B. Barden et C. Huttegger, Failure mechanisms in total hip and knee arthroplasty: A morphologic and radiologic study, MATER WERKS, 30(12), 1999, pp. 746-754
Failure mechanisms in hip and knee arthroplasty are controversial, as the s
ystem of bone and implant is rather complex due to permanent biological cha
nges, inhomogeneous Young's modulus and geometry of bone. Therefore, since
1989 534 failures were retrospectively analyzed in a clinical study: mean p
atients' age 67,9 years, range 21 - 91, cemented implants = 379, non-cement
ed = 155, group 1 = 286 acetabular revisions, group 2 = 165 femoral revisio
ns and group 3 = 83 knee revisions. Results: average survivorship of implan
ts (late loosening): group 1 = 10.2 years, group 1 = 12.1 years, group 3 =
7.3 years. The major failure mechanism of hip and knee implants in this stu
dy was revealed as loosening at the interface between bone and cement (poly
methylmethacrylate) or bone and implant: group 1 = 72 %, group 2 = 53 %, gr
oup 3 = 42 %. Other mechanisms seldom occurred: Debonding at the cement/pro
sthesis interface: group 1 = 4 %, group 2 = 12%, group 3 = 15 %. Crack prop
agation in the cement mantle: group 1 = 8 %, group 2 = 20 %, group 3 = 3 %.
Severe osteolysis due to granuloma: group 1 = 11 %, group 2 = 13 %, group
3 = 2 %. A high percentage of extensive bone defects was found: defect clas
sification according to Paprosky [20]: group 1: type 1 = 22,4 %, type 2 = 5
6,3 %, type 3 = 21,3 %, group 2: type 1 = 27.3 %, type 2 = 59,4 %, type 3 =
13,3 %. Defect classification according to Fitzek and Barden [10]: group 3
: type 1 = 44,6 %, type 2 = 41,0 %, type 3 = 14.5 %, type 4 = 0 %. While in
hip arthroplasty early loosening within five years after the preceding ope
ration was rare, half of the knee revisions had to be performed because of
early loosening due to the special complexity of knee arthroplasty. Partial
atrophy of the periprosthetic bone often occurred but could not be made re
sponsible for loosening. Loosening without local pain was noticed in 13,2 %
. There was close correlation between preoperative radio graphic analysis a
nd intraoperative Findings. In conclusion severe loss of bone stock without
clinical symptoms in loosening of hip and knee implants should be prevente
d by routine radiographic controls. Circumferential disintegration is most
common and the typical failure mechanism of hip and knee implants. Thus for
prophylaxis of loosening special emphasis should be laid on tight connecti
on at the interface between cement (or implant) and bone.