Secondary Logo

Journal Logo


Early Migration Pattern of the Uncemented CLS Stem in Total Hip Arthroplasties

Ström, Håkan; Nilsson, Olle; Milbrink, Jan; Mallmin, Hans; Larsson, Sune

Author Information
Clinical Orthopaedics and Related Research: January 2007 - Volume 454 - Issue - p 127-132
doi: 10.1097/01.blo.0000238785.98606.9d


Micromotion has been shown to predict clinical loosening of hip arthroplasties.3 Such implant motion can be determined with radiostereometry analysis (RSA). However, in some studies, the first RSA examination (ie, the baseline examination) was not done until a few days after surgery.3,4,6,11 This means that there is a knowledge gap on the possible migration during the very early postoperative period, ie, from the end of surgery until the patients regain muscle control and start early mobilization. Radiostereometry analysis with the baseline examination performed directly after surgery, with the patient still anesthetized, has been done on internally fixed femoral neck fractures,8 but not on uncemented femoral hip stems.

Rehabilitation programs after an uncemented total hip arthroplasty (THA) usually include partial weightbearing for 6-12 weeks. The hypothesis behind this regimen is that early weightbearing might increase early migration of the stem and thereby prevent bone ingrowth. However, clinical studies supporting this theoretical concern have not been conclusive.5,7,9,11

The aim of the investigation was to measure the very early postoperative micromigration of an uncemented stem in THA, and determine if unrestricted weightbearing combined with intensive physiotherapy has an effect on the pattern of early postoperative micromigration when compared with restricted weightbearing combined with restrictive training.


Our investigation was approved by the local ethics committee before commencement (Approval Number Ups 99242). All patients gave informed consent. From 2000-2003, 29 patients younger than 65 years with primary hip osteoarthritis (OA) had THAs with implantation of the uncemented CLS hip stem (Centerpulse™, Bern, Switzerland). The CLS stem is a collarless, three-dimensional tapered, straight titanium alloy stem with a grit-blasted surface (Fig 1). The stem size increases with increasing femur size, and we used templates provided by the manufacturer to estimate the stem size before surgery. The distribution of stem size for the two study groups is shown in Table 1. Three different uncemented cups were used: 15 Allofit (Center-pulse™), 11 Interop (Centerpulse™), and three Trilogy (Zimmer™, Warsaw, IN). Patients who weighed more than 100 kg, who had concomitant disease of the musculoskeletal system, or who had any history of drug abuse were excluded. The surgery was done using an anterolateral approach. All patients received spinal anesthesia. Directly after surgery, the patients were randomized to unrestricted weightbearing (UWB) (n = 16; seven women; average age, 54.2 years) or to partial weightbearing (PWB) (n = 13; seven women; average age, 55.3 years) on the surgically treated leg, using a closed envelope technique. No stratification was done. The patients were part of a larger randomized study evaluating the effect of different weightbearing regimens on the micromotion of uncemented stems. Bone mineral density (BMD) was measured at the proximal portion of the surgically treated femur before surgery with DXA and DPX-L (GE-Lunar, Madison, WI). We evaluated regions of interest (ROI) of the femoral neck and total hip for BMD and Z scores (BMD in relation to an age-matched, gender-matched, and weight-adjusted reference population of US Caucasians) using the standard software provided by the manufacturer (GE-Lunar). Two patients (one in each study group) did not have DXA measurement for practical reasons. Because two patients in the PWB group did not have measurement of the surgically treated hips before surgery, we used the values from the untreated hip for these two patients. The initial RSA investigation, ie, the baseline investigation, was done immediately after surgery when the patient still was anesthetized. Paracetamol (Glaxo, King of Prussia, PA) and morphine (Meda, Solna, Sweden) were given for postoperative pain control when indicated. Use of nonsteroidal antiinflammatory drugs (NSAIDs) was avoided the first week. The patients were treated with low molecular weight heparin (Aventis Pharma, Strasbourg, France) as antithrombotic prophylaxis for 7 to 10 days. The UWB group was encouraged to participate in unrestricted early weightbearing from the first postoperative day combined with intensive physiotherapy training during the first 3 months. During the first 3 weeks after surgery, this group was instructed to do exercises daily that included hip flexion, extension, and abduction. Four to 6 weeks postoperatively, the training program included water exercises two to three times a week. At 7 to 12 weeks after surgery, the rehabilitation program was extended to include training on an ergometer bicycle, and defined active exercises in a gym with special emphasis on hip extension and abduction using three weight-loaded training machines. The patients in the UWB group were allowed to use crutches if needed. Patients in the PWB group were instructed to walk with a load of approximately 15 kg on the surgically treated leg (to walk with a load corresponding to the weight of the leg) and were allowed to train on their own based on instructions in a short written training program that focused on mobility through exercises such as supported hip flexion and extension. After the first 3 months, the patients in the PWB group were allowed unrestricted weight-bearing and were given training instructions, if needed. The training 3 months after surgery and later was individually based for both groups.

