Cementless femoral stems have been widely used in performance of hip arthroplasty in recent years. Short-curved stems and standard-length single wedged stems have frequently been used in performance of cementless total hip arthroplasty (THA). Many of these types of stems that engage with the metaphysis of the femur have recently been designed. This type of stem can ensure minimal loss of metaphyseal bone during surgery and preserve bone stock in the long-term by providing greater physiological load transfer of the proximal femur, stimulating the peri-implant bone remodeling process and enhancement of osseointegration^1,2). The design purpose of the short-curved stem has been achieved in terms of proximal fixation and load transfer and its use has resulted in better preservation of the proximal femur bone stock^3,4). In addition, it is believed that use of short-curved stems may facilitate possible revision surgery in the future, although no study supporting this assumption has been reported in the literature so far.

The Fitmore^® hip stem (Zimmer) (Fig. 1A), a curved, uncemented, short curved stem prosthesis, was introduced into routine clinical practice in 2007⁵⁾. Its curved design is thought to enable proximal load transmission and thus provide an optimal fit in the calcar region. It is presumed that its primary stability is achieved by press-fit fixation based on the triple-tapered design.

Figure 1. (A) The Fitmore^® femoral stem (Zimmer). (B) The M/L taper^® femoral stem (Zimmer).

In a review of 500 cases with Fitmore^® hip stems, with a mean follow-up period of 1.3 years, Gustke⁶⁾ reported a survival rate of 99.4% with femoral revision for any reason as the endpoint and detected cortical hypertrophy in approximately 29% of his first 100 cases. He also reported a stem subsidence of ≥2 mm in 34% of the first 100 cases reviewed.

Because excellent mid-term or long-term results have been achieved with use of standard-length single wedged stems^7-12), critical evaluation of newly introduced short-stem designs for their clinical and radiographic outcomes is needed. Since 2006, uncemented standard-length single wedged stems (M/L taper^®; Zimmer) (Fig. 1B) have been used by our orthopedic department and satisfactory results have been achieved with use of this stem^13,14). Having experienced many cases, we assumed that enhancement of primary stability and satisfactory results could be achieved with application of the concept of a shorter stem design with greater functionality and biomechanical suitability compared to the previous stem, therefore, we began using a short-curved stem from 2012. The purpose of this study was to compare radiologic and clinical outcomes of use of short-curved (Fitmore^®) stems and standard-length single wedged stems (M/L taper^®).

The study was approved by the Institutional Review Board (IRB) of Chonnam National University Hwasun Hospital (IRB No. CNUHH-2020-241) and the written informed consent was waived by the IRB due to the retrospective nature of the study. This study had a retrospective case-control design and collection of data on primary THAs performed using the Fitmore^® femoral stem (140 hips, 134 patients) and the M/L taper^® femoral stem (206 hips, 198 patients) between October 2012 and June 2014 at Chonnam National University Hwasun Hospital was performed. Among these patients, ten patients (10 hips) who had undergone THA died from unrelated causes during a follow-up period of less than five years, and 14 patients (14 hips) were lost to follow-up. Finally, 127 hips (122 patients) with the Fitmore^® femoral stem and 195 hips (187 patients) with the M/L taper^® femoral stem were included for analysis (Table 1).

Table 1 . Demographics of the Patients.

Parameter	Fitmore® stem	M/L taper® stem	P-value
No. of hips	127	195
Sex, male/female	62/65	83/112	0.270
Age (yr)	55.1 (20.2-88.8)	54.0 (15.7-86.7)	0.510
Body mass index (kg/m2) Follow-up duration (yr)	21.5 (15.7-30.5) 5.8 (5.2-6.9)	22.0 (14.8-29.8) 6.0 (5.2-7.1)	0.743 0.422
Causes of THA			0.064
Osteonecrosis of femoral head	50 (39.4)	79 (40.5)
Osteoarthritis	46 (36.2)	69 (35.4)
Femoral neck fracture	16 (12.6)	7 (3.6)
Fracture non-union	2 (1.6)	1 (0.5)
Septic hip sequelae	4 (3.1)	10 (5.1)
Developmental dysplasia of hips	1 (0.8)	6 (3.1)
LCPD sequelae	4 (3.1)	11 (5.6)
Rheumatoid arthritis	2 (1.6)	4 (2.1)
Tumor	1 (0.8)	2 (1.0)
AS	0 (0)	4 (2.1)
Rapid destructive coxarthrosis	0 (0)	1 (0.5)
Resurfacing arthroplasty aseptic loosening	0 (0)	1 (0.5)
Others	1 (0.8)	0 (0)
Dorr classification			0.022
Type A	43 (33.9)	47 (24.1)
Type B	68 (53.5)	138 (70.8)
Type C	16 (12.6)	10 (5.1)

Values are presented as number only, mean (range), or number (%)..

