Approximately 40,000 total hip replacements (THRs) are performed in this country each year.1 Of the 62 primary THR prostheses available in this country, only 30% have results published in peer-reviewed journals.2 A method of predicting which prostheses are destined for early failure has been needed for quite some time. It recently has been shown that early mechanical failure can be predicted from excessive migration of a stem at 2 years.3
The most accurate method of measuring femoral stem migration is roentgen stereophotogrammetric analysis (RSA),4 but this method is prospective and expensive. Walker et al5 developed a method of measuring stem migration retrospectively from standard anteroposterior (AP) pelvic radiographs with an accuracy approaching that of RSA.
This study measured migration of the CLS (Cementless Total Hip Replacement System; Protek AG, Bern, Switzerland) uncemented femoral stem from standard AP radiographs using a modification of Walker's original method and correlated these data, with radiographic appearances and clinical outcome in terms of thigh pain.
Materials and Methods
The CLS femoral stem (Fig 1) is made of titanium alloy with a corundum-blasted surface designed as a three-dimensional taper for press-fit implantation. The distal part of the taper is thin to prevent canal fill and has proximal ridges to limit rotational migration and encourage bone ingrowth.
One hundred twenty-six uncemented CLS femoral stems were implanted in 117 patients over a 7-year period. Patient suitability for an uncemented prosthesis was determined preoperatively by the surgeon and was based on patient age, bone quality, and the proximal femoral morphology. Although a Spotorno score6 was not formally calculated preoperatively, it was estimated for the purposes of the study.
Mean patient age was 59.6 years (median: 61.1 years; range: 33.6-69.4 years) and mean weight was 73 kg (range: 39-115 kg). Fifty-nine women and 58 men comprised the study population. Preoperative diagnoses included osteoarthritis (114 cases), avascular necrosis secondary to fractured neck of femur (8 cases), and rheumatoid arthritis (4 cases).
Fig 1 : Photograph of the CLS femoral stem.
Surgery was performed using a transgluteal approach.7 A CLS femoral stem was implanted in all patients. A screw-threaded acetabular component (CLW) was implanted in the first 31 patients and a press-fit CLS acetabular component was implanted in the remaining 95 patients. The decision to change cups reflected the surgeon's (B. C. P.) preference for a press-fit prosthesis. A 32-mm femoral head was used in 100 cases and a 28-mm head was used in 26 cases. All of the prostheses in the study had a ceramic/poLyethylene articulation.
Fig 2: Migration of the CLS femoral stem.
Anteroposterior radiographs of the pelvis taken immediately postoperatively, at 3 months, at 1 year, and annually thereafter were used for the measurements. Software was developed specifically for the prosthesis and used in conjunction with a digitizer to measure the vertical migration of the stems. The geometry of the CLS stem and the absence of a collar meant that the software used in the original study5 had to be modified. The predetermined reference points were plotted, and migration distances were calculated. The technique is accurate to within 0.37 mm allowing for 10° of internal/external rotation or flexion/extension between the successive radiographs of any one patient.
Radiographs were reviewed by the principal author, and a radiographic assessment was carried out noting the following features: sclerotic and lucent lines adjacent to the prosthesis, the presence or absence of femoral shaft hypertrophy, spot welds, "corticalization," pedestal formation, calcar resorption, and debris lysis. The presence of heterotopic bone was documented and classified according to Brooker et al.8
A questionnaire to determine the incidence of thigh pain was mailed to all patients. For those patients who did not respond, their medical records were reviewed for documentation of thigh pain. All data were analyzed by a statistician.
Of the 126 hips, 1 1 with poor-quality radiographs were excluded, leaving 115 hips for review. The mean followup was 2.77 years (range: 6 months-7 years) for radiographic review and 3 years (range: 1-8 years) for clinical review.
Migration. Mean migration of the femoral stem was 2 mm at 2 years and 3.66 mm at 7 years. Migration data are plotted in Figure 2, and Table 1 details the number of complete sets of migration figures for each year of follow-up. Of those patients with 7 years of complete migration data, the mean migration at 2 years was 2.57 mm, marginally higher than the overall mean.
