck (Line a). (B) The femur–lesser–trochanter line is drawn through the center of the femur and the midpoint of the lesser trochanter (Line b). (C) Line c is tangent to the posterior aspect of the femur on the slice just above the attachment of the gastrocnemius. (D) A line through the sulcus of the medial epicondyle and the prominence of the lateral epicondyle (SEA) is shown. The angles formed between Line a and Line b, Line b and Line c, Line c and SEA, and Line a and SEA are regarded as the neck torsion, mid torsion, distal torsion, and total femoral torsion, respectively.

Figure 4. (A) Patella tilt is the angle formed between a line through the patella width (PWL) and a line tangent to the bilateral posterior condyles (PCRL). The red dotted line is parallel to the PCRL. (B) Congruence angle is defined as the angle between the sulcus angle bisector (dotted line b) and the line connecting the deepest portion of the sulcus and the apex of the patella ridge (dotted line c). (C) Patellar displacement is the distance between two vertical lines of the line tangent to the bilateral anterior condyles (line a), the one passing through the highest portion of the medial facet (line b), and the other one passing through the medial edge of the patella (line c).

Figure 4. (A) Patella tilt is the angle formed between a line through the patella width (PWL) and a line tangent to the bilateral posterior condyles (PCRL). The red dotted line is parallel to the PCRL. (B) Congruence angle is defined as the angle between the sulcus angle bisector (dotted line b) and the line connecting the deepest portion of the sulcus and the apex of the patella ridge (dotted line c). (C) Patellar displacement is the distance between two vertical lines of the line tangent to the bilateral anterior condyles (line a), the one passing through the highest portion of the medial facet (line b), and the other one passing through the medial edge of the patella (line c).

Figure 5. The results of the unadjusted binary regression model, showing the Odd Ratio and 95% confidence interval. (A) The dendrogram of TT-RA/TEW, TT-RA distance, trochlear depth, and LTI. (B) The dendrogram of patellar displacement, congruence angle, and HKA. (C) The dendrogram of patellar tilt, tibial torsion, and distal torsion. TT-RA, tibial tubercle-Roman arch distance; TEW, transepicondylar width; LTI, lateral trochlea inclination; HKA, hip–knee–ankle angle.

Figure 5. The results of the unadjusted binary regression model, showing the Odd Ratio and 95% confidence interval. (A) The dendrogram of TT-RA/TEW, TT-RA distance, trochlear depth, and LTI. (B) The dendrogram of patellar displacement, congruence angle, and HKA. (C) The dendrogram of patellar tilt, tibial torsion, and distal torsion. TT-RA, tibial tubercle-Roman arch distance; TEW, transepicondylar width; LTI, lateral trochlea inclination; HKA, hip–knee–ankle angle.

Figure 6. Receiver operating characteristic (ROC) curves of the anatomic parameters. (A) The ROC curves of the TT-RA distance, HKA angle, and TT-RA/TEW. (B) The ROC curves of patellar tilt, congruence angle, and patellar displacement. TT-RA, tibial tubercle-Roman arch distance; HKA, hip–knee–ankle angle; TEW, transepicondylar width.

Figure 6. Receiver operating characteristic (ROC) curves of the anatomic parameters. (A) The ROC curves of the TT-RA distance, HKA angle, and TT-RA/TEW. (B) The ROC curves of patellar tilt, congruence angle, and patellar displacement. TT-RA, tibial tubercle-Roman arch distance; HKA, hip–knee–ankle angle; TEW, transepicondylar width.

Table 1. Demographic characteristics of the participants.

Table 1. Demographic characteristics of the participants.

Study Group
(n = 17)Control Group
(n = 34)p ValueSex, n n.s Female1224  Male510 Age, median (IQR), y19.2 (7.0)19.0 (7.0)n.sSkeletal maturity, n%35.3%35.3%n.sSide of knee, n 0.569 a Left915  Right819 Interval time, month46 (IQR, 18)--

Table 2. Inter-and intra-observer reliability of each parameter, showing the ICC and 95% confidence interval.

Table 2. Inter-and intra-observer reliability of each parameter, showing the ICC and 95% confidence interval.