THA: total hip arthroplasty, LCPD: Legg-Calve-Perthes disease, AS: ankylosing spondylitis..

Minimally invasive THA was performed using the two-incision technique¹⁴⁾. Surgery was performed by two experienced hip surgeons in the lateral position using a modified two-incision technique. A cementless acetabular shell with a porous coating (Delta PF^®; Lima) was used in all cases. For the bearing surface, a ceramic liner and head (BIOLOX delta; CeramTec GmbH) was used in all cases.

Postoperatively, patients used an abduction pillow between the legs for one month and a pneumatic pump during admission periods. Quadriceps strengthening exercises were started on the day of surgery, and sitting alongside the bed was also allowed on the day of surgery, which was determined by patient comfort. Depending on the general condition, mobilization with weight-bearing was recommended, as tolerated, on postoperative day 1. Patients were discharged home when they were capable of using walking assists properly.

Patients were followed at six weeks, three months, six months, and one year, and then annually. Harris hip scores (HHS)¹⁵⁾ and Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) scores¹⁶⁾ were recorded during each visit. Patients were asked specific questions about episodes of instability or dislocation, thigh pain, and squeaking. Information on hospital stay, skin incision, operation time, blood loss, hemoglobin change, transfusion amount, the day of ambulation, and the duration of hospital stay was also obtained, and any intraoperative and postoperative complications were recorded.

Assessment of anteroposterior (AP) radiographs of both hips and the lateral radiograph was performed during each visit. All radiographic measurements were performed preoperatively by two independent observers, immediately after surgery, and during the last follow-up visit. Radiographic evaluations included preoperative Dorr classification¹⁷⁾ for evaluation of the femoral geometry, femoral component alignment within the femoral canal, and femoral component subsidence. Femoral component vertical subsidence was defined as change in the distance from the landmark of the femoral component to the most proximal point of the greater trochanter. A difference of 4 mm or more in the vertical direction between immediate postoperative measurements and measurements at the last follow-up was regarded as an indication of vertical subsidence. AP and frog leg lateral views were also examined for the presence of radiolucent lines or osteolysis at the femoral bone-prosthesis interface using the seven zones described by Gruen et al.¹⁸⁾. For evaluation of cortical hypertrophy, the locations were divided into medial, lateral, and both. Grading of heterotopic bone formation was performed using Brooker’s classification system¹⁹⁾. At each follow-up, evaluations were also performed for identification of any complications.

Statistical analysis was performed using IBM SPSS Statistical software (ver. 25; IBM Corp.), and P<0.05 was considered statistically significant. Comparison of the HHS, WOMAC score, and radiographic data in the two groups at the final follow-up was performed using an independent samples t-test for continuous variables. Chi-square (Fisher’s exact) test was used for categorical variables to determine differences in the incidence of complications and sex between the two groups. Binary logistic regression was performed for identification of significant factors associated with development of cortical hypertrophy Assessment of survival was performed using the Kaplan–Meier analysis method. A case of implant failure was defined as stem revision performed for any reason such as infection, aseptic loosening, and periprosthetic fracture as the endpoint.

The mean age of patients in the Fitmore^® group and the M/L taper^® group was 55.1 years (range, 20.2-88.8 years) and 54.0 years (range, 15.7-86.7 years) at the time of surgery over a mean follow-up period of 5.8 years (range, 5.2-6.9 years) and 6.0 years (range, 5.2-7.1 years), respectively (Table 1). There was no significant difference in demographic characteristics; however, a significant difference in preoperative Dorr classification of the femur was observed between the two groups (P=0.022). The Dorr classifications in the Fitmore^® group were as follows: 33.9%, A; 53.5%, B; and 12.6%, C; those in the M/L taper^® group were as follows: 24.1%, A; 70.8%, B; and 5.1%, C (Table 1). In the Fitmore^® group, standard offset (137°) was used in 104 hips and extended offset (129°) was used in 23 hips. The extended offset stem was used in the beginning; however, use of the standard stem was adopted later because femoral offset and leg length can be adjusted by the surgeon according to changing femoral head size or changing stem size. In the M/L taper^® group, the standard offset (137°) was used in all hips (Fig. 2).

