The annual incidence of hip arthroplasty has shown a significant increase¹⁾. The results of analysis of temporal patterns indicate a projected 40% increase in the rates of total hip arthroplasty (THA) by the year 2030. It is expected that the increase will be more pronounced in men, with a projected increase of 51%, compared to women, with an anticipated increase of 33%^2,3).

Leg length discrepancy (LLD) in the lower limbs is extremely common, occurring up to 30% of the time following hip replacement^1,2,4). According to some reports, the average mean LLD might vary from 3 to 17 mm^5-7). LLD following hip arthroplasty is a significant factor in patient disappointment and ranks as the second most prevalent cause for litigation associated with hip arthroplasty⁴⁾. LLD has been associated with certain adverse effects, including back problems, sciatica, peroneal palsy, gait abnormalities, implant dislocation, and aseptic loosening. These complications, which are considered a factor in increased morbidity rates, often require revision surgery^8-13). Several studies have reported an association of more than 10 mm LLD with low back pain^14,15). A study conducted by Konyves and Bannister¹⁶⁾ reported that patients with a greater leg length on the side who had undergone surgery had a more unfavorable functional outcome.

Methods that have been used successfully to minimize LLD following hip arthroplasty include a navigation system, bone-fixed Kirschner wires, using a lesser trochanter (LT) as an anatomical reference, and a templating approach^17-20). All currently available techniques are invasive, costly, associated with peri-pinhole fractures, and challenging to implement. Following hip arthroplasty, we introduced a method involving the use of the greater trochanter (GT) tip as an anatomical reference. Thus, the primary objective of this study is to report on a technique that uses the GT tip as an anatomical landmark to minimize LLD and examine its efficacy in clinical application.

1. Ethical Statement

The study was conducted according to the principles of the Declaration of Helsinki (2013). Patients were required to submit written informed consent prior to enrollment in this study. This study protocol was approved by the Ethics Committee of Faculty of Medicine, Pelita Harapan University (approval No. 193/K-LKJ/ETIK/XI/2023).

2. Patient Selection

Patients who underwent hip arthroplasty between January 2019 and August 2023 at Siloam Hospital Lippo Village were enrolled in this study. Baseline characteristic data including age, sex, diagnosis, body mass index (BMI), and type of arthroplasty were recorded. Inclusion criteria included (1) patients who underwent hip arthroplasty from 2019 to 2023, (2) available preoperative pelvic X-ray anteroposterior (AP) with symphysis pubis centration and 10°-15° of internal rotation, and (3) available postoperative erect pelvic X-ray AP with the criteria described above. Exclusion criteria included (1) comminuted GT, (2) comminuted LT, (3) poor pelvic X-ray position, (4) patients with bone tumours or hip dysplasia, and (5) bilateral pathologies. Intraoperative methods using GT tips as an anatomical reference were defined as GT groups. Patients using LT as an anatomical reference during hip arthroplasty were defined as the control group. Patients who underwent hip arthroplasty from 2019-2021 were enrolled in the control group and the GT group from 2022-2023. The flow for patient selection is shown in Fig. 1.

Figure 1. Patient selection method. GT: greater trochanter, LT: lesser trochanter.

3. Measurement Method

1) Greater trochanter group

Radiographic measurements were performed using image archiving and communications system software (INFINITT Healthcare Co.). Preoperatively, an AP pelvic X-ray was obtained with the patient in a supine position, and the midsagittal plane was adjusted to the center of the grid for symphysis pubis centration. Lower extremities and feet were rotated medially 10° to 15° unless contraindicated. On the healthy side, the midline of the proximal femoral shaft was drawn to obtain the femoral anatomical axis. An additional line was drawn perpendicular to the femoral axis, crossing the center of the acetabulum. The distance between the line and the tip of the GT was then measured [Lg]. The acetabular diameter was measured [D] by drawing a midline between the femoral head and the acetabulum for assessment of radiolucent cartilage. The ratio of [Lg] to [D] was calculated as [R], as shown in Fig. 2A. Intraoperatively, the largest diameter of the anatomic head was measured using a vernier caliper [D’]. [L’] is the measurement of prosthesis depth insertion, obtained by [R] multiplied by [D’].

Figure 2. (A) Greater trochanter tip method of measurement. (B) Control group (lesser trochanter) measurement. Lg: distance between the femoral head center and the greater trochanter, D: diameter of the femoral head, LI: distance between the femoral head center and the lesser trochanter.

