Background: In revision total knee arthroplasty (TKA), the femoral component and tibial baseplate are usually cemented. However, stems can be fixed either with cement or with a press-fit technique, with the latter resulting in hybrid fixation. There is no consensus on the preferred stem fixation technique. Therefore, we compared the stability of cemented TKA implants with the stability of TKA implants fixed with the hybrid technique in a prospective randomized trial using radiostereometric analysis (RSA).
Methods: Thirty-two patients with a Type-I or II bone defect who needed revision TKA and were randomly allocated at the time of surgery into either the cemented or hybrid-fixation group were included in the study. The radiographs for the RSA were obtained during hospitalization (baseline); at 6 weeks; and at 3, 6, 12, and 24 months. Migration of the femoral and tibial implants was measured using model-based RSA and expressed along or around the 3 orthogonal axes and as total translation (TT) and total rotation (TR). Clinical results were evaluated using the Knee Society Score (KSS), the Knee injury and Osteoarthritis Outcome Score (KOOS), active flexion, and visual analog scale (VAS) scores for pain and satisfaction. Mann-Whitney and chi-square tests were used to compare migration and clinical outcomes between the cement and hybrid techniques.
Results: At 24 months, no difference in median migration or the number of migrating components was found between the cemented and hybrid-fixation groups. In each group, approximately one-third of the tibial components had total rotation of >1°. The clinical scores did not differ between the techniques.
Conclusions: At 24 months after revision TKAs, cemented and hybrid-fixation replacements were equally stable. Unexpectedly, both groups had implants with >1 mm or >1° of micromotion although there were no clinical or radiographic signs of loosening. Whether these findings indicate the possibility of loosening with longer follow-up remains to be investigated.
Level of Evidence: Therapeutic Level I. See Instructions for Authors for a complete description of levels of evidence.
1Departments of Research (P.J.C.H.) and Orthopaedics (A.B.W. and G.G.v.H.), Sint Maartenskliniek, Nijmegen, the Netherlands
E-mail address for P.J.C. Heesterbeek: firstname.lastname@example.org
Achieving reliable implant fixation is critical to the success of revision total knee arthroplasty (TKA). Stems provide the possibility of bypassing damaged bone, securing diaphyseal fixation, reducing proximal strain, improving axial alignment, and permitting the use of augments and adjustment of component offset1-5. Numerous technical options are available, but the main issue remains: how to fix the stem? The femoral component and tibial baseplate are usually cemented whereas the stems are either cemented or press-fit, with the latter producing hybrid fixation. We are not aware of any evidence indicating which construct provides better stability; several investigators have reported that both options are at least clinically equivalent4,6-8.
The pros and cons of cemented compared with hybrid constructs have been reported in the literature. Hybrid fixation has the potential for bone stock preservation and enables better alignment of the intramedullary axis, whereas cemented stems allow more variability of component positioning to prevent lateral oversizing and/or excessive length6,8-10, which can be countered by using offset stems8,11,12. A disadvantage of hybrid fixation is the potential for end-of-stem pain13, although this can be diminished by the use of slotted stems14. The ideal fixation of modular revision TKA remains unclear. Hybrid and all-cement fixation techniques have been reported to provide equal stability, and experimental and clinical results (survival, aseptic loosening, and clinical outcome) have been reported to be comparable8,15. However, a need for high-quality research on this topic persists because of the low levels of evidence in previous studies8,15.
We are not aware of any reports of a clinical comparison between hybrid fixation and all-cement techniques in revision TKA in patients with (1) constructs of the same length, (2) the same Anderson Orthopaedic Research Institute (AORI) type of bone defect as well as an intact diaphysis and metaphysis, (3) implants with the same level of constraint, and (4) 1 revision only (i.e., re-revisions were excluded). In addition, the literature that we reviewed provided no precise outcome measures and the levels of evidence were low8,15. Therefore, we performed a randomized controlled trial comparing cemented and hybrid constructs in terms of implant stability and clinical outcome in patients with revision TKA. The hypothesis that we tested was that the use of either an all-cement or a hybrid fixation technique would result in equal implant micromotion and functional outcomes.
