- Email: [email protected]
Loosening of threaded acetabular components
The Journal of Arthroplasty
Loosening
of Threaded Components
Vol. 8 No. 6 1993
Acetabular
Radiographic Manifestations Kaj Tallroth, MD,* PZr Slitis, MD,+ Pekka Ylinen, MD,+ Pekka Paavolainen, MD,+ and Timo Paavilainen, MD-t
Abstract: Radiographic signs of loosening were assessed, and their sequential appearance recorded, in a retrospective analysis of 20 loosened, threaded acetabular cups. There was radiographic evidence of loosening in 19 (95%) cups. In decreasing order of frequency, the radiographic manifestations included bone resorption between the threads in 19 (95%) cups, increased bone density above the cup in 15 (75%), continuous lucency around the cup in 14 (70%), axial migration in 9 (45%), and rotation in 6 (30%). The order of appearance of these radiographic manifestations was the same as the order of frequency of occurrence of the alterations. In the one patient with no radiographically detectable signs of loosening, the cup did not move freely but was easily detached during surgery. Early radiographic diagnosis of loosening requires accurate positioning of the patient in order for the radiographs to reveal bone resorption between the threads of the cup. Key words: threaded acetabular component, total hip arthroplasty, radiology, loosening.
Cementless fixation of total hip implants has been developed in an attempt to reduce the frequency of loosening encountered with cemented prostheses. A large variety of acetabular components designed for noncemented use is available. The main methods of fixation currently used are press-fit, hemispherical, porous-coated cups, porous-coated cups secured by fixation pegs or traversing screws, and threaded cups with or without porous coating. In a recent study, however, Engh et al.* reported disappointing results after implantation of threaded acetabular components (TAC), as 18% of the implants migrated; this was interpreted as a sign of instability. Similarly, Snorrason and KBrrholm,’ using roentgen stereophotogrammetric analysis, demonstrated primary migration of the TAC in all of the patients they examFrom the Departments of *Radiology and torthopaedics, Hospital, Invalid Foundation, Helsinki, Finland.
ined, and additional rotatory movement in about half of them. The purpose of this retrospective analysis of 20 loosened, threaded cups was to identify the radiographic signs of early aseptic loosening.
Materials
and Methods
Patients Between 1986 and 1989 a total of 714 Bi-Metric (Biomet,
Warsaw,
IN) total hip arthroplasties
with
TAC were performed at our hospital (Orthopaedic Hospital, Helsinki, Finland). Until the end of 199 1, 132 (18.5%) of these prostheses have subsequently been revised, 115 ( 16.1%) of them due to a loosened TAC. Twenty of these patients ( 10 women and 10 men), with primary osteoarthritis as the indication for primary arthroplasty, comprised the subject group of this study. The mean age of the patients was 5 1.1 years (range, 34-70 years). In all of the patients the TACs were matched with a cementless, Bi-Met-
Orthopaedic
Reprint requests: Dr. Kaj Tallroth, Department of Radiology, Orthopaedic Hospital, Invalid Foundation, Tenholantie 10, SF00280 Helsinki, Finland.
581
582
The Journal of Arthroplasty
Vol. 8 No. 6 December
ric, collarless primary design femoral component made of titanium alloy. The most common clinical symptom indicating TAC loosening was increasing “start” pain in the groin or buttock. At revision surgery, the acetabular components were loose, whereas the femoral stems were well seated and fixed in every case. There was no loosening due to infection. Acetabular
Cup
The threaded, self-tapping, titanium, acetabular components used in this series were produced by Biomet Inc. (Warsaw, IN). The cups were comprised of a one-piece, wrought titanium shell with machined self-tapping threads (height, 3 mm) and a liner of ultra-high molecular weight polyethylene. The dome of the cup was sandblasted and had no porous coating. Operative
1993
of rotation of the hip and the inferior margin of the teardrop. Horizontal migration was defined as the change in distance between the center of rotation of the hip to a vertical line through the teardrop. A change of more than 2 mm in one or both of these was considered to indicate acetabular migration. Inclination of the lateral opening of the acetabular component was related to a horizontal line drawn perpendicular to a plumb line on the radiograph. A change of 4” was considered to indicate rotational migration. Bone resorption between the threads and bone texture around and above the cup in the acetabulum were noted. Both the width and extent of radiolucency adjacent to the cup were recorded at the laterosuperior and medioinferior zones, as was continuous radiolucency either at the dome of the cup or around the rim.
