Smartphone-assisted technique in total hip arthroplasty can improve the precision of acetabular cup placement: a randomised controlled trial

Abstract

Background:

Malposition of the acetabular component mainly results from intraoperative pelvic motion and manual errors during cup placement. We investigated the outcomes of a device that uses a level indicator application of a smartphone to function as a pelvic tilt goniometer to assess intraoperative motion. The cup positions outside Lewinnek’s safe zone were compared between the smartphone-assisted technique and the conventional method that uses a mechanical alignment guide.

Methods:

A randomised controlled trial was conducted among 64 patients receiving primary cementless total hip arthroplasties (THAs) via a posterolateral approach from June 2015 to February 2017. Acetabular components were implanted either by conventional technique (n = 32) or using smartphone technique (n = 32). Inclination and anteversion angles were measured in supine pelvic radiographs.

Results:

The inclination angle in the smartphone group was not significantly different from the conventional group (41.2° ± 3.9° vs. 40.3° ± 7.9°, p = 0.567). The anteversion angle was also similar (19.3 ± 3.8° vs 19.1° ± 5.9°, p = 0.856). However, the standard deviation of the angle in the smartphone group was significantly lower for inclination (p < 0.001) and anteversion (p = 0.016). There were 3 outliers (9.4%) in the smartphone group, but 13 (40.6%) in the conventional group (p = 0.008). The risk ratio was 0.23 (95% CI, 0.07–0.73). The risk difference was −0.31 (95% CI, –0.51– –0.11).

Conclusion:

The smartphone-assisted technique in THA improves the precision of cup placement and decreases the percentage of safe zone outliers.

Thai Clinical Trials Registry (ID: TCTR20151123002).

Keywords

Acetabular cup positioning Lewinnek safe zone smartphone total hip arthroplasty

Introduction

Proper acetabular cup position has a direct influence on the outcomes of total hip arthroplasty (THA). The most acceptable range of angles is that of the safe zone advocated by Lewinnek et al.¹ (15° ± 10° anteversion and 40° ± 10° inclination). The conventional methods of cup placement, guided by intraoperative anatomical landmarks or using mechanical alignment guides (MAG), can provide 25.7–70.5% of cup placement in Lewinnek’s safe zone.²

The safe zone outliers can be reduced if the surgeon uses a device such as a bubble level or a digital protractor to aim for a predetermined target.^3,4 The benefit of smartphone technology that uses the accelerometer and camera functions to improve the accuracy of cup placement was reported in both clinical and cadaveric studies.^5,6 However, they were based on the assumption of perfect patient positioning in lateral decubitus and no intraoperative pelvic motion. This assumption is unlikely to be found during THA.

We invented a device that uses a level indicator application of a smartphone to function as a pelvic tilt goniometer to assess pelvic motion. We hypothesised that our device would increase the precision of acetabular cup positioning within Lewinnek’s safe zone. This study aims to investigate the postoperative position of the acetabular component, comparing between the conventional method that uses a MAG alone, and the new technique that uses a MAG combined with our device.

Methods

Patients

A randomised controlled trial was conducted among patients receiving primary cementless THAs from June 2015 to February 2017. All surgeries were performed by the same surgeon using a posterolateral approach. The inclusion criteria were age >25 years old and hip osteoarthritis or femoral neck fracture with indications for THA. The exclusion criteria were patients with kypho-scoliosis or pelvic deformity.

A computer program generated a random sequence of the 2 techniques for each group and the result was kept in a sealed envelope. In the operating room, after the patient was positioned in lateral decubitus, the envelope was opened to indicate the group to which the patient had been allocated. This study received ethical approval from our institutional review board (Code 40/58) and registered in the Thai Clinical Trials Registry (ID: TCTR20151123002). Informed consent was given by every patient who participated in the study.

