Role and results of tapered fluted modular titanium stems in revision total hip arthroplasty

Tapered, fluted, modular, titanium (TFMT) stems have become the stem of choice for femoral revision total hip arthroplasty (THA) at our institution. They are relatively simple to use, versatile and offer reliable fixation across the spectrum of femoral deficiency. Compared to our experience with cylindrical, non-modular cobalt-chrome (CNCC) stems, we have demonstrated superior patient-reported quality of life outcome scores, improved preservation or restoration of bone stock, and lower rates of insertional intraoperative fracture.^1,2 Severe femoral deficiency with proximal bone loss remains a challenging management problem. These femoral deficiencies are classified by Paprosky as type-IIIB when there is less than 4 cm of diaphyseal bone available for distal fixation and type-IV when there is extensive diaphyseal damage with a widened canal and non-supportive isthmus.³ It has been suggested that TFMT stems may be the best option when revising these types of femoral deficiency.⁴ Our latest review of results for this group of patients revised using TFMT stems has shown an excellent rate of survival with similar maintenance of bone stock seen with less severe femoral deficiency. We present a review of our experience with TFMT stems used in revision THA at our institution over the last decade.

Patients and methods

In 2006 we published our early experience with TFMT stems² reporting outcomes from a cross-sectional matched cohort study. This compared 31 patients revised with a TFMT stem (ZMR Hip System; Zimmer, Warsaw, Indiana) to 189 patients revised with a CNCC stem (Solution System; DePuy, Warsaw, Indiana). Surgery was performed between August 1998 and December 2003. At this time the CNCC stem was used if > 4 cm of scratch fit could be obtained while the modular TFMT stem was used when < 4 cm of bone was available. Both groups were matched according to length of follow-up, age and comorbidity. Patients completed pre-operative questionnaires one to seven days prior to surgery and these were similar for the two groups. Patient-reported quality of life (QOL) was assessed post-operatively using five, validated scores: the Oxford Hip Score, the WOMAC Osteoarthritis Index, the SF-12, the Hip and Knee Arthroplasty Satisfaction questionnaire, and the UCLA activity score.

In 2010² following increased use of the TFMT stems, we published a retrospective cross-sectional cohort study of 109 patients revised using the same TFMT stem compared to 115 revised to a CNCC stem between January 2000 and March 2006. This included patients from the previous study. Ninety-five and 105 patients respectively were available for detailed follow-up. The study design included the same pre and post-operative assessment of QOL but evolved to assess complications and radiological evidence of changes in bone stock. Again the groups were matched for follow-up, age and comorbidity. Importantly, minimum follow-up was two years for both groups compared to one year in the earlier study. Mean follow-up was 37 months (24 to 83) for the TMFT group and 49 months (24 to 95) for the CNCC group. The pre-operative diagnosis was aseptic loosening in > 75% of cases with infection and fracture in the remainder. Radiographs were assessed for pre-operative femoral deficiency according to the classification of Paprosky,³ intra-operative fracture, subsidence and post-operative fixation according to the criteria of Engh et al.⁵ Post-operative changes in bone stock were classified using a method described by Bohm and Bischel⁶ as A: increasing defects; B: constant defects: or C: osseous restoration. There were significantly more (p <  0.0001) type-IIIB fractures in the TFMT cohort (58 cases vs 31).

Finally in 2012,⁷ we revisited the cohort published in 2010 and selected those patients in the TFMT group with type-IIIB and IV femoral deficiency. Sixty-five patients were identified and 48 of these were available for review with mean follow-up of 84 months (60 to 120). A total of 71% were revised for aseptic loosening, 12% for periprosthetic fracture and 16% for infection (2-stage revision). The methods used in the earlier study to assess QOL and pre and post-operative radiographs were reproduced. In 60% of cases, the stems were of the standard ZMR design, while 40% were the newer ZMR-XL system that has a strengthened modular junction. Cortical allograft struts were used in one third of cases to reconstruct the proximal femur.