Results
Technique
Arthroplasty was carried out with the patient in the lateral decubitus position using a direct lateral approach.3 Capsule and marginal osteophytes were excised, and the acetabulum was reamed until bleeding subchondra1 bone was visible. The self-tapping TAC was manually screwed into the acetabulum until close contact between the cup dome and bone was achieved. Lateral inclination of the cup was 3 5”-45” and ventral inclination 15”-20’.
Radiographic evidence of loosening was seen in 19 of the 20 patients. The frequency of the recorded signs of loosening is given in Table 1. From serial films it was also possible to determine the order in which the various signs appeared and progressed. The time interval between total hip arthroplasty and the radiologic diagnosis of loosening ranged from 9 to 36 months (mean, 17 months). Bone Resorption
Radiologic
Evaluation
The radiographs evaluated included anteroposterior, oblique, and lateral (frog and crosstable) views. The anteroposterior view was taken with the patient weight bearing, feet parallel, and a constant distance between the feet as described by Turula et al.’ The x-ray beam was centered on the symphysis pubis in order to obtain consistent and comparable films. The lateral frog view was obtained with the patient supine. When indicated, oblique views with both patient and tube tilted were obtained to assess the acetabular component with the threads viewed in profile. The radiographs were assessed by analyzing all of the postoperative films and comparing them with each other. Four sets of films per patient were taken on average before ihe cups were revised. The radiographs were interpreted according to the recommendations of the Societe Internationale de Chirurgie Orthopedique et de Traumatologie, the American Academy of Orthopaedic Surgeons, and the American Hip Society.’ Vertical migration was defined as the change in distance between the center
Between
the Threads
Loss of bone between the threads was the most common, and also the first, sign of loosening in all 19 cases. In 18 cases the loss of bone was seen as distinct osteolytic areas between the threads in the superolateral area (Fig. 1) . Uniform confluence and growth of the radiolucencies were characteristic of progressive loosening on subsequent films. Bone Formation
in the Acetabulum
Altogether, 15 of the 19 loose cups showed a sclerotic area above the cup. In 11 cases this was visible on the first film diagnostic of loosening. The bone
Table
1. Radiographic Signs of Loosening Threaded Acetabular Cups
Bone resorption between threads Sclerosis in acetabulum Continuous lucency around the cup Axial migration of the cup Rotation of the cup
in
No.
Percent
19 15 14 9 5
95 75 70 45 30
Threaded Acetabular
Fig. 1. Frontal view of the acetabular component showing a distinct osteolytic area (arrow) as the first sign of loosening between the outermost threads at the superolateral margin of the cup. No evidence of component migration.
formation appeared above the cup regardless position of the cup.
of the
Continuous Lucency of the Cup The confluence of the separate lucencies between the threads gave rise to larger lucent zones in 14 cases. They were more frequent at the inferior margin
Component
l
Tallroth et al.
583
Fig. 3. Radiographshowing axial subsidenceof the cup. There isa lucent band of uniform breadth at the rim of the cup (straight arrows) and a small area of bone sclerosis above the cup (curved arrow).
(14 cases) of the cup than the superior margin ( 11 cases). In cases with horizontal rotation of the cup, the lucencies were wider at the lower margin (Fig. 2). Axial subsidence was accompanied by a lucent band along the opening of the cup (Fig. 3). Migration Axial migration of the cup occurred in nine cups, rotational migration occurred in five, and the two types coexisted in three. All cups showing rotational instability had turned horizontally. Migratisn was observed in only three cases on the first film demonstrating loosening, and in each of these the migration was preceded by obvious lucency.