Figure 2. (A) Preoperative radiograph of a 57-year-old male with osteonecrosis of the left femoral head and Dorr type A femoral bone. (B) Immediate postoperative radiograph of the left hip with a Fitmore^® stem. (C) Anteroposterior X-ray of the left hip at one year after total hip arthroplasty shows a radiolucent line in Gruen zone I. (D-F) Anteroposterior X-rays of the left hip at two years, five years, and six years after total hip arthroplasty show cortical hypertrophy on the medial and lateral sides of the femur (red arrows).

1. Clinical Results

No significant difference in perioperative parameters was observed between the two groups (Table 2).

Table 2 . Comparison of the Perioperative Parameters.

Parameter	Fitmore® stem	M/L taper® stem	P-value
Duration of operation (min)	62.8 (40-105)	65.4 (35-110)	0.403
Mean estimated blood loss (mL)	578.9 (430-1,450)	613.2 (450-1,510)	0.129
Hb change (g/dL)	2.6 (1.2-4.6)	2.4 (1.1-4.2)	0.628
Preoperative Hb (g/dL)	11.6 (9.2-15.2)	11.4 (9.1-14.9)	0.453
Postoperative Hb (g/dL)	9.2 (7.1-13.3)	9.0 (6.9-12.8)	0.571
Transfusion (No. of units)	0.8 (0-3)	0.9 (0-3)	0.442
Ambulation by walker or crutch	2.3 (1-7)	2.1 (1-6)	0.673

Values are presented as mean (range)..

Hb: hemoglobin..

In the Fitmore^® group and the M/L taper^® group, respectively, significant improvement of the mean HHS score from 52.4 (range, 32-68) and 48.9 (range, 25-70) preoperatively to 93.3 (range, 82-100) and 94.5 (range, 81-100) was observed during the final follow-up. The mean WOMAC scores also showed significant improvement from 73.3 (range, 43-92) and 76.8 (range, 39-98) preoperatively to 22.9 (range, 13-34) and 25.6 (range, 13-36) at the final follow-up, respectively. However, no significant difference in HHS and WOMAC scores was observed between the two groups (P>0.05) (Table 3).

Table 3 . The Difference of Functional Status in Two Groups at Last Follow-up.

Outcome	Fitmore® stem			M/L taper® stem		P-value
	Preop.	Last follow-up		Preop.	Last follow-up
HHS	52.4 (32-68)	93.3 (82-100)		48.9 (25-70)	94.5 (81-100)	0.980
WOMAC	73.3 (43-92)	22.9 (13-34)		76.8 (39-98)	25.6 (13-36)	0.465

Values are presented as mean (range)..

P-value was calculated between the last follow-up clinical score of two groups..

Preop.: preoperative, HHS: Harris hip score, WOMAC: Western Ontario and McMaster Universities Osteoarthritis Index..

2. Radiographical Results

No significant difference in the stem position was observed between the two groups. Valgus greater than 3° was observed in 11 hips (8.7%) in the Fitmore^® group and four hips (2.1%) in the M/L taper^® group. Stems with varus greater than 3° were observed in three (2.4%) hips in the Fitmore^® group and three (1.5%) hips in the M/L taper^® group. A total of 113 hips (89.0%) and 189 hips (96.9%) were in a neutral position (–3 to +3) in each group.

Stem subsidence >4 mm was observed in six hips (4.7%) in the Fitmore^® group; however, two hips (1.0%) showed >4 mm subsidence in the M/L taper^® group (P=0.037).

No significant difference in the radiolucent line, observed on the last follow-up X-ray, was observed between the two groups (P=0.125). In the Fitmore^® group, on the femoral side, the radiolucent line was observed in zone I in 20 hips (15.7%), in zone II in one hip (0.8%), and in zone VII in three hips (2.4%). In the M/L taper^® group, on the femoral side, the radiolucent line was observed in zone I in seven hips (3.6%), in zone II in two hips (1.0%), and in zone VII in three hips (1.5%). No periacetabular osteolysis or radiolucent lines were observed in either group (Table 4).

Table 4 . Radiographical Follow-up Results.

Parameter	Fitmore® stem (n=127)	M/L taper® stem (n=195)	P-value
Femoral stem subsidence			0.005
<2 mm	108	186
2-4 mm	13	7
>4 mm	6	2
Femoral side radiolucent line			0.125
I	20	7
II	1	2
III	0	0
IV	0	0
V	0	0
VI	0	0
VII	3	3
Cortical hypertrophy	28	1	<0.001
Medial side	12	1
Lateral side	11	0
Medial and lateral side	5	0

Values are presented as number only..

