Reading Radiographs after Total Hip Arthroplasty

Hong Kong J Orthop Surg. 2006;10(1):1-9. Reading Radiographs after Total Hip Arthroplasty REVIEW ARTICLE Reading Radiographs after Total Hip Arthrop...
Author: Amy Cross
2 downloads 3 Views 184KB Size
Hong Kong J Orthop Surg. 2006;10(1):1-9.

Reading Radiographs after Total Hip Arthroplasty

REVIEW ARTICLE Reading Radiographs after Total Hip Arthroplasty Chiu KY, Yau WP, Tang WM, Ng TP Division of Joint Replacement Surgery, Department of Orthopaedics and Traumatology, The University of Hong Kong, Pokfulam, Hong Kong

ABSTRACT Indefinite follow-up following total hip arthroplasty is mandatory, and follow-up of these patients without radiographs is incomplete. In the follow-up radiographs, the surgeon may find clues on why the patient needed a total hip arthroplasty, can note details about the fixation and design features of the prosthesis, and can assess whether the total hip arthroplasty procedure was properly done. The surgeon can also discern the present status of the total hip arthroplasty — if the components are well fixed or loose; if the radiographs show any wear and osteolysis — and decide how to proceed with the follow-up, i.e., whether the patient can be safely asked to come back a year later for the next follow-up, or needs close radiographic monitoring or even early surgical intervention. Key Words: Arthroplasty, replacement, hip, Hip joint, Postoperative complications, Radiography, Review, Treatment outcome



! 

!"#$%&'()*+u 

!"#$!%#&!'()

 !"#$%&'()*+,-./0&12 X  !"#$%&'( X  !"#$%&'()*+,  !"#$%&'()*+,-./01!234 56789:;?@AB!567CD!EF !"#$ X !"#$%&'()*+ ,-./!0123456 X !"#$%&'()*+

INTRODUCTION Failures after total hip arthroplasty (THA) increase with the passage of time, so it is mandatory to follow up with THA patients indefinitely. Contrary to the customary school of thought that it is more important to look at the patient than the radiograph, for patients who have had THA, it is perhaps more important to look at the radiographs. Failures can be spotted much sooner with radiographs than by the onset of symptoms. For example, wear and osteolysis may occur without any symptoms, and mechanical loosening is often much progressed by the time the patient complains of pain. Follow-up of THA patients, without accompanying serial radiographs is, thus, both incomplete and insufficient. In this paper, we take another

look at the current method of radiographic evaluation and describe our approach to reading the radiographs of patients who have undergone THA (Table 1).

THE RADIOGRAPH Reading of radiographs after THA needs to be done in a systematic manner. Look at the label, note the patient’s identity and date of radiograph, and make sure it is the correct side of the radiograph. Assess the quality of the radiograph; note whether the whole prosthesis is included in the film and whether the pelvis or femur is properly rotated. A standard set should include an anteroposterior (AP) radiograph of the pelvis, with the beam being centred at the pubic symphysis, and AP and lateral radiographs of the replaced hip

Correspondence: Dr KY Chiu, Associate Professor and Chief, Room 216, New Clinical Building, Queen Mary Hospital, 102 Pokfulam Road, Pokfulam, Hong Kong. Tel: (852) 2855 4259; Fax: (852) 2817 4392; E-mail: [email protected] © 2006 Hong Kong Orthopaedic Association & Hong Kong College of Orthopaedic Surgeons.

1

Hong Kong J Orthop Surg.

Table 1

Chiu KY, Yau WP, Tang WM, et al

Checklist for reading radiographs of patients who have had total hip arthroplasty.

The radiograph Date, patient identity, side, quality The patient Clues about pre-existing hip disease (Figure 1) The prosthesis Fixation — cemented, cementless, hybrid (Figure 2) Design features of the socket and stem (Figure 3) Articulation The procedure Socket — lateral opening angle, horizontal and vertical positions, version (Figure 4) Stem — varus, neutral, or valgus; grades of cement mantle for cemented stem (Figure 3); canal fill for cementless stem The present status Socket — migration, radiolucent line (Figure 5) Cemented stem — definite, probable, or possible loosening (Figures 5 and 6) Cementless stem — bone ingrowth, stable fibrous ingrowth, unstable (Figures 5 and 7) Stress shielding (Figures 7 and 10) Wear and osteolysis (Figures 8, 9, and 10) Heterotopic ossification Actions

showing the entire prosthesis with at least 20 to 30 mm of pelvic bone above the acetabular component (socket) and below the femoral component (stem). Four key questions need to be addressed with regard to the THA itself — concerning the patient, the prosthesis, the procedure, and the present status — before deciding on any future actions.