Discussion Bobyn et al.’ reported a case in which microradiography of a well-fixed TAC that had been removed showed much less bone in the grooves of the threads than had been expected. Histologic sections proved Fig. 2. Radiograph showing a continuous lytic demarca-
tion around the cup. The threads have lost their grip and the cup has turned horizontally, leaving behind a wide
lucency at the inferior margin of the acetabulum (arrow). A large area of sclerotic bone is seenabove the cup.
that there had been resorption of bone and ingrowth of dense fibrous tissue. They suggested that the bone resorption might reflect ether thermal injury, mechanical instability, excessive stress concentration, or localized ischemia due to damage to the microvascu-
584
The Journal of Arthroplasty
Vol. 8 No. 6 December
lature caused by the sharp threads as the implant is screwed into position. Using a finite element method, Huiskes4 demonstrated that the load transfer from a noncemented threaded cup to the bone in the acetabulum is concentrated mostly at the upper and lower corners and at the first and last threads. In view of the high, local stress concentration, he emphasized that bone yielding, remodeling, and densification could occur in the immediate vicinity of the cup. Since the acetabular components are inserted in approximately 15” anteversion to mimic the anatomic inclination of the acetabulum, the grooves of the cup are obscured by the threads on an anteroposterior radiograph. This was recognized by Shaw et al.6 who considered that radiographic determination of component stability is very difficult with noncemented acetabular components and particularly difficult with TACs, and conceded that the rate of loosening may be higher than reported. Likewise, Snorrason and K9rrholm,7 although they observed primary migration in 20 of the TACs they examined, admitted that they were not able to evaluate radiolucent zones reliably owing to the lack of oblique views. However, the grooves of the cup and the bone ingrowth can easily be visualized on views in which the x-ray beams are on line with the threads. To obtain oblique films of this kind, the patient and often the x-ray tube have to be tilted. To ensure that the threads are in profile, we recommend that the patient be positioned under fluoroscopic guidance. This also saves examination time and reduces radiation exposure for the patient. In our study, the availability of focused oblique films of the threads helped in most cases to detect bone resorption between the threads even before other radiographic manifestations of loosening were present. In this review of 20 verified loosened TACs, the primary radiographic signs of loosening and their further migration displayed uniform features suggesting elements that may explain the mechanism of loosening. The threads of the cup cause considerable hoop stress in the acetabulum during and after insertion and may cause both microfractures and spatial
1993
compression of the adjacent acetabular bone. Blood perfusion of the area is easily affected, giving rise to local ischaemia of the bone. Necrotic bone will be resorbed and replaced by fibrous tissue. The first sign of loosening is osteolysis around the threads, usually within 1 year from the operation. The cup, if not secured by ingrowth of bone in the dome area, will now be loose, and forces transmitted from the lower leg through the implant will tend to displace the inserted cup. Increased stress above the cup is reflected by increased bone density in the supraacetabular area. As the migration continues, the radiolucent areas grow, especially in the medioinferior region, eventually causing the cup to tilt. Slight pain occurs at an early stage. Loosening can be confirmed radiographically by using dedicated views that outline the grooves of the threads in profile. Due to the high rates of loosening of this particular type of threaded cup, we have abandoned its use in favor of hemispherical, press-fit, porous-coated components, or conventional cemented cups.