Fig 1
Fig 1:
The CLS stem is shown.
Distribution of Stem Size

Because of the intensive physiotherapy in the UWB group, only patients living in the neighborhood of the hospital could be included in the study. Another limitation was that the RSA facilities were not always available for the baseline investigations immediately after surgery. Therefore the study was not consecutive.

Five surgeons who were blinded to the postoperative regimen performed all the surgeries.

The RSA procedure has been described.4,10 At the time of surgery, five to eight tantalum ball markers (1 mm) were inserted in the proximal femur. The stems were marked with five tantalum balls; one at the tip, two on each side of the body, and two on each side of the hump distal to the neck. Radiostereometric analysis radiographs were taken with two x-ray tubes positioned with a 40° angle. Both radiographs were exposed simultaneously. The radiographs were digitized by a scanner (UMAX, Umeå, Sweden) and were measured using a semiautomated digital technique. We used the UMRSA computer program (RSA Innovations AB, Umeå, Sweden) to calculate migration. The translations and rotations of the rigid body of the stem in three different directions were calculated. The precision error (the value of precision) of the RSA measurements was determined on double examinations postoperatively, and a 99% confidence interval was chosen (Table 2).6 Comparisons were made between the two groups and within groups over time. The initial RSA was done within 1 hour after completion of surgery (before the spinal anesthesia had subsided). Additional RSA investigations were done at 1 day, 1 week, 1 month, 3 months, and 1 year postoperatively. No patient was lost to followup in the study. The RSA investigations were done by a specially trained nurse who was blinded for the randomization.

99% Confidence Limits for Significant Movements

Mann-Whitney U tests were used for comparison between groups, and two-tailed Student's t tests were used for comparison within groups over time, with the examination at 1 hour used as baseline. Probability values less than 0.05 were considered significant. Migration values are given as signed values with average ± standard deviations. We used an 80% power to calculate the difference that could be detected to 0.47 mm.


