Cementless straight, tapered rectangular cross section titanium steins have
been used by the senior author since 1979. Thirty-four stems retrieved pos
tmortem, after between 10 days and 15.2 years in situ, were studied morphom
etrically and histologically. Nineteen steins were first generation (1979-1
986), with an incomplete sagittal taper and a mean surface roughness of 1 m
um (Ra 1.23 +/- 0.3 mum); 15 stems were second generation after 1986, with
a full biplane taper and a mean surface roughness of 4 mum (Ra 4.14 +/- 0.3
6 mum). Implant surface bony coverage was determined morphometrically in 10
segments of the stems, and expressed as a bone implant contact index. Hist
ologically, there were no differences between implants with different level
s of roughness. Morphometrically, the first-generation steins showed signif
icant differences in coverage (distal > proximal); second-generation stems
had a more uniform pattern. Stems retrieved early after arthroplasty had a
mean bone implant contact index of 10%. The mean bone im-plant contact inde
x showed attainment of maximum coverage by 5 years after arthroplasty, with
out additional apposition or loss thereafter. Patients younger than 65 year
s at arthroplasty bad similar bone implant contact indices to patients 65 y
ears or older; coverage in the six patients 80 years of age at retrieval di
d not differ from the rates in the other patients. Morphometry was able to
provide objective evidence of design change effects. No differences in cove
rage were found in terms of times in situ, patient age at arthroplasty and
at retrieval, and degree of stem surface roughness.