A Radiographic Evaluation of Short Monolithic Femoral Hip Stem (SMF) for Dysplastic Osteoarthritis: Does Stem Alignment Influence on the Stability? ()
1. Introduction
In Japan, which has the highest longevity in the world, the diagnosis that most often requires total hip arthroplasty (THA) is hip dysplasia [1] . Dysplastic patients are relatively young and have high activity [1] [2] . For such young and active patients, femoral bone preservation is an essential issue, considering the revision in the future [3] [4] . Recently many types of uncemented short stems have been designed to preserve the bone stock and at the same time to ease the insertion procedure in MIS-THA [4] [5] . Each type has different concept to obtain the initial and long-term stability in the femur [4] [5] . Short Monolithic Femoral hip stem (SMFTM) (Smith & Nephew, Inc., Memphis, TN, USA) (Figure 1) is a short stem and was improved as a modification of SYNERGY Stem (Smith & Nephew, Inc.) [6] . SMFTM is a proximal fixation type stem and has double tapered shape in the medial-lateral and anterior-posterior planes to provide good proximal fit and fixation [7] . Because of the shape and structural property, the recommended alignment of SMFTM is varus at least for primary osteoarthritis [7] . Dysplastic hips often have dysplastic femurs [8] . Therefore, a question comes if the recommended alignment above is also true in the dysplastic hips. We evaluated the influence of the SMFTM stem alignment on the stability in the dysplastic hips.
2. Methodology
We evaluated 29 hips in 28 dysplastic patients (5 males and 24 females) after THA using SMFTM stems through 2 or more years follow-up. The average age at the surgery was 60 years old (39 to 80). The follow-up period was 2 to 6 years (3.5 years in average). The body weight ranged 41.1 to 91.0 kg and was 58.8 kg in average at the surgery. The average body mass index (BMI) was 23.2 (17.2 - 31.9). Those who belonged to Crowe’s classification I were 19 and those of II
![]()
(A) ![]()
(B)
Figure 1. Short monolithic femoral hip stem (SMF). (A) anterior-posterior view; (B) superior-inferior view.
were 10 hips [9] . The shape of the femur was classified as Dorr’s Type A in 5, B in 21, and C in 3 hips [10] . All patients started full weight bearing gait from within the day of surgery or next day.
Stem alignment was determined by Kramhøft’s classification and mode of stem fixation by Nakata’s and Luger’s [11] [12] [13] . We evaluated the stem subsidence, stress shielding, radiolucent lines, spotwelds (cancellous condensation), stress shielding, and cortical hypertrophy on the consecutive radiographs [14] [15] [16] . The locations of the radiographic findings are described along Gruen’s zone definition [17] . All the THA were performed through an anterolateral MIS procedure [1] . Neither navigation system nor image intensifier was used. Combined cups were R3 Cup (Smith & Nephew, Inc.) in all hips [1] . The length of SMFTM is about 20% shorter than that of other primary stems to preserve the bone stock [18] . SMFTM stem has a circumferential STIKTITE porous coating (three-dimensional porous structure made of sintered titanium powder, 60% of porous rate) to obtain early bone in-growth in the most proximal part, circumferential grid blasting for bone on-growth in the middle, and satin finish in the distal part (Figure 1) [18] [19] . All stems in this study are mono-block. Preoperative plannings were performed manually on two dimensional images in all patients.
Although the number of subjects was small, statistical analyses were carried out using unpaired Student’s t-test, the Chi square test, Pearson’s correlation coefficient test, and Fisher’s Z transformation. Significant differences were reported at p < 0.05 in all statistical analyses.
3. Results
The alignment of the stem was varus (on anteroposterior view)-flexion (on lateral view) in 15 (57.7%) (Figure 2), neutral-flexion in 7 (24.1%), varus-neutral in 6 (20.7%), and neutral-neutral in 1 hip (3.4%) (Figure 3). Crowe’s Grade did not have influence on the stem alignment (Table 1). The ratio of non-varus alignment (i.e., neutral in anterolateral view) was more with Dorr’s Type C than with others (t = 4.209, p = 0.122) (Table 1). The mode of fixation was mediolateral fit in 23 (79.3%) and multi-point in 6 hips (20.7%). Achieving rate of mediolateral fixation was significantly higher in the varus stems (all hips) than in non-varus ones (2 hips) (t = 14.312, p = 0.000). The significant subsidence (Table 2) occurred in 3 hips (10.3%) although all stems became stable within 6 months. The average depth of subsidence was 0.8 (0 - 4.5) mm. Significant correlations were not observed between the body weight and the depth (r = 0.136, p = 0.475), nor between BMI and the depth (r = 0.222, p = 0.246). The depth was significantly lower in the varus stems than in non-varus ones (t = 1.768, p = 0.044) (Table 2). The average depth was 0.7 in mediolateral fixation and 1.5 mm in the multi point fixation (t = 1.437, p = 0.081). Stress shielding was observed in 23 (79.3%) hips. More prevalence of stress shielding was observed in non-varus alignment (t = 10.07, p = 0.007) (Table 3). Radiographic reactions such as spotwelds or cortical hypertrophy were mainly observed at the portion where the stem contacted
![]()
Table 1. Stem alignment [11] vs. femoral shape.
*: t = 0.420, p = 0.517 †: 2.971, p = 0.226.
![]()
Table 2. Stem alignment vs. subsidence [16] .