Materials and Methods
This single-center, patient-blinded, parallel-group study with an allocation ratio of 1:1 was conducted from January 2008 to April 2010 at Sint Maartenskliniek, Nijmegen, the Netherlands. Block randomization (block sizes of 6, 4, and 4) was performed by a computer-generated random-number list prepared by an independent investigator. The allocation sequence was concealed, in sequentially numbered, opaque, sealed envelopes, from the surgeons and research nurse enrolling and assessing participants. During surgery, on preparation for implantation of the stems, the envelope was opened by the surgical technician to reveal the allocation.
To detect a difference of 1 mm or 1° of micromotion at 1 year (which was arbitrarily considered to be clinically relevant and feasible with regard to the accuracy and precision) with a 2-sided 5% significance level and a power of 90%, a sample size of 16 patients per group was necessary; this took into account an anticipated number of missing data because of marker occlusion.
The study protocol was approved by the hospital’s investigational review board and the Arnhem-Nijmegen Medical Ethical Review Board (2007/083). The trial is registered at the Nederlands Trial Register (http://www.trialregister.nl), identified with number NTR1315. This study was conducted in accordance with the Declaration of Helsinki, CONSORT (Consolidated Standards of Reporting Trials) guidelines, and ISO (International Organization for Standardization) 16087:2013 for radiostereometric analysis (RSA). Written informed consent was obtained from all participating patients.
Eligible participants were adults on the waiting list for a total system revision of a primary TKA for whom either implantation technique (all-cement or hybrid) was indicated. Patients had to be in stable health and free from, or under treatment for (with stabilization of their condition), cardiac, pulmonary, hematological, or any other conditions that might pose an excessive operative risk or affect compliance with the standard postoperative rehabilitation protocol. Exclusion criteria were radiographically or perioperatively defined Type-III bone loss (deficient metaphyseal bone with bone loss compromising a major portion of the condyle or plateau, occasionally with detachment of a collateral ligament or patellar tendon) according to the Anderson Orthopaedic Research Institute (AORI) classification16, a body mass index (BMI) of >35 kg/m2, or a known sensitivity to materials in the device.
All patients received the Legion revision TKA (Smith & Nephew). All surgical procedures were performed by 2 experienced orthopaedic knee surgeons (A.B.W. and G.G.v.H.) who had >10 years’ experience using both the all-cement and hybrid techniques. All patients received spinal anesthesia or regional femoral and sciatic nerve blocks. The femoral components and tibial baseplates were cemented in all cases, whereas the patella was selectively resurfaced, with 14 patellae remaining unresurfaced. Instrumentation specific to both stem-fixation techniques was used as described in the surgical technique brochure provided by the manufacturer (Smith & Nephew).
The all-cement treatment involved fully cementing the stems of both the tibia and the femoral component. A Weber PE (polyethylene) cement plug (Zimmer) was placed, after which pulse lavage was performed, vacuum-mixed Palacos R+G or COPAL cement (Heraeus Medical) was injected without a pressurizer and with a retrograde technique, and the femoral and tibial implants were inserted; at least 1 cm of cement was placed between the cement plug and the end of the stem.
With the hybrid implantation technique (a cemented femoral component or tibial baseplate with a press-fit stem), the diaphyseal fixation was obtained using a canal-filling stem in order to provide a substantial length of diaphyseal cortical contact. The stem thickness was patient-specific, and was chosen either by fitting the stem into the medullary canal with cement (with the cemented stems available in 2-mm increments) or by placement of a press-fit stem (available in 1-mm increments). A shorter stem length was chosen if an offset coupler (length, 30 mm) was used (in both cemented and hybrid constructs), and a longer stem length was chosen if no offset coupler was used. The total length of the construct was 150 or 160 mm for the femur and 120 or 130 mm for the tibia, depending on the use of an offset coupler (Fig. 1). All of the longer, press-fit stems (no offset coupler) were slotted stems. Slotted shorter stems were not available.
All procedures were performed in 1 stage except in 3 patients with infection as the indication for the revision, who were treated with a 2-stage revision. All patients received the same preoperative, perioperative, and postoperative standard care; they were allowed full weight-bearing using crutches immediately postoperatively. After preparation of the bone and before placement of the implants, 1-mm tantalum beads for RSA were placed in the femur and tibia (Fig. 2).