One patient in the UWB group died from a pulmonary embolus 15 days after surgery and was excluded from the study. No other perioperative or postoperative complications occurred. There were no significant differences in BMD of the neck or BMD of the total hip for the UWB or PWB groups: 1.05 (0.80-1.41) g/cm2 vs 0.97 (0.77-1-10) g/cm2 and 1.0 (0.8-1.25) g/cm2 vs and 0.95 (0.74-1.15) g/cm2, respectively. Both groups had normal age and gender-matched and weight-adjusted Z scores: 0.81 vs 0.18 and 0.0 vs 0.3. The patient who had revision surgery had a normal BMD, a neck BMD of 1.1 (Z score 0.8), and a total hip BMD of 1.09 (Z score 0.3). There was no significant micromigration during the first week in any of the groups. The UWB group had subsidence of −0.03 (range, −0.25-0.16) mm at 1 week, and the PWB group had subsidence of 0.01 mm (range, −0.05-0.8 mm). Retroversion at 1 week was 0.16° (range, −0.24°-0.59°) and 0.02° (range, −0.38°-0.39°) for the UWB and PWB groups, respectively. The migration in other directions was not significant. Patients in the UWB group did not have significant subsidence at 3 months (−0.69 mm; p = 0.08) but continued up to −1.01 mm (range, −5.86-0.04 mm) at 12 months (p = 0.04) (Fig 2). Patients in the PWB group had no significant subsidence up to 3 months (−0.47 mm), and remained stable at 12 months postoperatively (−0.51 mm; range, −3.58-0.34 mm) (Table 3). There was no statistical difference between the groups in subsidence at any time. The majority of the stems showed no subsidence (Fig 3), and stems that subsided were stable at 3 months (except for the stem that was revised which subsided continuously). Similarly, the rotation into retroversion at 1 year was 1.79° (range, −0.26°-13.19°) (p = 0.02) in the UWB group and 0.85° (range, −0.18°-5.23°) (p = 0.07) in the PWB group, the difference between the two groups was not significant (Fig 4). The angulation into varus was −0.44 mm (range, −3.66-0.10 mm) in the UWB group and −0.06 mm (range, −0.50-0.24 mm) in the PWB group at 1 year after surgery. There was a difference in anteroposterior (AP) translation between the two groups at 3 months, although the difference was small and at 1 year there was no difference between groups.

Fig 2
Fig 2:
The graph shows subsidence (mm) of the CLS stem in the UWB and PWB groups.
Migration of the Uncemented CLS Stem Measured with RSA
Fig 3A
Fig 3A:
B. The graphs show subsidence (mm) of the CLS stem for all patients in the (A) UWB and (B) PWB groups.
Fig 4
Fig 4:
The graph shows retroversion (degrees) of the CLS stem in the UWB and PWB groups.

The two groups showed the same migration pattern, which occurred mainly during the first two months (Fig 2). In the UWB group, eight of 16 stems had subsided more than the value of precision at 1 year. Of these eight stems, seven exceeded the value of precision at 3 months, and four stems exceeded the value 1 month after surgery. In the PWB group, four of 13 stems had subsided at 1 year after surgery, all four already had subsided at 3 months after surgery, and two of these had subsided at 1 month after surgery.


Most RSA studies on stem migration use data from an examination usually performed during the first week after surgery, but not immediately after surgery, as the baseline examination.3,4,6,11 We wanted to study the very early migration pattern of the uncemented CLS, with direct postoperative values as the baseline. Furthermore we wanted to compare early weightbearing and aggressive rehabilitation with restricted weightbearing in patients after surgery.

We did not find any significant micromotion of the CLS stems from the time of surgery until the day after surgery nor during the first week after surgery, regardless whether the patients were prescribed unrestricted or partial weight-bearing. The standard start point for the first RSA measurement within 1 week after THA therefore seems to be sufficient when investigating uncemented femur implants.

Nearly all stems (11 of 12) that subsided more than the value of precision of the RSA (0.24 mm) at 1 year also exceeded the value of precision at 3 months after surgery. Of the eight stems that subsided more than 1 mm at 1 year postoperatively, six also had subsided more than 1 mm at 3 months. We think the 3-month RSA results are a good predictor of the subsidence that will be seen 1 year after surgery. This finding is in contrast to that of Davies et al2 who described continuous migration up to 1 year with the same stem we studied. The small early subsidence followed by stabilization indicates that after surgery, the stems became stabilized by the load. This pattern was the same in both study groups despite the fact that the patients in the PWB group were not allowed full weightbearing during the first 3 months after surgery.