In the Fitmore^® group, cortical hypertrophy was detected on the final follow-up AP radiographs in 28 hips (Fig. 3). In the M/L taper^® group, cortical hypertrophy was only detected in one hip (Table 4). The number of cases with cortical hypertrophy was significantly greater in the Fitmore^® group than in the M/L taper^® group (P<0.001): 12 hips with medial cortical hypertrophy, 11 hips with lateral cortical hypertrophy, and five hips with medial and lateral hypertrophy. However, the M/L taper^® stem group included only one hip with medial hypertrophy. The regression analysis for assessment of cortical hypertrophy in the Fitmore^® group found no significant factors; intraoperative femoral crack (P=0.232), gender (P=0.156), body mass index (P=0.552), Dorr type A (P=0.117), and Dorr type B (P=0.201), and Dorr type C (P=0.438).

Figure 3. (A) Preoperative radiograph of a 58-year-old male with osteonecrosis of the left femoral head. (B) Immediate postoperative radiograph of the left hip with an M/L taper^® stem. (C-F) Anteroposterior X-rays of the left hip at three months, six months, three years, and six years after total hip arthroplasty show no radiolucent line or cortical hypertrophy around the femoral stem.

3. Complications and Revision

Development of an intraoperative calcar crack occurred in 15 hips (Fitmore^® group: 14 hips; M/L taper^® group: 1 hip, P<0.001), which were treated with cable wiring during surgery. Two intraoperative fractures of the femoral shaft were treated with wiring in the M/L taper^® group and there were none in the Fitmore^® group (P=0.252). A complaint of squeaking was reported for three hips in the Fitmore^® group and seven hips in the M/L taper^® group (P=0.745), and a complaint of thigh pain was reported for two hips in the Fitmore^® group and 11 hips in the M/L taper^® group (P=0.086).

Arthroscopic decompression and debridement of the hip was required at 18 months postoperatively for one patient in the Fitmore^® group who developed hemorrhagic iliopsoas bursitis. Except for this patient, no other cases required reoperation. The M/L taper^® group included one hip with bone growth failure and subsidence, which required stem revision. There was no case involving dislocation or infection in either group (Table 5).

Table 5 . The Difference of Perioperative Complications between Two Groups.

Complication	Fitmore® stem (n=127)	M/L taper® stem (n=195)	P-value
Intraoperative crack	14	1	<0.001
Intraoperative femur shaft facture	0	2	0.252
Squeaking	3	7	0.745
Iliopsoas bursitis	1	0	0.394
Stem loosening	0	1	>0.999
Dislocation	0	0	-
Infection	0	0	-
Thigh pain	2	11	0.086
Nerve palsy	0	0	-

Values are presented as number only..

4. Implant Survivorship

The survival rates with an endpoint of revision for any reason at five years were 100.0% in the Fitmore^® group, and 99.5% in the M/L taper^® group (95% confidence interval, 97.2%-100%). No significant difference in survivorship was observed between groups (log-rank, P=0.418; Fig. 4).

Figure 4. Kaplan–Meier survival curves for survivorship with an end point of any revision.

Currently, the popularity of the cementless femoral stem has increased in many countries due to its acceptable long-term survival rate. Compared with the cemented stem, the improved surface treatment is the main factor in their popularity, along with the more uncomplicated surgical technique. In addition, the cementless femoral stem is more suitable for minimally invasive surgeries such as the direct anterior approach.

In our study, the Fitmore^® group included a significantly greater number of cases with cortical hypertrophy compared with the M/L taper^® group. However, no significant difference in HHS and WOMAC scores was observed between the two groups. According to Thalmann et al.²⁰⁾, cortical hypertrophy was detected on the distal side in 46% of cases, mainly in Gruen zones II, III, V, and VI. In his 5-year follow-up study, no correlation was observed between cortical hypertrophy and clinical outcomes, such as surgical or implant factors. However, stress shielding with cortical thinning was more commonly observed in Gruen zone VII. In a study of patients with cortical hypertrophy, Cho et al.²¹⁾ reported a higher incidence of thigh pain in 27 hips (6.2%); however, cortical hypertrophy had no effect on clinical outcomes or femoral stem stability. In the current study, cortical hypertrophy was detected in 28 of 127 hips (22.0%). Consistent with previous studies^5,22), our findings also demonstrated that cortical hypertrophy had no significant effect on clinical outcomes during mid-term follow-up. Even though 11 of 28 hips that showed cortical hypertrophy were observed as a radiolucent line around the stem, predominantly in Gruen zone I, cortical hypertrophy and the radiolucent line were found to have a negative impact on mid-term survival. Evaluation of hips with cortical hypertrophy over a longer period of follow-up is needed for assessment of the potential for failure of the femoral stem.