At 2 years, 15 stems had migrated >3 mm: 10 had migrated 3-3.99 mm, 4 had migrated 4-4.99 mm, and 1 had migrated 9.27 mm. Interestingly, this latter prosthesis had migrated already 9.26 mm at 3 months and did not subsequently migrate, and the patient remains asymptomatic.
Radiographic Assessment. The results of the radiographic assessment are summarized in Table 2. Sclerotic lines adjacent to the prosthesis were noted in 62 (49.2%) stems while a nonprogressive lucent line was noted in 4 (3.17%) stems. Fifty percent of the sclerotic lines were confined to Gruen zone 1 at die shoulder Of the prosthesis and were not progressive (Fig 3). The only other consistent pattern was of nonprogressive sclerotic lines in Gruen zones 1 and 7; this was noted in 11 (8.7%) cases. One case showed a continuous sclerotic line in all seven Gruen zones.
Patients in the migration study for each year of follow-up
Radiographic features identified in relation to the femoral stems
Shaft hypertophy was noted in four patients, two of whom had sustained a perioperative calcar split that required cerclage wire insertion. Of the 13 prostheses demonstrating pedestal formation, 1 1 were in a neutral position and 2 in varus.
Cortiealization is a term that has been used to indicate a generalized increase in calcification of the cancellous bone adjacent to the prosthesis; this was observed in 1 8 cases. Varying degrees of calcar resorption was seen in 21 cases, and proximal focal endosteal lysis was noted in three patients. There were 47 (37.3%) and 22 (17.5%) cases of Brooker grades 1 and 2 heterotopic bone formation, respectively. The remaining 57 hips had no heterotopic bone. The mean Spotorno score was 4.49 (range: 1-9).
Clinical Results. Of the 1 17 patients with 126 THRs, 3 (2.4%) patients specifically complained of thigh pain or had the presence of thigh pain documented in their hospital record.
Statistical Analysis. Analysis of the clinical data revealed no statistical relationship between migration of the femoral stem at any time and age, weight, patients' perceived outcome, or Spotorno score (Spearman's rank correlation test). There also was no statistical association between migration of the femoral stem at any time and sex, pain, site of pain, or severity of pain (visual analogue scale) using the Mann-Whitney U test. No significant relationship could be detected between any of the radiographic features in Table 1 and migration of me femoral stem (Mann- Whitney U test). No association between heterotopic bone and stem migration was demonstrated using Spearman's rank correlation test.
In the 1980s, many surgeons believed that failure of cemented hip arthroplasties in younger and more active patients made uncemented arthroplasty an attractive option. However, failure due to thigh pain was common and could be as high as 26% at 4 years using porouscoated implants.9 At the same time, improvements in cement techniques10 have thrown into question the need for uncemented implants at all. However, if a stable uncemented implant could be achieved so that early failure and thigh pain were not a problem, then cemented prostheses would have no advantages over uncemented prostheses, and failure in both groups eventually would be the result of endosteal lysis due to the generation of wear particles in most cases.
Fig 3: Radiograph showing nonprogressive sclerosis in Gruen zone 1 .
This study has shown that the incidence of thigh pain with the CLS femoral stem is low (2.3%) and that its presence is not related to excessive migration of the stem. Of the 15 patients with >3 mm migration at 2 years, only 1 patient complained of thigh pain. No specific radiographic features were identified in those patients who complained of thigh pain. Other studies of the CLS hip have shown it to have good short- and medium-term results,"·12 and Uns is supported by our study.
The migration rates are comparable to those of cemented Chamley and Stanmore hips5 and suggest that the design is inherently stable. Initial "high" migration between 0 and 3 months is to be expected with a cementless stem as bone resorption occurs at local high-stress regions and the prosthesis settles. Migration then slows, presumably due to the establishment of osseous integration in keeping with the design concept of the CLS stem.13
According to Rondia et al,14 osseous integration of the CLS stem is defined by calcar atrophy and shaft hypertrophy in Gruen zones 2 and 6, while fibrous integration is represented by a continuous sclerotic line around the entire prosthesis. An even line was said to represent stability, while an uneven line associated with localized reactive shaft hypertrophy at the tip of the stem implied instability.