References 1. Bobyn JD, Engh CA, Glassman AH: Radiography and histology of threaded acetabular implant: one case studied at two years. J Bone Joint Surg 70B:302, 1988 2. Engh CA, Griffin W, Marx CL: Cementless acetabular components. J Bone Joint Surg 72B:53, 1990 3. Hardinge K: Direct lateral approach to the hip. J Bone Joint Surg 64B: 17, 1982 4. Huiskes R: Finite element analysis of acetabular reconstruction: noncemented threaded cups. Acta Orthop Stand 58:620, 1987 5. Johnston RC, Fitzgerald RH Jr, Harris WH et al: Clinical and radiographic evaluation of total hip replacement. J Bone Joint Surg 72A:161, 1990 6. Shaw JA, Bailey JH, Bruno A, Greer RB III: Threaded acetabular components for primary and revision total hip arthroplasty. J Arthroplasty 5:201, 1990 7. Snorrason F, Karrholm J: Primary migration of fully threaded acetabular prostheses: a roentgen stereophotogrammetric analysis. J Bone Joint Surg 72B:647, 1990 8. Turula KB, Friberg 0, Haajanen J et al: Weight-bearing radiography in total hip replacement. Skeletal Radio1 14:200, 1985
Loosening
of Threaded Components
Vol. 8 No. 6 1993
Acetabular
Radiographic Manifestations Kaj Tallroth, MD,* PZr Slitis, MD,+ Pekka Ylinen, MD,+ Pekka Paavolainen, MD,+ and Timo Paavilainen, MD-t
Abstract: Radiographic signs of loosening were assessed, and their sequential appearance recorded, in a retrospective analysis of 20 loosened, threaded acetabular cups. There was radiographic evidence of loosening in 19 (95%) cups. In decreasing order of frequency, the radiographic manifestations included bone resorption between the threads in 19 (95%) cups, increased bone density above the cup in 15 (75%), continuous lucency around the cup in 14 (70%), axial migration in 9 (45%), and rotation in 6 (30%). The order of appearance of these radiographic manifestations was the same as the order of frequency of occurrence of the alterations. In the one patient with no radiographically detectable signs of loosening, the cup did not move freely but was easily detached during surgery. Early radiographic diagnosis of loosening requires accurate positioning of the patient in order for the radiographs to reveal bone resorption between the threads of the cup. Key words: threaded acetabular component, total hip arthroplasty, radiology, loosening.
Cementless fixation of total hip implants has been developed in an attempt to reduce the frequency of loosening encountered with cemented prostheses. A large variety of acetabular components designed for noncemented use is available. The main methods of fixation currently used are press-fit, hemispherical, porous-coated cups, porous-coated cups secured by fixation pegs or traversing screws, and threaded cups with or without porous coating. In a recent study, however, Engh et al.* reported disappointing results after implantation of threaded acetabular components (TAC), as 18% of the implants migrated; this was interpreted as a sign of instability. Similarly, Snorrason and KBrrholm,’ using roentgen stereophotogrammetric analysis, demonstrated primary migration of the TAC in all of the patients they examFrom the Departments of *Radiology and torthopaedics, Hospital, Invalid Foundation, Helsinki, Finland.
ined, and additional rotatory movement in about half of them. The purpose of this retrospective analysis of 20 loosened, threaded cups was to identify the radiographic signs of early aseptic loosening.
Materials
and Methods
Patients Between 1986 and 1989 a total of 714 Bi-Metric (Biomet,
Warsaw,
IN) total hip arthroplasties
with
TAC were performed at our hospital (Orthopaedic Hospital, Helsinki, Finland). Until the end of 199 1, 132 (18.5%) of these prostheses have subsequently been revised, 115 ( 16.1%) of them due to a loosened TAC. Twenty of these patients ( 10 women and 10 men), with primary osteoarthritis as the indication for primary arthroplasty, comprised the subject group of this study. The mean age of the patients was 5 1.1 years (range, 34-70 years). In all of the patients the TACs were matched with a cementless, Bi-Met-
Orthopaedic
Reprint requests: Dr. Kaj Tallroth, Department of Radiology, Orthopaedic Hospital, Invalid Foundation, Tenholantie 10, SF00280 Helsinki, Finland.
581
582
The Journal of Arthroplasty
Vol. 8 No. 6 December
ric, collarless primary design femoral component made of titanium alloy. The most common clinical symptom indicating TAC loosening was increasing “start” pain in the groin or buttock. At revision surgery, the acetabular components were loose, whereas the femoral stems were well seated and fixed in every case. There was no loosening due to infection. Acetabular
Cup
The threaded, self-tapping, titanium, acetabular components used in this series were produced by Biomet Inc. (Warsaw, IN). The cups were comprised of a one-piece, wrought titanium shell with machined self-tapping threads (height, 3 mm) and a liner of ultra-high molecular weight polyethylene. The dome of the cup was sandblasted and had no porous coating. Operative
1993
of rotation of the hip and the inferior margin of the teardrop. Horizontal migration was defined as the change in distance between the center of rotation of the hip to a vertical line through the teardrop. A change of more than 2 mm in one or both of these was considered to indicate acetabular migration. Inclination of the lateral opening of the acetabular component was related to a horizontal line drawn perpendicular to a plumb line on the radiograph. A change of 4” was considered to indicate rotational migration. Bone resorption between the threads and bone texture around and above the cup in the acetabulum were noted. Both the width and extent of radiolucency adjacent to the cup were recorded at the laterosuperior and medioinferior zones, as was continuous radiolucency either at the dome of the cup or around the rim.