THE PATIENT The reason for performing the THA is sometimes obvious in radiographs; clues may be present in the

Figure 1 A woman had bilateral cementless total hip arthroplasty (THA) [Anatomic Medullary Locking, AML; DePuy, Johnson and Johnson] done when she was 35 years old. There were multiple metallic clips over the right lower quadrant of the abdomen. She had a history of renal transplantation and the reason for THA was steroid-induced avascular necrosis of the femoral heads.

2

contralateral hip, especially if it has not been replaced. For example, radiological features of avascular necrosis, osteoarthritis, or developmental dysplasia may be present. Some features may also be present in the replaced hip, such as residual hardware after internal fixation of the acetabular or proximal femoral fractures, or holes indicative of previous internal fixation devices. For radiographs showing very small-sized bones and prosthesis, the possibility of juvenile chronic arthritis or bone dysplasias should be considered. In patients with ankylosing spondylitis, changes in the sacroiliac joints and the lumbosacral spine may be seen in the AP radiograph of the pelvis, and the obturator foramen morphology may be changed.1 Any other unusual feature in the radiographs should also be noted (Figure 1).

Figure 2 A 77-year-old man had bilateral total hip arthroplasty (THA) more than 25 years ago. All components were actually cemented, but not visible in the radiograph as radiolucent cement was used at that time. The socket was all-polyethylene in the left THA (Muller). The right THA had a metal-metal articulation (McKee-Ferrar).

Reading Radiographs after Total Hip Arthroplasty

THE PROSTHESIS Unless the surgeon is very familiar with the specific prosthesis, it is often difficult to recognise the actual model and brand of the prosthesis used; it is also unnecessary. Nevertheless, the method of fixation of the prosthesis, and the features of the socket, stem, and articulation can be commented on.

stem that is longer than usual may sometimes suggest problems at the time of surgery. For the articulation, the surgeon can comment on the size of the femoral head. A large femoral head (≥32 mm) was commonly used in the early days. However, it became unpopular as it made the polyethylene (PE) liner too thin and thus increased the wear rate.

The fixation of components may be cemented or cementless. While cement is usually made radioopaque by the addition of barium sulphate or zirconium dioxide, it may be radiolucent, especially if the THA was done a long time ago (Figure 2). For cementless prosthesis, part of the surface may appear rough on the radiograph; it is then either porous coated, hydroxyapatite coated, or simply has a roughened surface. It can also be a hybrid fixation, with usually a cementless socket and a cemented stem. The socket may be all-polyethylene (mostly cemented) or have a metal shell (mostly cementless). For the allpolyethylene cemented sockets, there are metallic markers or wires to enable the surgeon to assess the alignment and wear. For the metal-backed cementless sockets, spikes, pegs, tabs or screws that serve to improve the stability of the cup can be used. However, many surgeons prefer to use the press-fit technique where such supplementary fixation devices may not be seen. For cemented stems, the presence of a collar should be noted. The collar is designed to prevent subsidence of the stem and to facilitate vertical load transfer from the stem to the femoral neck. Cemented stems designed to subside in the cement mantle for better loading of the latter, such as the Exeter stem (Howmedica), have a polished surface and are tapered and collarless. The presence of a centraliser distally can also be looked out for (Figure 3). The latter is usually radiolucent and may be mistaken for bubbles or voids. For cementless stems, the shape, presence of a collar, extent of the coating, and length can be commented on. The stem may be curved and bowed so as to fit with the actual anatomy of the medullary canal, or it may be a straight stem design that makes use of the 3-point fixation principle. The distal part of the stem may be tapered or cylindrical. The use of a collar in cementless stems is controversial. For the extent of coating, it may be proximal (metaphyseal loading) or extensive or fully coated (distal loading). The actual length of the stem varies among different models. A

Figure 3 The cemented stem (Elite plus; DePuy, Johnson and Johnson) was inserted with improved cementing techniques. A cement restrictor (R) was used and the cement column was stopped abruptly. There was a centraliser (C) that appeared as a radiolucent bubble distal to the stem tip. The stem alignment was neutral. There was a complete cement mantle of >2 mm. Apart from minor radiolucencies, there was white-out at the distal medullary canal. This was, strictly speaking, a grade B cementation.

3

Hong Kong J Orthop Surg.

Still, it has recently become popular again because of the introduction of a highly cross-linked PE, the resurgence of metal-metal articulation, and the reduced risk of dislocation associated with a large hip ball diameter. It may not be possible to tell whether the hip ball is made of metal or zirconia, although alumina hip balls may appear more radiolucent than the usual metal ones.