2) Control group

In the control group, the method suggested by Wang et al. was adopted to determine femoral stem depth insertion¹⁹⁾. The midline of the proximal femoral shaft was drawn to obtain the femoral anatomical axis. In addition, the perpendicular line that was drawn was identical to that in the GT group. However, the calculation was obtained by measuring the distance between the center of the femoral head and the most prominent tip of the LT [Ll]. The diameter of the healthy femoral head was measured [D], as shown in Fig. 2B. The ratio [R] of [Ll]/[D] was also calculated.

4. Surgical Technique

All surgical procedures were performed by the senior author, with 18 years of experience using a posterior approach. The patient was positioned in a lateral decubitus position. After draping the affected limb, a skin incision was made just posterior to the GT. The incision was then used to cut through the fascia and tensor fascia latae, which cover the gluteus maximus muscle. Finally, the muscle was carefully divided to access the short external rotators. The sciatic nerve, which passes directly posterior to the short external rotators, is protected due to its anatomical course. After making a short rotator excision, an L-shaped capsule incision was performed, followed by identification of the femoral head.

Following exposure of the hip, a cut was made to the femoral neck. A level of neck resection is located along a line drawn from a point approximately 1 cm in proximity to LT to a point laterally at the base of the neck, and the angle of the cut was matched to a trial or broach. Femoral head extraction was done using an extractor, and the measurement of the femoral head was performed using a vernier caliper [D’]. In hip hemiarthroplasty (HHA) procedure, the chosen size of the femoral head prosthesis should be identical to the size of the diameter of the anatomic femoral head. However, the size of the implant cup was used as a reference for [D’] during performance of the THA procedure. Thus, [D’] was multiplied by [R] to determine the distance from the top of the GT to the implant center. Fig. 3 shows two lines drawn on the broach (lines A and B) based on the [L’] to determine the depth of stem insertion based on a previous calculation using a marking pen. It was used to guide the surgeon in the stem depth insertion. Indeed, the use of this technique relies only on eyesight when adjusting the depth, thus precision and accuracy are required. The prosthesis was then inserted until the A mark was parallel to the tip of the GT.

Figure 3. Intraoperative method of femoral stem insertion.

Several prostheses were used in this study, including Corail^® (DePuy Synthes Co.) (n=11), Summit^® (DePuy Synthes Co.) (n=9), and the UTF and Matrix Stem (United Orthopedic Corp.) (n=8). However, cemented prostheses are utilized in every patient. While a variety of femoral stem neck lengths have been employed to achieve offset balance, the standard size was utilized when performing this procedure due to the single stem size available in our country. Variations in neck length have been associated with disparities during performance of intraoperative measurements, although the difference will not be significant. Use of a downsized stem is recommended when the prosthesis is between sizes, with adjustment of the intraoperative measurement accordingly. During performance of THA, the acetabulum of the affected site must be at the same level as that on the healthy side and the alignment of the prosthesis during insertion must be neutral. Thus, the measurement of L’ is not determined based on the cup position because it is placed in the neutral position of the acetabulum origin. A preoperative X-ray was used to ensure that the acetabular level was equal on both sides. The acetabular insertion proceeded with a 45° inclination and 20° anteversion using a transverse acetabular ligament as a landmark.

5. Leg Length Discrepancy Measurement

Assessment of the vertical lengths parallel to the line connecting the lower margin of the teardrop to the most prominent part of LT on a standing AP pelvic X-ray with symphysis pubis centration to determine LLD after the surgical intervention and 10°-15° of internal rotation taken one month after the procedure²¹⁾ (Fig. 4). LLD was assessed one month post-surgery, as we observed that patients were capable of full weight bearing at that point. This timeframe is considered safer due to the use of cemented prostheses in all patients. Postoperative LLD measurements were performed independently by three independent examiners; disparity among writers was resolved by discussion. Final decisions were rendered by the senior author in cases without agreement. A negative value indicates that the operative side was shorter than the healthy side. A positive value infers that the operative side was longer than the contralateral side.

Figure 4. Leg length discrepancy measurement.