Results
Technique
Arthroplasty was carried out with the patient in the lateral decubitus position using a direct lateral approach.3 Capsule and marginal osteophytes were excised, and the acetabulum was reamed until bleeding subchondra1 bone was visible. The self-tapping TAC was manually screwed into the acetabulum until close contact between the cup dome and bone was achieved. Lateral inclination of the cup was 3 5”-45” and ventral inclination 15”-20’.
Radiographic evidence of loosening was seen in 19 of the 20 patients. The frequency of the recorded signs of loosening is given in Table 1. From serial films it was also possible to determine the order in which the various signs appeared and progressed. The time interval between total hip arthroplasty and the radiologic diagnosis of loosening ranged from 9 to 36 months (mean, 17 months). Bone Resorption
Radiologic
Evaluation
The radiographs evaluated included anteroposterior, oblique, and lateral (frog and crosstable) views. The anteroposterior view was taken with the patient weight bearing, feet parallel, and a constant distance between the feet as described by Turula et al.’ The x-ray beam was centered on the symphysis pubis in order to obtain consistent and comparable films. The lateral frog view was obtained with the patient supine. When indicated, oblique views with both patient and tube tilted were obtained to assess the acetabular component with the threads viewed in profile. The radiographs were assessed by analyzing all of the postoperative films and comparing them with each other. Four sets of films per patient were taken on average before ihe cups were revised. The radiographs were interpreted according to the recommendations of the Societe Internationale de Chirurgie Orthopedique et de Traumatologie, the American Academy of Orthopaedic Surgeons, and the American Hip Society.’ Vertical migration was defined as the change in distance between the center
Between
the Threads
Loss of bone between the threads was the most common, and also the first, sign of loosening in all 19 cases. In 18 cases the loss of bone was seen as distinct osteolytic areas between the threads in the superolateral area (Fig. 1) . Uniform confluence and growth of the radiolucencies were characteristic of progressive loosening on subsequent films. Bone Formation
in the Acetabulum
Altogether, 15 of the 19 loose cups showed a sclerotic area above the cup. In 11 cases this was visible on the first film diagnostic of loosening. The bone
Table
1. Radiographic Signs of Loosening Threaded Acetabular Cups
Bone resorption between threads Sclerosis in acetabulum Continuous lucency around the cup Axial migration of the cup Rotation of the cup
in
No.
Percent
19 15 14 9 5
95 75 70 45 30
Threaded Acetabular
Fig. 1. Frontal view of the acetabular component showing a distinct osteolytic area (arrow) as the first sign of loosening between the outermost threads at the superolateral margin of the cup. No evidence of component migration.
formation appeared above the cup regardless position of the cup.
of the
Continuous Lucency of the Cup The confluence of the separate lucencies between the threads gave rise to larger lucent zones in 14 cases. They were more frequent at the inferior margin
Component
l
Tallroth et al.
583
Fig. 3. Radiographshowing axial subsidenceof the cup. There isa lucent band of uniform breadth at the rim of the cup (straight arrows) and a small area of bone sclerosis above the cup (curved arrow).
(14 cases) of the cup than the superior margin ( 11 cases). In cases with horizontal rotation of the cup, the lucencies were wider at the lower margin (Fig. 2). Axial subsidence was accompanied by a lucent band along the opening of the cup (Fig. 3). Migration Axial migration of the cup occurred in nine cups, rotational migration occurred in five, and the two types coexisted in three. All cups showing rotational instability had turned horizontally. Migratisn was observed in only three cases on the first film demonstrating loosening, and in each of these the migration was preceded by obvious lucency.