The primary end point with respect to implant stability was micromotion, measured with RSA using a uniplanar setup with ceiling-mounted x-ray tubes. Patients were in a supine position, with standardized foot rotation to enable marker visibility throughout the follow-up period. Migration was expressed along or around the 3 orthogonal axes: longitudinal, transverse, and sagittal (Fig. 3).
Laser-scanned, reverse-engineered models were used. Accuracy was determined with a phantom study and, for the femoral component, was ±0.063, ±0.021, and ±0.167 mm for translation along the transverse, longitudinal, and sagittal axes, respectively, and ±0.127°, ±0.095°, and ±0.042° for rotation around these axes. For the tibial component, the accuracy was ±0.043, ±0.030, and ±0.075 mm for translation along the transverse, longitudinal, and sagittal axes, respectively, and ±0.079°, ±0.122°, and ±0.072° for rotation around these axes. Double examinations were carried out for all subjects at the 6-week follow-up evaluations to assess precision. Precision was calculated as the 95% confidence interval (1.96 × standard deviation [SD]) around the mean migration between the two examinations (Table I). Patients were evaluated preoperatively and at 6 weeks (±1 week), 3 months (±2 weeks), 6 months (±3 weeks), 1 year (±1 month), and 2 years (±2 months) postoperatively.
Model-based (MB) RSA measurements were performed with MBRSA software (RSAcore). Migration of the femoral component was calculated with reference to the femoral bone, and migration of the tibial component was calculated with reference to the tibial bone. Since the stem is firmly attached to the femoral component or tibial baseplate, the two are considered to behave as one rigid body. A cutoff level of <150 was used for the condition number. Total translation (TT) was calculated as TT = √(Tx2+Ty2+Tz2) and total rotation (TR) was calculated as TR ≈ √(Rx2+Ry2+Rz2)17, with x, y, and z defined as shown in Figure 3, and these were used as summarizing variables for the translation along and rotation around the 3 axes.
During all follow-up visits, the research nurse collected data during a physical examination. The clinical, functional, and total Knee Society Score (KSS) and active flexion of the affected knee were assessed. The patients marked visual analog scales (VASs) for pain and satisfaction and filled out the Knee injury and Osteoarthritis Outcome Score (KOOS)18. Mid-thigh or mid-shin pain was scored as “yes” or “no” at 6, 12, and 24 months postoperatively and patients were considered to have such pain when they answered “yes” at one or more of the follow-up visits.
Statistically, all RSA data were assessed as not having a normal distribution; the medians and interquartile ranges of the TT and TR are presented graphically. The primary end points were the TT and TR values at 2 years. Additional analyses were done to compare the number of migrating implants according to different thresholds of TT and TR (0.5, 1.0, 1.5, and 2.0 mm or degrees) between the cemented and hybrid fixation groups at the 2-year follow-up assessment. Groups were compared using the Wilcoxon-Mann-Whitney test, and categorical values were compared using the chi-square or Fisher exact test. Medians and ranges are presented for the RSA results as well as for the clinical outcome parameters. During the trial, no interim analyses were carried out. P values of <0.05 were considered significant.
The patient flow diagram is presented in Figure 4. Of the 34 patients randomized, 1 did not receive the allocated intervention and 1 (with a hybrid stem) was excluded from the analysis because the size of this patient’s femoral component was not available for MBRSA measurements. Of the 32 patients included in the study, 30 (15 with cemented stems and 15 with hybrid stems) completed the 2-year follow-up. There was no statistically significant difference between the treatment groups with regard to the baseline demographics, clinical characteristics, or indications for revision surgery, except for active flexion of the affected knee (Tables II and III19).
At 2 years, the median TTfemur and TRfemur were 0.31 mm (range, 0.15 to 1.14 mm) and 0.62° (range, 0.13° to 2.10°) for the cemented constructs and 0.45 mm (range, 0.13 to 0.89 mm) and 0.57° (range, 0.13° to 1.12°) for the hybrid group, with no significant difference between the groups (p = 0.66 for both TT and TR). The median TTtibia and TRtibia were 0.40 mm (range, 0.14 to 1.02 mm) and 0.86° (range, 0.24° to 1.88°) for the cemented group and 0.44 mm (range, 0.16 to 2.05 mm) and 0.65° (range, 0.28° to 2.95°) for the hybrid group. Again, there was no significant difference between the treatment groups (p = 0.68 and p = 0.89 for TTtibia and TRtibia, respectively).