Inter-Observer 1Inter-Observer 2Intra-Observer 1Intra-Observer 2Dejour classification a0.91 (0.85, 0.95)0.94 (0.88, 0.97)0.93 (0.85, 0.96)0.87 (0.74, 0.94)LTI0.88 (0.83, 0.93)0.85 (0.77, 0.92)0.86 (0.79, 0.91)0.90 (0.85, 0.94)Trochlear depth0.78 (0.66, 0.85)0.80 (0.68, 0.89)0.82 (0.77, 0.90)0.80 (0.69, 0.88)TT-RA distance0.91 (0.83, 0.93)0.88 (0.84, 0.92)0.83 (0.75, 0.91)0.95 (0.92, 0.98)Total femoral torsion0.93 (0.88, 0.96)0.94 (0.89, 0.97)0.85 (0.74, 0.92)0.93 (0.89, 0.97)Neck torsion0.95 (0.92, 0.98)0.94 (0.91, 0.96)0.96 (0.94, 0.98)0.97 (0.95, 0.98)Mid torsion0.93 (0.88, 0.97)0.91 (0.85, 0.94)0.90 (0.82, 0.94)0.92 (0.87, 0.98)Distal torsion0.95 (0.91, 0.98)0.94 (0.90, 0.97)0.90 (0.83, 0.94)0.96 (0.94, 0.98)Knee joint rotation0.93 (0.88, 0.97)0.89 (0.79, 0.94)0.93 (0.85, 0.97)0.96 (0.92, 0.99)Tibial torsion0.96 (0.94, 0.99)0.93 (0.90, 0.96)0.97 (0.95, 0.99)0.92 (0.86, 0.95)PCA0.94 (0.89, 0.96)0.93 (0.85, 0.97)0.87 (0.78, 0.92)0.96 (0.93, 0.99)TEW0.96 (0.93, 0.99)0.92 (0.83, 0.96)0.91 (0.80, 0.95)0.94 (0.87, 0.99)Patellar height0.90 (0.83, 0.96)0.89 (0.81, 0.94)0.82 (0.74, 0.91)0.91 (0.82, 0.98)HKA angle0.95 (0.87, 0.98)0.92 (0.83, 0.97)0.87 (0.75, 0.94)0.92 (0.88, 0.97)Patellar tilt0.88 (0.81, 0.93)0.86 (0.77, 0.94)0.91 (0.84, 0.95)0.93 (0.84, 0.97)Congruence angle0.87 (0.81, 0.95)0.91 (0.85, 0.96)0.92 (0.83, 0.96)0.89 (0.82, 0.94)Patellar displacement0.93 (0.85, 0.97)0.91 (0.85, 0.94)0.88 (0.78, 0.93)0.92 (0.86, 0.97)

Table 3. Differences in the parameters between the study group and the control group via the Wilcoxon rank-sum test, showing the median and IQR.

Table 3. Differences in the parameters between the study group and the control group via the Wilcoxon rank-sum test, showing the median and IQR.

Study GroupControl Groupp ValueLTI, °11.9 (9.4–13.6)13.9 (11.1–16.2)<0.001Trochlear depth, mm3.8 (2.6–4.2)4.1 (3.3–4.5)<0.001TT-RA distance, mm24.1 (20.9–25.7)19.5 (17.9–22.2)<0.001Total femoral torsion, °19.2 (14.7–22.7)18.5(16.6–21.3)0.551Neck torsion, °23.3 (20.2–25.3)22.3 (19.1–25.4)0.375Mid torsion, °27.1 (26.3–29.5)27.2 (24.6–29.8)0.302Distal torsion, °14.4 (11.1–16.6)14.6 (10.9–18.3)0.057Knee joint rotation, °9.5 (7.3–12.1)9.4 (7.4–12.6)0.295Tibial torsion, °32.1 (28.7–35.5)30.8 (23.1–32.7)0.001PCA, °4.4 (2.2–5.4)4.4 (2.8–4.7)0.867TEW, mm72.7 (70.2–74.6)73.1 (70.3–74.9)0.122Patellar height1.32 (1.22–1.39)1.34 (1.12–1.50)0.863HKA angle, °1.9 (1.3–2.8)0.8 (−0.5–1.8)<0.001Patellar tilt, °23.9 (20.9–33.1)17.8 (13.5–22.4)<0.001Congruence angle, °10.2 (9.0–13.0)9.2 (6.5–10.5)<0.001Patellar displacement, mm9.9 (9.3–11.0)7.6 (6.4–9.9)<0.001TT-RA/TEW, %32.7 (29.8–34.9)27.6 (24.6–31.3)<0.001Dejour classification
(A/B/C/D)1/2/7/74/12/12/6<0.001 a

Table 4. Diagnostic ability of the parameters for subsequent contralateral LPD.

Table 4. Diagnostic ability of the parameters for subsequent contralateral LPD.

VariablesAUCp ValueCutoff ValueSensitivity, %Specificity, %LTI0.6690.051---Trochlear depth0.5930.285---TT-RA distance0.7270.00920.082.344.1Distal torsion0.5480.583---Tibial rotation0.5450.603---HKA angle0.797<0.0011.382.470.6Patellar tilt0.7030.01919.088.250.0Congruence angle0.7250.0098.782.451.1Patellar displacement0.817<0.0019.288.264.7TT-RA/TEW0.7410.00629.583.361.7