Thigh pain following primary THA is a complication related to stem design and implant stiffness. For the short curved cementless stem, there is an association of thigh pain with high levels of activity and Dorr type C femoral bone shape²³⁾. Amendola et al.²⁴⁾ reported a substantially higher rate for 25% of patients with a short stem who complained of persistent thigh pain during short-term follow-up. In that study, although the thigh pain was concerning, most patients reported an excellent return to function, and bone ingrowth with minimal stress shielding was achieved for most stems. A series of 123 THAs with a short stem reported thigh pain for 16% of hips and distal cortical hypertrophy for 75% of hips; however, no association was observed between cortical hypertrophy and thigh pain²⁰⁾. In the current study, lower rates of thigh pain were observed in the Fitmore^® stem group compared with the M/L taper^® stem group. In addition, no significant correlation was observed between cortical hypertrophy and thigh pain.

In the current series, the survival rate of femoral stems, with revision for any reason as the endpoint, was excellent at 100% in the Fitmore^® stem group. In a study of 123 THAs using the Fitmore^® hip stem during a mean follow-up period of five years, Thalmann et al.²⁰⁾ reported a survival rate of 99.2% for the femoral stem with revision for any reason as the endpoint, similar to our results. Femoral stem revision was required for only one hip due to bone ingrowth failure of the stem. In the current series, there was no dislocation or infection in the Fitmore^® stem group during the 5.9-year follow-up period. However, a larger number of cases developed a proximal femur crack (14 hips, 11.0%) in the Fitmore^® stem group than in the M/L taper^® group. In patients with a Fitmore^® stem, we believe that use of cerclage wiring during insertion of the femoral stems would have been helpful in management of this troublesome issue and the effort to obtain satisfactory results.

In this regard, the short stem was developed to support further preservation of femoral bone stock and easier insertion during performance of minimally invasive surgery. The Fitmore^® stem was also developed with these two objectives in mind, and the reported short-term results were as good as those for any other cementless stem. However, only a few medium- or long-term outcomes have been reported for short femoral stems. The Fitmore^® stem has shown in vitro primary stability comparable to that for the CLS (Zimmer) and Mayo prostheses (Zimmer). The early radiolucent line appeared postoperatively at six months and did not progress. We suggest that radiolucent lines may be a result of initial failure of bone ingrowth of the proximal part of the femur, particularly in Gruen zones I and VII and are a consequence of technical problems. The Fitmore^® stem was developed to provide greater convenience when using a direct anterior approach in performance of THA, where the fashion of broaching and stem insertion should be curved from lateral to medial of the metaphyseal area of the femur. Using the two-incision technique with a small posterior incision, we were not able to see the direction of the stem during the procedure, thus following the technical guidelines recommended by the manufacturer was more difficult.

This study has several limitations. First, this study did not have a randomized prospective design. The acetabular component was obtained from a different manufacturer. Some surgeons may question the usage of implants from different manufacturers in performance of one THA; however, based on our experience, we used this combination because this implant is more suitable for achievement of acetabular fixation than cups from the same manufacturer as that of the femoral stem. Second, a statistically significant difference in the Dorr type of the femur was observed between the two groups. This disparity was attributed to the morphology of the femoral stem. We favored the use of the short-curved stem in Dorr type B femurs over type A because the diameter of the short-curved stem is relatively larger compared to the single wedged stem, facilitating a better fit during preoperative templating. This difference is regarded as a limitation of this retrospective study. Third, only a limited radiological evaluation of cortical hypertrophy was performed using X-rays, which affected the results. In addition, the results of our study were restricted to mid-term results showing cortical hypertrophy. Conduct of a long-term clinical and radiographic study is needed for determination of other differences.

Similarly favorable clinical and radiographic outcomes have been achieved with use of short-curved stems and standard-length single wedged stems; however, a higher rate of cortical hypertrophy and femoral crack and a lower rate of thigh pain were observed with use of Fitmore^® femoral stems. In addition, no significant difference in survival rates was observed between the two groups over a minimum follow-up period of five years. Caution should be used for prevention of crack or fracture of the calcar during intraoperative insertion of the Fitmore^® femoral stem. Additional clinical and radiologic follow-up will be necessary for evaluation of any differences in long-term outcomes of use of the two cementless femoral stems.

This research was supported by the best annual research grant of the Korean Hip Society 2022.

Kyung-Soon Park has been an editorial board member since January 2024, but had no role in the decision to publish this article. No other potential conflict of interest relevant to this article was reported.