‡: t = 1.065, p = 0.044.
to the lateral cortex (Table 4-6). All radiological findings brought about no negative clinical symptoms. Neither intraoperative fracture nor postoperative dislocation occurred. No revision was necessary for all patients.
4. Discussion
As the limitation of this study, number of the patients are small and follow-up
![]()
Table 3. Stem alignment vs. stress shielding [14] .
• : t = 10.07, p = 0.007.
![]()
Table 4. Stem alignment vs. cortical hypertrophy [15] .
![]()
Table 5. Stem alignment vs. radiolucent line [15] .
![]()
Table 6. Stem alignment vs. Spotwelds [15] .
term provides only for short term evaluation, while mid- or long-term result of this stem is relatively few in the database, this study can be evaluated as the report for dysplastic hips [6] [7] [20] . SMFTM stem is a proximal fixation type as its proximal surface structure to aim at bone affinity indicates [6] [7] . As shown in our result, relatively high prevalence of distal radiolucent line was observed. This line seemed to be brought about by the successful proximal fixation [6] [7] . At the same time, as far as the alignment was varus, good proximal fixation and proper stress distribution seemed to be provided in SMFTM stem, because the prevalence of stress shielding was significantly lower in varus than in non-varus alignment.
Because of recent spread of MIS procedure, the usage of short stems has been increasing in THA [4] . Short stems seem to allow preservation of bone stock, with decreased stress shielding and also a lower incidence of thigh pain, compared with conventional stems [21] . On the other hand, a concern in short stems exists, such as more frequent risk of subsidence or radiologic reactions than those of conventional uncemented stems [22] . Subsidence of uncemented stems can be generally accepted within the first three months, but after that osseointegration and stability should have occurred [23] . In our case, 2 out of 29 hips showed 3 mm or more subsidence though they became stable and no aggravation occurred after that. In both of these two hips, the stems were multi-point fit as the result of undersizing of the stem. On the other hand, some other stems were inserted in non-varus alignment to avoid the “undersizing”. We could not judge which was the better, “undersizing” or “non-varus alignment” in this study, though both of them should be avoided of course, if possible.
Dysplastic hips are often accompanied by coxa valgus as the result of sub-dislocation [24] [25] . We concerned offset when varus insertion. In addition, when the stem is inserted deeper than expected, a longer head (for mono-block stems) can be generally adapted to compensate the leg length. However, such compensation emphasizes the excessive offset with varus insertion (Figure 4). Fortunately, we had no problem of the offset in this study.
This type of stem should be inserted along the femoral neck axis [4] . Thus, reserving the calcar ring (as the guide of insertion) enables the varus insertion and the dispersion of excessive valgus load. When the neck preservation is insufficient, the stem will be inserted into neutral alignment along the metaphyseal cortex and the lateral cortex cannot resist the valgus load. In dysplastic hips, shortening and/or and hypoplasia of the femoral neck is often observed [8] [24] . Such cervical bony shape may have contributed to our result. In addition, preserving enough neck length often disturbs the surgical procedure in the shallow, narrow, and small bony acetabulum against the contracture and/or shortening of the leg length due to sub-dislocation in dysplastic hips [8] [24] . The inferior portion (and also posterior one in anterior approaches) of the acetabulum is often
![]()
(A)![]()
(b)
Figure 4. Excessive offset (80-year-old female). (A) Before the surgery the femur showed coxa valga. (B) The stem was inserted in a varus alignment and also into the canal deeper the expected. An extra-long head was adapted to compensate the leg length and resulted in much more offset than the contralateral hip while the patient did not complain of the rotator muscles irritation or the internal rotation disturbance due to the excessive offset (6 years after THA).
hidden by the medial part of the neck when the osteotomy is performed just below the capital. This may disturb the procedure to set the cup in the anatomical (i.e., lower and more medial than the sub-dislocated alignment) in dysplastic acetabulum [8] . In dysplastic hips, we often encounter the excessive anterior-torsion of the neck (Figure 3(B)) [24] . This deformity, when the enough neck length is preserved, increases the difficulty in accessing to the acetabulum and also complicates the insertion of rasps or stem into the twisted femoral canal (although SMFTM has the configurational advantage of more easiness to slip through the twisted neck until the proper fit than other short stems) [6] [7] . Thus, we are often in a dilemma to decide the priority between “enough resection of the neck to secure the access to the acetabulum” and “preserving the neck”, especially in MIS-THA for dysplastic hips [24] . Recent trend in MIS-THA is to preserve maximally the ligaments around the hip and the preserved remaining soft tissue tension makes the dilemma more serious [26] . On the other hand, it was also true that even stems inserted in non-varus alignment or even stems showed early subsidence finally became stable and required no revision. In other words, SMF is a stem that has a potential to accept the non-varus insertion, at least in dysplastic hips.
As an additional concern, the surface finish and the anteroposterior bulkiness of the stem may give more difficulty at the time of stem removal and more bone damage in the proximal femur would be considered at the future revision THA with SMFTM than with other straight even-surface stems (e.g., SL PLUSTM stem), while the preservation of proximal cortex in diaphysis is much attractive nonetheless [7] [27] .
5. Conclusion
The varus insertion of the stem seemed more secure also in dysplastic femurs, but even non-varus was acceptable as it brought about no severe problem. Comprehensively evaluating the result, careful selection of the patients is essential to take the advantage of and to overcome the disadvantage of SMFTM for dysplastic hips.
Acknowledgements
The authors respectfully dedicate this work to our late colleague, Prof. Yoshikatsu Kuroki.