The radiographic findings in our study differ, with only four cases of shaft hypertrophy. A continuous sclerotic line was encountered in only one case, and there were no cases in which the stem tip had produced a reactive shaft hypertrophy. In our study, radiographic appearance varied little over the study period, and there was no statistical relationship between any of the radiographic appearances and femoral stem migration or clinical outcome in terms of thigh pain.
Spotorno6 has defined the indications for a CLS stem as a Spotomo score <4. In our study, where the mean Spotorno score was 4.49, there was no correlation between the Spotorno score and migration of the stem or the presence of thigh pain. Furthermore, there have been no cases of mechanical failure. While this is encouraging, longer term follow-up is necessary before we are able to relate our measurements to failure.
The choice of which THR to use is beyond the scope of this study, but we believe migration data as well as clinical and radiographic results should be available for all prostheses. This method is a simple retrospective technique that could be used to produce migration profiles for other total hip prostheses.
1. Williams M, Frankel S, Nanchahal K, Coast J, Donovan J. Epidemiologically Based Needs Assessment. Total Hip Replacement. DHA Project: Research Programme Commissioned By the NHS Management Executive. Crown Publisher; 1992. Report 2a:87.
2. Murray DW, Carr AJ, Bulstrode CJ. Which primary total hip replacement? J Bone Joint Surg Br. 1995; 77:520-527.
3. Freeman MAR, Plante-Bordeneuve P. Early migration and late aseptic failure of proximal femoral prostheses. J Bone Joint Surg Br. 1994; 76:432-438
4. Selvik G. Roentgen stereophotogrammetry: a method for the study of kinematics of the skeletal system. Acta Orthop Scand. 1989; 232(suppl):l-51.
5. Walker PS, Mai SF. Cobb AG. Bentley G. Hua J. Prediction of clinical outcome of THR from migration measurements on standard radiographs. J Bone Joint Surg Br. 1995; 77:705-7 14.
6. Spotorno L, Schenk RK, Dietschi C. Romagnoli S, Mumenthaler A. Unsere Erfahrungen mit nicht-zementierten Prosthesen. Orthopade. 1987; 16:225-238
7. Nazarian S, Tisserand PH, Brunet CH. Muller ME. Anatomical basis of the transgluteal approach to the hip. In: Surgical-Radiologic Anatomy. Springer- Verlag; 1987:27-35.
8. Brooker AF. Bowerman JW, Robinson RA, Riley LH. Ectopic ossification following total hip replacement: incidence and a method of classification. J Bone Joint Surg Am. 1973; 55:1629.
9. Heekin RD. Callaghan JH. Hopkinson WJ, Savory CG, Xenos JS. The porous-coated anatomic hip prosthesis inserted without cement. Results after 5 to 7 years in a prospective study. J Bone Joint Surg Am. 1993; 75:77-91 .
10. Barrack RL. Mulroy RD. Harris WH. Improved cementing techniques and femoral component loosening in young patients with hip arthroplasty. J Bone Joint Surg Br. 1992; 74:385389.
11. Blasius K, Cotta H, Schneider U. Thomsen M. CLS multicentre study: 8 year results. Z Orthop Ihre Grenzgeb. 1993; 131:542547.
12. Spotorno L, Romagnloi S, Ivaldo N, et al. The CLS system. Theoretical concept and results. Acta Orthop BeIg. 1993; 59(suppl 0:144-148.
13. Blaha JD, Spotorno L, Romagnoli S. CLS press-fit arthroplasty. Techniques Orthop. 1991; 6(3):80-86.
14. Rondia J, Al bass ir A, Litt R, Le-Maire R. Le composant fémoral non cimenté CLS étude de 1 79 cas avec un recul de 2 à 5 ens. Acta Orthop BeIg. 1993; 59(suppl I):289-293.
Patients in the migration study for each year of follow-up
Radiographic features identified in relation to the femoral stems