The history and development of the Exeter hip

Original design
The first total hip arthroplasty to be carried out in Exeter was performed in April of 1965 at the Princess Elizabeth Orthopaedic Hospital, utilising the McKee-Farrar metal-on-metal implant.  Over the next 4½ years, more than 350 of these arthroplasties were implanted at the Princess Elizabeth.  Aseptic loosening became such a problem that it was clear that the continued use of this device would not be justified.  Rightly or wrongly, the metal-on-metal articulation was regarded as part of the problem with the McKee-Farrar and, therefore, the adoption of the Ring metal-on-metal device was not considered.  By 1969, most of the surgeons at the Princess Elizabeth had adopted the posterior approach for total hip arthroplasty and did not wish to take up the lateral approach via a trochanteric osteotomy, reasonably regarded at that time as an essential part of the operative technique for the use of the Charnley low-friction arthroplasty.  There were no other options available in 1969  and therefore a decision was taken to develop a new implant specifically intended for use with cement through the posterior approach to the hip.  The Exeter Hip was the outcome and was developed jointly at the Princess Elizabeth Orthopaedic Hospital & the Dept. of Engineering Science of the University of Exeter throughout 1969 and early 1970, being introduced in to clinical practice at the Princess Elizabeth in November 1970.

The Femoral Component:
The design of the original Exeter femoral component is shown in Fig. 1. The device was double tapered throughout its length and was unique (for 1969) in being totally devoid of any form of collar.

In follow-up studies of the cemented McKee-Farrar and Thompson femoral components in Exeter, an x-ray finding that was regularly seen was resorption of the neck of the femur beneath the collar of the implant.  This was so common that it led inevitably to the conclusion that in vivo, as opposed to in vitro and theoretical studies 1,2,3,4, the collar could not regularly load the proximal cut surface of the neck of the femur.  As a consequence, the decision was made to remove entirely the collar from the new femoral component.

A double-taper shape was adopted for the stem because it was believed that this would represent the optimal shape for the extrusion of acrylic cement dough in to the endosteal surface of the femur during stem insertion.  This was an intuitive decision at the time that has since been supported by experimental evidence 5,6. The taper was carried right up to the base of the neck on the anterior & posterior surfaces of the stem.

A stem offset of 44mms was chosen as being appropriate for the majority of hips requiring replacement (Fig.1).

The original Exeter stems were manufactured from the rather ductile stainless steel EN58J (also used for the original polished flat-back Charnley stem).  The British Standard 7 for the use of this alloy in the manufacture of surgical implants demanded that such implants  should have a polished surface.  The fact that the surface finish of the device might be functionally important was not at that time considered and the reason that the surface finish of the stem was polished was simply the demands of the British Standard for that particular alloy.

In 1969, no one knew what the optimum head size would eventually turn out to be.  As a consequence, a compromise between the 41mm head of the McKee-Farrar stem and the 22.5mm head of the Charnley LFA was adopted with an head size of 29.75mm. (Fig. 1)  The stem was originally made available in two sizes known as standard and lightweight.

Pre-clinical Testing:
In 1969, the pre-clinical testing of implants was in its infancy.  However, since by the standards of contemporary  (i.e., 1969) femoral components, the new stem design was unorthodox, a series of comparative loading tests was performed in the Department of Engineering Science at the University of Exeter in order to establish whether the proposed new stem design was as capable of transmitting load in to the femur as other contemporary stems. Samples of all the femoral components in contemporary use in the UK in 1969 were cemented into cadaveric femora and their displacement under load compared with the new stem. These tests were relatively unsophisticated, not least because they only included axial loading as, at that time, there was no knowledge of the importance of torsional forces as contributing to the loading of the femoral head in vivo. Nevertheless, the outcome of these tests suggested that the new stem was at least as effective as other contemporary stems as far as axial load transmission was concerned (Fig. 2).

Instrumentation:
The femoral instrumentation was unsophisticated.  A taper pin reamer, curette and modified long-handled Capener spinal gouge were used for femoral preparation.  Trial stems were available, each of which had a series of holes from anterior to posterior in the base of the neck of the device, corresponding to the positions of etched marks on the necks of the implantable stems. A pin that would rest on the cut surface of the femoral neck was passed through one of the holes in base of the neck of the trial stem and so allow an appropriate position of the stem to be estimated at trial reduction for the accurate reproduction of leg length.  A femoral introducer was utilised that held the stem with its long axis exactly in the line of the handle of the introducer so as to enable the surgeon to introduce the device down the middle of the long axis of the femoral canal.  Retractors were developed specifically to improve the exposure of the upper end of the femur and achieve appropriate opening of the base of the neck and superior surface of the trochanteric region to allow central insertion of the stem.

The Acetabular Component:
The cup was manufactured from ultra-high molecular weight polyethylene.  As a 29.75mm head diameter had been selected, the problem arose of how to obtain a thick layer of polyethylene in the projected wear path of the head without going to an excessively large outside diameter of cup.  This was addressed by making the inside of the cup eccentric with respect to the outside (Fig. 3a), a design feature later recognised 8  as inappropriate and changed in 1988. The vertical skirt (Fig.3b) on the lateral aspect of the face of the cup was intended to protect against dislocation, and in practice, was usually placed posterolaterally rather than superior by rotating the cup posteriorly through 30-450 in the long axis of the cup introducer before insertion. This defeated the object of the eccentricity because it brought a thin layer of polyethylene into part of the projected wear path of the femoral head. Guides on the introducer helped to orientate the cup appropriately in abduction and anteversion..