ABSTRACT
Background:
The Legg-Calvé-Perthesian disease (LCPD) which affects hips in childhood, represents an increased risk of developing hip arthrosis later in life. Modified mechanical bone properties as a consequence of femoral head osteonecrosis in childhood, may lead to growth problems and, consequently, deformities of the hip joint and pelvis in adulthood. Biomechanical hip analysis may provide additional insights into the mechanisms leading to avascular osteonecrosis in the hip and its subsequent consequences.
Methods:
We have analysed the biomechanical parameters of 141 hips that were subjected to LCPD in childhood, and distinguished between them and 114 contralateral hips without record of disease and taken as controls. Biomechanical parameters were determined by mathematical models for resultant hip force in one-legged stance and for contact hip stress, which use as an input the geometrical parameters assessed from anteroposterior radiographs.. Analysis took into account also the influence of the possible deformation of the articular sphere in frontal and axial plane on those parameters. Because there were no important influence of femoral head deformation with assumption of sphericity of articular sphere in different planes (different frontal and axial radii of articular sphere) found, we upgraded the mathematical model for stress to account for the deviation of the femoral head shape from spherical with assumption of ellipsoid deformation of articular surface (circular surface in frontal plane and ellipsoid in axial plane). We performed additional analysis of radiograms in both planes with estimation of possible influence of elipsoid deformation on biomechanical parameters. Additionaly we also performed analysis after subgrouping hips accordig to the size of femoral head developed after LCPD and looked for possible indicators responsible for higher risk for cox-arthrosis later in life.
Results:
Measurement of the deformation of the articular surface showed that some hips (in the test and in the control groups) are flattened, and others are extended in the anteroposterior direction, however, on average, they are mostly flattened. As expected, affected hips (test hips) were considerably (by 83%) and statistically significantly more flattened (p<10-3) with larger articular surface radius in the frontal plane, later being favourable as regards hip stress. However, all other biomechanical parameters that exhibited statistically significant differences (centre-edge angle, functional angle of load bearing area, position of the pole, index of hip stress gradient and its normalised value), were less favourable in the test population. No statistically significant differences in peak stress, its normalised value and size of the load bearing area were found between the test and the control groups. We compared 90 hips that developed frontal radius of the articular sphere r larger than 2.65 cm with 165 normal-sized articular surfaces (r smaller than 2.65 cm). Among hips with r>2.65 cm, 72 (80%) pertained to the group of affected hips and 18 (20%) to the control group. Hips with r>2.65 were considerably and statistically significantly more flattened (by 125%), had smaller centre-edge angle and functional angle of the weight bearing area, and had a positive hip stress gradient index but there was no statistically significant difference between both subgroups of hips in peak contact stress and the normalised contact hip stress. We have also focused on the population of 69 affected hips (49% of hips pertaining to the test group) which later developed normal-sized articular surfaces r <2.65 in comparison with 96 contralateral normal-sized articular surfaces r<2.65. The affected hips were considerably (by 55%) and statistically significantly more deformed than the contralateral normal sized hips. Simillary, there was no statistically significant difference between both subgroups of hips in peak contact stress and the normalised contact hip stress but all other biomechanical parameters showed more unfavourable values in hips after LCPD. Analysis of dysplastic and non-dysplastic hips according to the HIPSTRESS method, with hips being dysplastic when hip stress gradient index is positive (Gp>0) and non-dysplastic when hip stress gradient index is negative (Gp<0), showed 26% of all hips being dysplatic (n=66). Grouping hips as dysplastic to the acknowledged criterion of centre-edge angle (<20°) showed 43 dysplastic hips representing 17% of all hips. Dysplastic hips regarding HIPSTRESS and centre-edge angle were mostly from tested group of hips (85% and 93% resopectively). All biomechanical parameters were considerably and statistically significantly more favourable in nondysplastic hips according to both classifications. Dysplastic hips also had considerably and statistically significantly larger radius of the articular sphere in the frontal plane and more deformed articular surface than normal hips. Dyspalstic group according centre-edge angle had for 19% (p<10-13) larger radius and were for 89% (p<10-3) more deformed and dysplastic group according HIPSTRESS criterion had for 17% (p<10-13) larger radius of the articular sphere and were for 86% (p<10-4 more deformed. Correlations between parameters that are not connected functionally within the model showed statistically significant correlations. Strong and positive correlation between the radius of the articular surface in the frontal plane and the deformation of the head were seen, mostly as the contribution of the affected hips although singular unaffected hips which developed large articular surfaces fit well into the correlation pattern. Also, most of the femoral heads are flattened in the anteroposterior direction ( is positive). Amidst the hips that were extended in the anteroposterior direction, the hips affected by the Perthes Disease in childhood are, on average, more deformed. Most of the unaffected hips lie well below the threshold for hip dysplasia (Gp=0) according HIPSTRESS method while, regardless of the deformation parameter , the affected hips are close to the borderline or over the borderline.
Conclusion:
Results confirm previous indications that head enlargement after the Legg-Calve-Perthes Disease compensates the values of hip stress and supports the suggested role of hip gradient index in the hip developement. Furthermore, it was found that an increased risk for coxarthritis development after the disease is secondary to concomitant hip dysplasia, with considerable and statistically significantly lower centre-edge angle and unfavourable distribution of stress.
|