6. Statistical Analysis

Windows IBM SPSS software (ver. 25; IBM Corp.) was used in performance of statistical analysis. Baseline characteristics were compared between the two groups. Assessment of non-parametric numerical and categorical data was performed using the Mann–Whitney U and chi-square tests, respectively (if the expected count <5 more than 20% was continued using the Fisher exact test). Analysis of the interrater reliability during LLD measurement was performed using Cohen’s kappa analysis. A kappa score >0.80 was considered excellent agreement. Assessment of the normality of the data was performed using the Saphiro–Wilk test, and the comparative mean LLD was measured using an independent t-test. P<0.05 was considered statistically significant. Linear regression analysis was performed to examine the association between the estimation of femoral stem depth insertion intraoperatively and subsequent postoperative assessment.

1. Demographic Characteristic

The control group included 11 patients, while the GT group included 17 patients. The mean age of patients in the GT group was 74 years, while the mean age of those in the control group was 73 years. The mean BMI for the GT and control groups is 20.8 kg/m² and 22.2 kg/m², respectively. No statistical difference in terms of age, sex, or BMI was observed between the two groups. However, a significant statistical difference was observed in the type of arthroplasty (P=0.016), as shown in Table 1.

Table 1 . Baseline Characteristics.

Variable	GT group (n=17)	Control group (n=11 )	P-value
Sex, M/F	4/13	1/10	0.619
Age (yr)	74±17	73±10	0.926
BMI (kg/m2)	20.8±5.5	22.2±6.3	0.430
Surgery	THA (n=7) HHA (n=10)	THA (n=0) HHA (n=11)	0.016
OA	3	0	-
AVN	4	0	-
Fracture	Column femur fracture (n=8) Intertrochanteric fracture (n=2)	column femur fracture (n=10) Intertrochanteric fracture (n=1)	-

Values are presented as number or mean±standard deviation..

GT: greater trochanter, M: male, F: female, BMI: body mass index, THA: total hip arthroplasty, HHA: hip hemiarthroplasty, OA: osteoarthritis, AVN: avascular necrosis..

2. Femoral Stem Depth Insertion

As shown in Table 2 and Fig. 5, the results of linear regression analysis showed an association of postoperative measurements with intraoperative planning and yielded significant values (t=2.705, r²=0.672, P=0.016).

Figure 5. (A) Linear regression between variables expressed as a scatter plot. (B) Scatter dot model of leg length discrepancy (LLD) measurement between the two groups. GT: greater trochanter.

Table 2 . Measurement of Femoral Stem Depth Insertion.

Dependent variable	Independent variable	t value	r2 value	P-value
Postoperative measurement	Intraoperative calculation	2.705	0.672*	0.016

t value: β-coefficients..

*Coefficient of determination..

3. Leg Length Discrepancy

The results of LLD in this study demonstrated adherence to the normal distribution and were summed up by measures of central tendency, specifically the mean and median. As shown in Table 3, the LLD range between groups was –4.5 to +4.8 mm in the GT group and –15.3 to +8.7 mm in the control group. Interrater reliability for LLD measurement was excellent between examiners (kappa value=0.855). The median postoperative LLD was +1.4 mm in the GT group and +0.65 mm in the control group. The median postoperative LLD was significantly lower in the GT group compared with the control group (P=0.024). A statistical difference (P=0.004) in the mean postoperative LLD was observed between the GT group and the control group. In addition, a smaller mean LLD was observed in the GT group (+0.97 mm) compared with the control group (–1.44 mm).

Table 3 . Leg Length Discrepancy (LLD) Measurement.

	GT group (n=17)	Control group (n=11)	P-value
LLD (mm)	–0.6	–15.3
	+0.9	–4.7
	+1.9	+8.7
	–1.0	+6.9
	–0.5	+0.6
	+0.9	–2.6
	–0.7	+2.0
	+3.0	–6.0
	+1.4	+1.6
	+1.6	–8.8
	+2.9	+1.5
	–0.3
	+1.9
	+3.3
	–4.5
	+1.7
	+4.8
Median	+1.4	+0.65	0.024
Mean	+0.97	–1.44	0.004

LLD, an important variable, has shown a correlation with favorable results of joint arthroplasty procedures. The current study included a comparison of the mean and median LLD values in both groups. Based on the results shown in Table 3, significantly lower median and mean LLD values were observed in the GT group compared with the control group (P=0.024 and P=0.004, respectively). LLD ranged from –4.5 to +4.8 mm in the GT group and from –15.3 to +8.7 mm in the control group. As shown in Table 2, our method for planning femoral stem depth insertion was significantly associated with postoperative measurement (t=2.705, r²=0.672, P=0.016). Regarding our r² value, it should be noted that the femoral neck stem may be too wide for some of our patients; therefore, the benefit of increasing the depth for adjustment of LLD would be uncertain.