The median translations and rotations at all follow-up intervals for both groups (Tables IV and V) were at or below the precision limits, indicating that there was not a significant amount of implant migration on the group level.
The number of implants showing translation or rotation of >1 mm or >1° was comparable between the groups at all follow-up interval (p > 0.26 for the femoral implants and p > 0.32 for the tibial implants). At 2 years, 2 cemented femoral components had translated >1 mm compared with 0 hybrid femoral components and 3 cemented femoral components had rotated >1° compared with 1 hybrid femoral component (Fig. 5). Two tibial components (1 cemented and 1 hybrid) had translated >1 mm at 2 years and 10 (5 cemented and 5 hybrid) had rotated >1° (Fig. 6). Also, no differences between the groups were found when the numbers of migrating components were compared on the basis of different thresholds of TT and TR (0.5, 1.0, 1.5, and 2.0 mm or degrees) (Figs. 7 and 8).
At 2 years, the clinical outcomes were comparable between the all-cement and hybrid fixation techniques (Table VI). In the cemented group, 8 patients reported mid-thigh or mid-shin pain during at least 1 follow-up visit; in the hybrid group, 6 patients reported mid-thigh or mid-shin pain.
The occurrence and types of serious adverse device-related effects are listed in Table VII. The most frequent was persistent or increasing pain, with some patients referred to a pain clinic. There was no difference in the occurrence of serious adverse device-related effects between the 2 groups (p = 0.29).
In this randomized controlled trial, we compared implant stability and clinical outcomes between fully cemented revision TKAs and revision TKAs performed with a hybrid fixation technique. The 2 procedures were done with the same implant, the same length of construct, and the same level of implant constraint in patients with a Type-I or II bone defect. Implant migration was similar between the 2 groups at the 2-year follow-up evaluation. The median translation and rotation remained at or below the precision level in both groups, indicating that the amount of implant migration was not significant at the group level. In addition, the number of implants showing >1 mm of translation or >1° of rotation was comparable between the groups. Our hypothesis that there was equal implant stability could not be rejected.
By studying patients with the same length of construct, same level of bone loss, and same level of implant constraint; by excluding re-revisions; and by using a randomized design and a precise measurement technique we were able to focus on the optimal stem fixation strategy. We believe this to be the first and only randomized controlled trial in which implant micromotion of revision TKAs was measured with RSA and thus that there is no literature with which to compare our results. However, this topic has been investigated in some studies with a lower level of evidence (III or IV), including a review and a recent meta-analysis, which showed the implant survival rate, rate of aseptic loosening, and clinical outcomes to be comparable between the cemented and hybrid techniques8,15. The conclusions of our study support those findings. However, we were surprised by the high number of fully cemented constructs with micromotion of >1 mm or >1° as we expected “rock solid” constructs. It is unclear whether the relatively high number of components (in both groups) that migrated >1 mm or >1° will eventually have loosening, or whether the arbitrary choice of 1 mm or 1° to indicate migration was too conservative. In a previous study of primary tibial components, a maximal total point motion of >1.6 mm indicated an unacceptable risk of revision at 5 and 10 years20. Extended follow-up of the patients in the current study continues, with collection of 5 and 10-year results, to ascertain whether the amounts of migration found in the present study will lead to early re-revisions.
Clinical scores were comparable with those found in other studies8, although some patients in the present study did have substantially lower scores. Mid-thigh or mid-shin pain was reported by almost half of the patients but was equally distributed between the groups and therefore not specific for the hybrid fixation. The prevalence was a bit higher than has been reported in the literature21, but counting a patient as having mid-thigh or shin pain when he/she reported it at one or more follow-up visits might have resulted in an overestimation. Although one explanation may be a low-threshold referral strategy to our own pain clinic, the high prevalence, particularly in the group with cemented constructs, warrants further investigation.