Operative technique

In the conventional group, the operating table was set at 0° of tilt before the patient was positioned in lateral decubitus by using 2 supports (over the anterior superior iliac spine (ASIS) and the sacrum). The acetabular components were placed using a MAG with the target of 37° operative inclination (OI) and 25° operative anteversion (OA) in order to achieve 40° radiographic inclination (RI) and 20° radiographic anteversion (RA).⁷ This could be achieved by moving the cup positioner until the MAG, set at 45° inclination and 20° anteversion, was about 8° below the horizontal and pointed about 5° posterior to the shoulder of the patient.

In the smartphone group, a skin staple was put over the tip of the coccyx as a radiographic marker prior to positioning the patient. The phone (Lenovo A390, China, with Smart Level application, version 1.0, androidboy1.blogspot.com) was enclosed in a zip bag and placed in a holder (Yunteng self picture monopod YT-188, China) that was connected to a detachable stainless stem. After skin preparation and partial draping, this device was securely installed on the iliac tubercle by 2 self-drilling screws (Figure 1 (A) and (B)). Pelvic position was standardised by using cross-table fluoroscopy. A plumb was dangled on the image-reception screen and a long K-wire was taped on the screen to represent this line. The operating table was tilted until the K-wire was parallel to the inter-teardrop line and the coccyx marker was in the midline of pubic symphysis. The C-arm was rotated horizontally until the positional relationship between the tip of the coccyx and the superior border of the pubic symphysis was similar to that in the preoperative supine antero-posterior (AP) radiograph (Figure 1 (C) and (D)). To mark this plane, a cross-line laser level was touched onto the flat surface of the C-arm. It projected a perpendicular laser line to the border of the smartphone (Figure 1 (E) and (F)). The axis of the smartphone was adjusted horizontally to be parallel with the laser line. The relative position of the phone to the ground was calibrated to 0° along both axes by touching the ‘calibrate’ screen-icon (Figure 2 (A)). The surgeon disconnected the inclinometer from its base and then finished draping. Before making the skin incision, the operating table was tilted back to its original position in order to eliminate the effect of fluoroscopy.

Figure 1.

The pelvic inclinometer comprised a smartphone, a plastic holder, a stainless steel connector and base. It was installed to the iliac tubercle by 2 screws ((A) and (B)). A long K-wire was taped on the fluoroscopic image-reception screen to represent a plumb line. The operating table was tilted until the K-wire was parallel to the inter-teardrop line and the coccyx marker was in the midline. The C-arm was rotated horizontally until the positional relationship between the coccyx and the pubic symphysis was similar to that in the preoperative radiograph ((C) and (D)). The plane of pitch axis was identified by placing a cross-line laser level on the C-arm. The smartphone holder was adjusted horizontally to be parallel with the laser line ((E) and (F)).

At the time of cup placement, the inclinometer was reconnected to its base and the operating table was returned to the previous standardised position by obtaining 0° in both axes of the application. Another smartphone (Lenovo A390) with the Clinometer application (www.plaincode.com) was placed onto the cup positioner and they were moved vertically together until 37° OI was achieved. Thereafter, they were moved horizontally to aim the MAG, set at 20° OA, about 5° posteriorly to the border of the inclinometer before final cup impaction (Figure 2 (B) and (C)).

Figure 2.

After calibration to 0°, the tilting degrees shown in the Smart Level application represented the intraoperative pelvic motion from the standardised position (A). The positioner was aimed at 37° OI verified by another smartphone with the Clinometer application (B). The MAG, set at 20° OA, was pointed 5° posterior to the border of the inclinometer smartphone (C).

In both groups, the short external rotators and posterior capsule were repaired. The postoperative antibiotic was intravenous cefazolin for 24–48 hours. The sample size, calculated to provide an 80% statistical power of detection and a 2-sided type I error level of 0.05 plus 10% drop-out, was 32 hips per group. This was based on expected cup placements within Lewinnek’s safe zone of 88% in the smartphone group similar to computer navigation,⁸ and 55% in the conventional group according to our previous THAs.