Discussion Bobyn et al.’ reported a case in which microradiography of a well-fixed TAC that had been removed showed much less bone in the grooves of the threads than had been expected. Histologic sections proved Fig. 2. Radiograph showing a continuous lytic demarca-
tion around the cup. The threads have lost their grip and the cup has turned horizontally, leaving behind a wide
lucency at the inferior margin of the acetabulum (arrow). A large area of sclerotic bone is seenabove the cup.
that there had been resorption of bone and ingrowth of dense fibrous tissue. They suggested that the bone resorption might reflect ether thermal injury, mechanical instability, excessive stress concentration, or localized ischemia due to damage to the microvascu-
584
The Journal of Arthroplasty
Vol. 8 No. 6 December
lature caused by the sharp threads as the implant is screwed into position. Using a finite element method, Huiskes4 demonstrated that the load transfer from a noncemented threaded cup to the bone in the acetabulum is concentrated mostly at the upper and lower corners and at the first and last threads. In view of the high, local stress concentration, he emphasized that bone yielding, remodeling, and densification could occur in the immediate vicinity of the cup. Since the acetabular components are inserted in approximately 15” anteversion to mimic the anatomic inclination of the acetabulum, the grooves of the cup are obscured by the threads on an anteroposterior radiograph. This was recognized by Shaw et al.6 who considered that radiographic determination of component stability is very difficult with noncemented acetabular components and particularly difficult with TACs, and conceded that the rate of loosening may be higher than reported. Likewise, Snorrason and K9rrholm,7 although they observed primary migration in 20 of the TACs they examined, admitted that they were not able to evaluate radiolucent zones reliably owing to the lack of oblique views. However, the grooves of the cup and the bone ingrowth can easily be visualized on views in which the x-ray beams are on line with the threads. To obtain oblique films of this kind, the patient and often the x-ray tube have to be tilted. To ensure that the threads are in profile, we recommend that the patient be positioned under fluoroscopic guidance. This also saves examination time and reduces radiation exposure for the patient. In our study, the availability of focused oblique films of the threads helped in most cases to detect bone resorption between the threads even before other radiographic manifestations of loosening were present. In this review of 20 verified loosened TACs, the primary radiographic signs of loosening and their further migration displayed uniform features suggesting elements that may explain the mechanism of loosening. The threads of the cup cause considerable hoop stress in the acetabulum during and after insertion and may cause both microfractures and spatial
1993
compression of the adjacent acetabular bone. Blood perfusion of the area is easily affected, giving rise to local ischaemia of the bone. Necrotic bone will be resorbed and replaced by fibrous tissue. The first sign of loosening is osteolysis around the threads, usually within 1 year from the operation. The cup, if not secured by ingrowth of bone in the dome area, will now be loose, and forces transmitted from the lower leg through the implant will tend to displace the inserted cup. Increased stress above the cup is reflected by increased bone density in the supraacetabular area. As the migration continues, the radiolucent areas grow, especially in the medioinferior region, eventually causing the cup to tilt. Slight pain occurs at an early stage. Loosening can be confirmed radiographically by using dedicated views that outline the grooves of the threads in profile. Due to the high rates of loosening of this particular type of threaded cup, we have abandoned its use in favor of hemispherical, press-fit, porous-coated components, or conventional cemented cups.
References 1. Bobyn JD, Engh CA, Glassman AH: Radiography and histology of threaded acetabular implant: one case studied at two years. J Bone Joint Surg 70B:302, 1988 2. Engh CA, Griffin W, Marx CL: Cementless acetabular components. J Bone Joint Surg 72B:53, 1990 3. Hardinge K: Direct lateral approach to the hip. J Bone Joint Surg 64B: 17, 1982 4. Huiskes R: Finite element analysis of acetabular reconstruction: noncemented threaded cups. Acta Orthop Stand 58:620, 1987 5. Johnston RC, Fitzgerald RH Jr, Harris WH et al: Clinical and radiographic evaluation of total hip replacement. J Bone Joint Surg 72A:161, 1990 6. Shaw JA, Bailey JH, Bruno A, Greer RB III: Threaded acetabular components for primary and revision total hip arthroplasty. J Arthroplasty 5:201, 1990 7. Snorrason F, Karrholm J: Primary migration of fully threaded acetabular prostheses: a roentgen stereophotogrammetric analysis. J Bone Joint Surg 72B:647, 1990 8. Turula KB, Friberg 0, Haajanen J et al: Weight-bearing radiography in total hip replacement. Skeletal Radio1 14:200, 1985