This study has some limitations. First, during the course of the trial an additional exclusion criterion was added (the need for a 2-stage revision). This change was made because the surgeons became concerned that they might be cementing stems in patients who had an infection at the site of the implant. Since these patients were excluded before randomization took place, we do not believe this affected the outcome at the group level, but it is still a flaw. Second, the baseline RSA was not always performed on the same postoperative day because of logistic reasons, although it was always carried out after mobilization and as late as possible during the hospital stay. There were no differences between the groups in terms of the day of the RSA, and we are confident that this did not introduce bias. Third, the sample size in the present study was small, a problem inherent to RSA studies. While the sample was sufficient for investigation of micromotion, no conclusions can be drawn regarding clinical outcomes. Fourth, this study was limited to patients with Type-I, IIA, or IIB defects. We cannot predict the results for patients with Type-III osseous defects, for those undergoing re-revision, or for those treated with other stem lengths or in whom hybrid fixation techniques with secondary osseous ingrowth possibilities were used. The total length of the construct was not the same for all patients; it differed by 10 mm between those treated with and those treated without an offset coupler. Although the stem lengths were equally distributed between the randomization groups, theoretically this could have led to bias as randomization took place before the decision to use offset couplers. However, Wood et al. found results of hybrid and cement fixation techniques to be comparable irrespective of the degree of constraint, use of augments, or length or thickness of the stems4.
The results of our study show that both stem fixation techniques resulted in equally stable constructs. Recent development of ingrowth metaphyseal sleeves and porous metal cones has resulted in a trend toward using shorter stems in combination with solid metaphyseal fixation22,23. Stability of this type of construct compared with the ones that we investigated will likely stimulate debate and will be a topic of future investigations.
In conclusion, all-cement and hybrid fixation of revision TKAs resulted in equal stability at 24 months in patients with a Type-I or II bone defect, the same length of construct, and the same level of implant constraint. There were implants with micromotion of >1 mm or >1° in both groups, but there were no clinical or radiographic signs of loosening. Whether the micromotion might result in loosening at a later follow-up interval remains to be investigated.
NOTE: The authors thank research nurse Saskia Susan for her efforts in patient recruitment and data management.
Investigation performed at the Departments of Orthopaedics and Research, Sint Maartenskliniek, Nijmegen, the Netherlands
Disclosure: The institution of the authors received funding for this study from Smith & Nephew (Memphis, Tennessee), the manufacturer of the device used in this study. Smith & Nephew had no role in interpretation or reporting of the results of the study. On the Disclosure of Potential Conflicts of Interest forms, which are provided with the online version of the article, one or more of the authors checked “yes” to indicate that the author had a relevant financial relationship in the biomedical arena outside the submitted work; “yes” to indicate that the author had a patent and/or copyright, planned, pending, or issued, broadly relevant to this work; and “yes” to indicate that the author had other relationships or activities that could be perceived to influence, or have the potential to influence, what was written in this work.
1. Sah AP, Shukla S, Della Valle CJ, Rosenberg AG, Paprosky WG. Modified hybrid stem fixation in revision TKA is durable at 2 to 10 years. Clin Orthop Relat Res. 2011 ;469(3):839–46. Epub 2010 Sep 28.
2. Haas SB, Insall JN, Montgomery W 3rd, Windsor RE. Revision total knee arthroplasty with use of modular components with stems inserted without cement. J Bone Joint Surg Am. 1995 ;77(11):1700–7.
3. Jazrawi LM, Bai B, Kummer FJ, Hiebert R, Stuchin SA. The effect of stem modularity and mode of fixation on tibial component stability in revision total knee arthroplasty. J Arthroplasty. 2001 ;16(6):759–67.
4. Wood GC, Naudie DDR, MacDonald SJ, McCalden RW, Bourne RB. Results of press-fit stems in revision knee arthroplasties. Clin Orthop Relat Res. 2009 ;467(3):810–7. Epub 2008 Nov 26.
5. Morgan-Jones R, Oussedik SIS, Graichen H, Haddad FS. Zonal fixation in revision total knee arthroplasty. Bone Joint J. 2015 ;97-B(2):147–9.
6. Gililland JM, Gaffney CJ, Odum SM, Fehring TK, Peters CL, Beaver WB. Clinical & radiographic outcomes of cemented vs. diaphyseal engaging cementless stems in aseptic revision TKA. J Arthroplasty. 2014 ;29(9)(Suppl):224–8. Epub 2014 May 24.