Wang et al.¹⁹⁾ proposed a method using LT as an anatomical reference (reported mean LLD 4.4±3.2 mm in 47 hips); however, it would be challenging to implement due to the round shape of LT. Therefore, determining the tip can be challenging, and is dependent on the quality of the X-ray. Ranawat et al.¹⁷⁾ and McGee and Scott¹⁸⁾, using a bone-fixed K-wire, reported a mean LLD of 5.4 mm and 7.4 mm, respectively, following hip arthroplasty; this method involves the intrusive procedure of drilling bones, which has been associated with the risk of a pin tract fracture. Grosso et al.²²⁾, using the navigation method, obtained promising results with a mean LLD of 1.3 mm in 25 patients. However, it should be noted that the utilization of navigation fixation markers may require supplementary incisions on the skin and has been associated with pin-sit pain and irritation⁵⁾. Despite the availability of pinless navigation systems, they require additional equipment that increases operation time and costs.

The method we recommend employs the use of GT as an anatomical landmark and an AP pelvic X-ray before surgery for measurement of the distance between the center of the femoral head prosthesis and GT. Even though a higher mean LLD was obtained when using our proposed method compared with the navigation system, it is less invasive, lower cost, and time-efficient. To the best of our knowledge, this is the first study to propose and compare between the use of the GT tip and LT tip as an anatomical reference for measuring the depth of femoral stem insertion. We believe that this method is more accessible, lower cost, and more time-efficient. Using the GT tip as an anatomical reference can enable greater tip visibility, resulting in more precise measurements.

However, our study has several limitations, including a limited sample size and the use of a non-randomized design. Due to cartilage-induced interference, accurate measurement of the diameter of the femoral head via X-ray imaging is challenging. The circle used to measure the diameter of the femoral head was positioned to pass through the middle of the space between the uppermost portion of the femoral head and the acetabulum. However, the effect due to the presence of cartilage would be of no significance. In measuring the LLD, the technique employed by Woolson et al. for assessing LLD, which was used in our study, did not account for the potential influence of hip flexion or abduction deformities during the X-ray examination. The presence of these deformities tends to be associated with underestimating the assessed LLD. In addition, the process did not account for other factors unrelated to hip conditions that might contribute to LLD. According to Heaver et al.²³⁾, the interischial line was identified as the most precise pelvic marker for assessing LLD. However, according to Meermans et al.²⁴⁾, the teardrop line indicated a higher level of accuracy. According to the findings reported by Tipton et al.²⁵⁾, measurement of LLD determined using an AP radiograph of the pelvis cannot be equated with the LLD determined from full-length radiographs of the limb. Thus, relying only on a pelvic radiograph is not acceptable for assessing LLD. Nevertheless, our adopted method for measuring LLD is the most adopted method worldwide.

During performance of this procedure, a marking pen was used by the surgeon to determine the precise depth of stem insertion. Use of this technique, which relies solely on visual perception to optimize the depth, requires meticulousness and precision. However, the GT tip is more clearly visible than the LT tip, providing enhanced accuracy even with visual perception alone. This procedure is applicable only to patients with neutral alignment. The position of the head center will change if the stem is not inserted in neutral alignment. Based on our findings, none of our patients exhibited any noticeable femoral bowing, leading us to conclude that neutral alignment was achieved in all patients. This technique may not be applicable for dysplastic acetabular cases where the affected acetabulum is not at the same level as the healthy side.

One weakness of this study is the lack of consideration for the horizontal offset factor when using different neck lengths for achievement of soft tissue balance. Although this factor can alter limb height, we believe that its impact is minimal. In addition, we were restricted to using a standard stem size, as it was the only size available in our country. Furthermore, when performing THA, this method may prove advantageous exclusively for patients whose acetabulum is positioned neutrally and parallel to the healthy side.

Our data support the application of an efficient and reliable technique that yielded satisfactory outcomes in reducing LLD following a hip arthroplasty procedure. Utilization of this method in clinical practice can be beneficial, particularly in situations where navigation systems and other advanced equipment used in performing measurements are not available. Conduct of additional study comparing this method with other available methods is warranted.

No funding to declare.

No potential conflict of interest relevant to this article was reported.