THE PROCEDURE It can be discerned whether the THA was done well and if any problems were encountered at the time of the THA from the postoperative radiographs. When such patients come back for follow-up, more caution should be exercised to actively watch out for any early indicators of failure. For the sockets, the lateral opening or abduction angle, and the vertical and horizontal positions can be commented on. The lateral opening angle is subtended by a line joining the superolateral and inferomedial corners of the socket and the inter-teardrop line. Most surgeons try to make the lateral opening angle at 45°. If the socket is too vertical, the PE wear rate is increased and the chance of having a fracture of the ceramic-ceramic articulation is higher. The inferomedial corner of the socket is normally at about the same horizontal level as the bottom of the teardrop while the medial part of the socket is close to the lateral border of the teardrop. Protrusio acetabuli is present if the socket breaches the ilio-ischial line or the Kohler’s line. The vertical and horizontal distances of the hip joint centre from the bottom of the corresponding teardrop can also be measured, and then compared with those in the contralateral hip if it is normal. The anteversion of the socket is often difficult to ascertain in standard radiographs. Many cemented sockets

Figure 4 The metal shell of cementless socket is projected as an ellipse. There is little or no version if the 2 halves of the ellipse overlap (left) in the anteroposterior pelvic radiograph. A broader ellipse signifies a bigger version angle (right).

4

Chiu KY, Yau WP, Tang WM, et al

have wire markers to indicate its position, such as a ring near the edge so that it is projected as an ellipse when seen at an angle in the radiograph. The metal shell of the cementless socket is also projected as an ellipse. Mathematical formulae for evaluating the version angle by calculating the maximum and minimum diameters of the projected ellipse have been proposed, but they are too complicated to be of general use. A broader ellipse signifies bigger version angles, and there is little or no version if the 2 halves of the ellipse overlap in the AP pelvic radiograph (Figure 4). It must be pointed out that the same ellipse may be seen with the socket in anteversion or retroversion. The version of the socket is an important factor when a patient has suffered from recurrent dislocation of the THA. For the stem, it should be noted whether it is in the neutral, varus, or valgus position. One can tell whether a cemented stem was inserted with early or improved techniques by looking at the cement distally. With improved cementing techniques, there is usually an abrupt cut off of the cement in the medullary canal because of the use of a distal plug (Figure 3). If a commercially available plastic plug was used, a metal marker can usually be seen. If there is a long column of cement extending down the medullary canal, the cemented stem was likely to be inserted with early cementing techniques. Barrack et al described the grading of the cement mantles around stems.2 The cement mantle is grade A if there is complete filling of the medullary cavity by cement, or 'white-out' of the bone-cement interface. In grade B mantles, there is slight (1 mm at any point. For the cementless socket, the description of RLLs is similar to that for the cemented socket. Migration of a cementless socket is defined by a change in the opening angle of >8° or a difference in the component position of >3 mm when comparable radiographs are compared.7 For a cementless socket with additional fixation devices, like screws, reactive lines and 'halo' around such devices in a loose socket may be seen. Shedding of metal particles from the porous-coated surface also implies loosening of the cementless socket. For the stems, RLLs may be present around the entire surface or it may only be present over certain areas. The latter can be recorded by dividing the stem surface into 7 zones in the AP radiograph as described by Gruen et al and 7 additional zones in the lateral radiograph as described by Johnston et al (Figure 5).8,9 Apart from RLLs, such zonal evaluation can also be used to

It is extremely useful to have serial radiographs after THA for comparison, especially in the presence of radiolucent lines (RLLs). RLLs usually refer to black lines, either at the cement-prosthesis interface or the cement-bone interface, that were absent in the early postoperative radiographs. Radiolucencies seen between the cement and the endosteal surface of the femoral medullary canal may be just cancellous bone not filled up with cement, rather than RLLs. In cementless prosthesis, white lines, usually referred to as reactive lines, that run parallel but very close to the surface of the prosthesis may be seen. In addition to the lines, the surgeon has to decide whether or not the components are well fixed, and whether there is wear and osteolysis. Other problems, such as heterotopic ossification, may also be present. DeLee and Charnley described 3 types of cement-bone demarcations in cemented sockets.5 The RLL occupies only the lateral one-third in type I, and also involves the superomedial quadrant in type II. In type III, the entire circumference is involved. While types I, II, and III actually represented 3 different extents in the original article, many authors have subsequently used types I, II, and III to indicate 3 different zones (Figure 5). Hodgkinson et al described a type IV demarcation if there was socket migration.6 A socket was considered to be loose if a continuous RLL was evident in all 3 zones (type III) or if the cup had migrated (type IV).

Figure 5 The surface of the socket is divided into 3 zones and that of the stem into 7 zones in the anteroposterior radiograph (left). There are 7 more zones around the stem in the lateral radiograph (right).

5

Hong Kong J Orthop Surg.

Chiu KY, Yau WP, Tang WM, et al

describe the cement mantle thickness, remodeling changes including atrophy and hypertrophy, and osteolytic lesions.

occupies >50% but

Suggest Documents