7. Greene JW, Reynolds SM, Stimac JD, Malkani AL, Massini MA. Midterm results of hybrid cement technique in revision total knee arthroplasty. J Arthroplasty. 2013 ;28(4):570–4. Epub 2012 Nov 24.
8. Beckmann J, Lüring C, Springorum R, Köck FX, Grifka J, Tingart M. Fixation of revision TKA: a review of the literature. Knee Surg Sports Traumatol Arthrosc. 2011 ;19(6):872–9. Epub 2010 Sep 7.
9. Peters CL, Hennessey R, Barden RM, Galante JO, Rosenberg AG. Revision total knee arthroplasty with a cemented posterior-stabilized or constrained condylar prosthesis: a minimum 3-year and average 5-year follow-up study. J Arthroplasty. 1997 ;12(8):896–903.
10. Leta TH, Lygre SH, Skredderstuen A, Hallan G, Furnes O. Failure of aseptic revision total knee arthroplasties. Acta Orthop. 2015 ;86(1):48–57. Epub 2014 Sep 30.
11. Conlisk N, Gray H, Pankaj P, Howie CR. The influence of stem length and fixation on initial femoral component stability in revision total knee replacement. Bone Joint Res. 2012 ;1(11):281–8. Epub 2012 Nov 1.
12. Nanson CJ, Fehring TK. Stem fixation in revision total knee arthroplasty. Tech Knee Surg. 2009;8(3):161–5.
13. Barrack RL, Rorabeck C, Burt M, Sawhney J. Pain at the end of the stem after revision total knee arthroplasty. Clin Orthop Relat Res. 1999 ;367:216–25.
14. Barrack RL, Stanley T, Burt M, Hopkins S. The effect of stem design on end-of-stem pain in revision total knee arthroplasty. J Arthroplasty. 2004 ;19(7)(Suppl 2):119–24.
15. Wang C, Pfitzner T, von Roth P, Mayr HO, Sostheim M, Hube R. Fixation of stem in revision of total knee arthroplasty: cemented versus cementless-a meta-analysis. Knee Surg Sports Traumatol Arthrosc. 2015 . Epub 2015 Dec 19.
16. Engh GA, Ammeen DJ. Classification and preoperative radiographic evaluation: knee. Orthop Clin North Am. 1998 ;29(2):205–17.
17. Stilling M, Madsen F, Odgaard A, Rømer L, Andersen NT, Rahbek O, Søballe K. Superior fixation of pegged trabecular metal over screw-fixed pegged porous titanium fiber mesh: a randomized clinical RSA study on cementless tibial components. Acta Orthop. 2011 ;82(2):177–86. Epub 2011 Mar 24.
18. Roos EM, Lohmander LS. The Knee injury and Osteoarthritis Outcome Score (KOOS): from joint injury to osteoarthritis. Health Qual Life Outcomes. 2003;1:64. Epub 2003 Nov 3.
19. van Kempen RW, Schimmel JJ, van Hellemondt GG, Vandenneucker H, Wymenga AB. Reason for revision TKA predicts clinical outcome: prospective evaluation of 150 consecutive patients with 2-years followup. Clin Orthop Relat Res. 2013 ;471(7):2296–302. Epub 2013 Mar 30.
20. Pijls BG, Valstar ER, Nouta KA, Plevier JW, Fiocco M, Middeldorp S, Nelissen RG. Early migration of tibial components is associated with late revision: a systematic review and meta-analysis of 21,000 knee arthroplasties. Acta Orthop. 2012 ;83(6):614–24. Epub 2012 Nov 9.
21. Mihalko WM, Whiteside LA. Stem pain after cementless revision total knee arthroplasty. J Surg Orthop Adv. 2015 ;24(2):137–9.
22. Lachiewicz PF, Soileau ES. A 30-mm cemented stem extension provides adequate fixation of the tibial component in revision knee arthroplasty. Clin Orthop Relat Res. 2015 ;473(1):185–9.
Copyright 2016 by The Journal of Bone and Joint Surgery, Incorporated
23. Patel AR, Barlow B, Ranawat AS. Stem length in revision total knee arthroplasty. Curr Rev Musculoskelet Med. 2015 ;8(4):407–12.