Despite finding no substantial differences between the groups, stems in the UWB group tended to subside and rotate backward into retroversion slightly more than the stems in the PWB group. Most of the migration occurred between 1 week and 3 months. However, both groups had increasing retroversion between 3 months and 1 year. A similar migration pattern has been described before.1,11,13

Migration greater than 1-1.5 mm in cementless stems during the first 2 years after surgery has been shown to predict an increased risk of early or midterm revision.4 The mean subsidence in the UWB group was 1.01 mm at 1 year, thereby, approaching this limit. Most subsidence occurred between 1 week and 3 months. The major part of the migration occurred in six stems in the UWB group and four stems in the PWB group. One stem in each group subsided up to 6 mm by 3 months, but then remained stable. The only stem that showed continuous subsidence was a stem that was revised 1.5 years after surgery, which was after the end of the current study. The remaining implants in the UWB group remained stable and had the same migration pattern during the first postoperative year as the PWB group. The high precision of RSA measurements allows for small numbers of patients in studies focused on specific questions (eg, implant design). However, when more complex problems are studied, such as the effects of weightbearing on implant stability, the number of patients may have to be increased to avoid Type 2 errors. We found no difference between the two regimens regarding the pattern of micromotion or migration rates, but there was a tendency toward slightly greater values in the UWB group. However, the number of patients in each group was small which makes the conclusions for the two groups less reliable.

Because the CLS stems were stable from 1 hour postoperatively in both groups, the first week can be used as a starting point for RSA measurements. The 3-month RSA results predict subsidence up to 1 year postoperatively, although retroversion seems more difficult to predict. The migration pattern was similar in the PWB group compared with the UWB group, with slightly less migration in the PWB group. There was no statistical difference between the groups. Early weightbearing did not seem to affect micromigration. However, the slightly greater values, loosening of one stem in the UWB group, and the small study groups are concerns. A larger group of patients with longer periods of observation should be analyzed before making definite conclusions concerning the safety of early full weightbearing.


1. Carlsson AS, Gentz CF, Stenport J. Fracture of the femoral prosthesis in total hip replacement according to Charnley. Acta Orthop Scand. 1977;48:650-655.
2. Davies MS, Parker BC, Ward DA. Migration of the uncemented CLS femoral component. Orthopedics. 1999;22:225-228.
3. Kärrholm J, Borssen B, Löwenhielm G, Snorrason F. Does early micro motion of femoral stem prostheses matter? 4-7-year stereo-radiographic follow-up of 84 cemented prostheses. J Bone Joint Surg Br. 1994;76:912-917.
4. Kärrholm J, Herberts P, Hultmark P, Malchau H, Nivbrant B, Thanner J. Radiostereometry of hip prostheses: review of methodology and clinical results. Clin Orthop Relat Res. 1997;344:94-110.
5. Kishida Y, Sugano N, Sakai T, Nishii T, Haraguchi K, Ohzono K, Yoshikawa H. Full weight-bearing after cementless total hip arthroplasty. Int Orthop. 2001;25:25-28.
6. Mjöberg B. Loosening of the cemented hip prosthesis: the importance of heat injury. Acta Orthop Scand Suppl. 1986;221:1-40.
7. Pilliar RL, Maniatopoulos JM. Observations on the effect of movement on bone ingrowth into porous-surfaced implants. Clin Orthop Relat Res. 1986;208:108-113.
8. Ragnarsson J, Hansson L, Karrholm J. Stability of femoral neck fractures: a postoperative roentgen stereophotogrammetric analysis. Acta Orthop Scand. 1989;60:283-287.
9. Rao RR, Sharkey PF, Hozack WJ, Eng K, Rothman RH. Immediate weightbearing after uncemented total hip arthroplasty. Clin Orthop Relat Res. 1998;349:156-162.
10. Selvik G. Roentgen stereophotogrammetry: a method for the study of the kinematics of the skeletal system. Acta Orthop Scand Suppl. 1989;232:1-51.
11. Ström H, Mallmin H, Pétren-Mallmin M, Milbrink J, Kolstad K. The cone hip stem: a prospective study of 13 patients followed for 5 years with RSA. Acta Orthop Scand. 2003;74:525-530.
12. Woolson S, Adler N. The effect of partial or full weightbearing ambulation after cementless total hip arthroplasty. J Arthroplasty. 2002;17:820-825.
13. Wroblewski BM. The mechanism of fracture of the femoral prosthesis in total hip replacement. Int Orthop. 1979;3:137-139.
© 2007 Lippincott Williams & Wilkins, Inc.