We investigated the effect of simulated muscle loading and the contrib
ution of the radial head to stability of the Norway elbow in 6 cadaver
s using an electromagnetic tracking device. The kinematics of the elbo
w after implantation of the prosthesis were similar to the intact elbo
w in their valgus-varus orientation, however, the forearms were slight
ly externally rotated, probably due to a small amount of external rota
tion of the humeral components at the time of implantation. The valgus
-varus laxity limit of the implants were greater than in the intact sp
ecimens averaging 8.0 and 5.6 degrees, respectively. Simulated muscle
loading stabilized both the intact and the Norway elbows. Excision of
the radial head after implant arthroplasty increased their valgus-varu
s laxity, suggesting that preservation of the radial head may be indic
ated if it is not too severely involved by the underlying disease proc
ess. The laxity permitted by the prosthesis articulation is greater th
an that measured after implantation of the Norway arthroplasty. This s
uggests that the prosthesis may behave as an unconstrained arthroplast
y. This should minimize the stress experienced by the bone-cement inte
rface and may reduce the incidence of loosening. The laxity of the elb
ows after joint arthroplasty were only slightly greater than normal, p
ossibly explaining the low incidence of prosthesis dislocation which h
as been observed with clinical use.