Introduction- Locating the
hip center (HC) is an important part of any gait analysis study that entails
calculation of pelvi-femoral motion and hip muscle moments. Some investigators
has estimated its location from their general knowledge of external landmarks
(Eberhart and Inman, 1951; Harrington, 1976; Paul, 1965; Pedotti, 1977). Others
have used radiographs to precisely locate the HC (Crowninshield et al., 1978;
Ellis et al., 1979; Johnston et al., 1979). At least two investigators developed
methods to estimate HC location from certain palpable landmarks (Andriacchi
et al., 1980, 1982; Andriacchi and Strickland, 1983; Tylkowski et al., 1982).
Specifically, Andriacchi's group predicted that the HC would lie 1.5-2 cm directly
distal to the midpoint of a line between the pubic symphysis and the anterior
superior iliac spiners (ASIS) in a frontal plane projection, and directly medial
to the greater trochanter in the sagittal plane. Tylkowski's group predicted
the HC would lie 11% of the distance between the ASIS
Medial to, 12% distal to, and 21% posterior to the Asis. While we have previously
shown that the approaches of HC location estimation of Tylkowski's group and
Andriacchi's group are more accurate than 'freehand' external estimation (within
3.3 cm of the three-dimensional HC location (Tylkowski), and within 1.8 cm of
the frontal plane HC location (Andriacchi) 95% of the time in adults), our measurements
were taken from the bony landmarks on radiographs and probably underestimate
the actual errors associated with marker placement on living subjects with variable
amounts of overlying tissues (Bell et al., 1989).
Cappozzo (1984) described a method of HC estimation based on the premises that
(1) the thigh is a rigid body, and (2) the HC is the center of a sphere described
by the three-dimensional rotation of a point on that body, but reported no error
estimates. The purpose of this study was to estimate the accuracy with which
this rotational method could estimate HC location in a series of live subjects,
and compare that to the accuracy that could be obtained by the methods of Andriacchi'
group and Tylkowski's group in the same subjects.