Results

The results in 2006¹ were very encouraging. All QOL scores favoured the TFMT stem group (Table I). Significantly higher WOMAC pain (p = 0.03) and function (p = 0.02) scores were observed. Oxford Hip (p = 0.006) and satisfaction (p < 0.0001) scores were also significantly higher in the TFMT group.

For the larger cohort reported in 2010² there were six TFMT stem revisions of 109 patients (6%) including four implant fractures, one infection and one aseptic loosening. In the CNCC group ten stems from 115 (9%) were revised including two implant fractures, four infections and four aseptic loosening. Superior QOL outcome scores were again observed in the TFMT group (Table 1). Despite the TFMT stems being used in cases with more severe bone loss, WOMAC pain (p = 0.04) and stiffness scores (p = 0.009), Oxford Hip (p = 0.008) and overall satisfaction (p = 0.009) scores were significantly higher in the TFMT group.

There were more (p = 0.042) intra-operative fractures in the CNCC group (29 of 114) than the TFMT group (9 of 103). Dislocation was similar for both groups (12 of 114 for the CNCC group, 8 of 103 for the TFMT group). The use of an ETO did not increase the risk of fracture or dislocation.

More patients in the TFMT group (p < 0.0001) had an increase in osseous restoration while more patients in the CNCC group (p = 0.02) had an increase in bone loss (Table II). All surviving implants in both groups were well fixed.

For the final analysis of TFMT stems used in severe bone loss,⁷ survival at five years was 94%. At ten years survival was 84% after five stems failed due to fracture at the modular junction. Average time to fracture was three years and all occurred with the standard ZMR stem. Body mass index (BMI) was elevated in these patients to a mean of 35 (30 to 40). Stem diameter was 19 mm or less in 4 cases and only one case had been supported with an allograft strut. There was no evidence of subsidence beyond the first year. Six patients had early subsidence with a mean of 11 mm but no difference in functional outcome could be demonstrated when compared with stems that had not subsided. No stem was revised for subsidence or loosening. There was no significant decrease in the QOL scores during the period of follow-up.

Discussion

TFMT stems have demonstrated excellent performance over the medium term. In our institution they have demonstrated excellent survival and radiological results with superior QOL outcomes.

Bone stock following revision with TFMT stems has been observed to increase. This must be viewed within the limitations of a plain radiograph’s ability to detect bone mineral density change. Regardless, the severe stress shielding observed with CNCC stems has not been seen with the TFMT stems. This is most likely due to the lower modulus of elasticity titanium used for these stems. The success of the TFMT stem in cases with severe femoral bone deficiency is most encouraging as the same general technique and systems can now be applied across a range of cases.

QOL scores suggest a difference in outcome in favour of TFMT stems at two years. There were potential confounders as there was less than 100% data capture for both pre- and post-operative QOL scores and more of the TFMT group had severe femoral defects. Theoretically these should be negative cofounders for the TFMT group, therefore we feel that the conclusions are justifiable.

Fracture of the stem was concerning although this has also been observed with CNCC stems. We have adopted the more recently introduced ZMR-XL system with a strengthened modular junction, but only use this in large femurs as it can result in removal of excessive bone in smaller patients.

Continuing follow-up will demonstrate the long-term outcome for TFMT stems and also their efficacy in the management of periprosthetic fractures. We believe that these systems are a very attractive option in the management of femoral revision regardless of diagnosis and bone deficiency.

The author or one or more of the authors have received or will receive benefits for personal or professional use from a commercial party related directly or indirectly to the subject of this article. In addition, benefits have been or will be directed to a research fund, foundation, educational institution, or other non- profit organization with which one or more of the authors are associated.

This paper is based on a study which was presented at the Winter 2011 Current Concepts in Joint Replacement meeting in Orlando, Florida, 7th – 10th December.