Statistical analysis

The study parameters were diagnosis, age, gender, body mass index, operative time, postoperative dislocation and infection. The radiological results were measured in the pubic-symphysis-centred pelvic anteroposterior (AP) radiographs, taken in the supine position at 6 weeks postoperatively. The cup inclination was determined by measuring the angle between the inter-teardrop line and a line drawn parallel to the opening plane of the cup. The anteversion was measured using the method described by Widmer.⁹

The primary outcome was the percentage of cup placements outside Lewinnek’s safe zone. The secondary outcomes were the RI, RA angles and the deviation of the measured angle from the target position (40° RI, 20° RA). The radiographic assessments were performed by 2 orthopaedic residents, who did not know to which group the patient was assigned and repeated again 2 weeks later. The average of 4 measurements was assumed as the real cup position. The intra-class correlation coefficients (ICC) with the 2-way random-effects model and absolute agreement were calculated for intra-and inter-observer reliability. We used the exact probability test to compare categorical data. Continuous data were analysed by using the t-test and Mann-Whitney U test. A p-value of <0.05 was considered significant.

Results

Demographic data did not differ significantly between the 2 groups. Plasmafit cups (Aesculap, Tuttlingen, Germany) were implanted in 41 cases (64%), Allofit cups and Trilogy cups (Zimmer Biomet, Warsaw, Indiana, USA) in 15 cases (23%) and 8 cases (13%) respectively (Table 1).

Table 1.

Patients’ baseline characteristics and prosthesis data.

Data	Conventional(n = 32)	Smartphone(n = 32)	p-value
Age mean (SD)	52.6 (8.0)	56.0 (7.1)	0.076
range	35–64	41–71
Gender M:F	20:12	15:17	0.315
BMI mean (SD)	21.9 (4.1)	23.5 (3.8)	0.118
range	17.6– 31.6	17.5–30.4
Diagnosis:
Osteonecrosis of femoral head	16 (50.0%)	13 (40.6%)	0.601
Femoral neck fracture	6 (18.8%)	4 (12.5%)
Hip dysplasia	5 (15.6%)	5 (15.6%)
Inflammatory joint disease	3 (9.4%)	4 (12.5%)
Others	2 (6.2%)	6 (18.8%)
Cementless cup type:
Plasmafit	17 (53.1%)	24 (75.0%)	0.115
Allofit	11 (34.4%)	4 (12.5%)
Trilogy	4 (12.5%)	4 (12.5%)
Femoral head size:
28 mm	4 (12.5%)	4 (12.5%)	0.755
32 mm	2 (6.3%)	4 (12.5%)
36 mm	26 (81.2%)	24 (75.0%)

BMI, body mass index; SD, standard deviation.

The mean inclination was 41.2° ± 3.9° (range 32.9–48.9°) in the smartphone group and 40.3° ± 7.9° (range 26.5–56.5°) in the conventional group (p = 0.567). The mean anteversion was 19.3° ± 3.8° (range 12.3–28.0°) in the smartphone group and 19.1° ± 5.9° (range 0.5–28.4°) in the conventional group (p = 0.856) (Table 2). The standard deviation (SD) of the angle in the smartphone group was significantly lower for inclination (3.9° vs. 7.9°, p < 0.001) and anteversion (3.8° vs. 5.9°, p = 0.016) (Figure 3).

Table 2.

Comparison of radiographic measurements and outliers.

Data	Conventional(n = 32)	Smartphone(n = 32)	p-value
Inclination angle (°) mean (SD)	40.3 (7.9)	41.2 (3.9)	0.567
95% CI	37.5–43.2	39.8–42.6
range	26.5–56.5	32.9–48.9
Anteversion angle (°) mean (SD)	19.1 (5.9)	19.3 (3.8)	0.856
95% CI	16.9–21.2	17.9–20.6
range	0.5–28.4	12.3–28.0
Deviation from 40° inclination (°) mean (SD)	0.3 (7.9)	−0.7 (3.8)	0.494
95% CI	−2.5 to 3.2	−2.0 to 0.6
range	−14 to 17	−7 to 9
Deviation from 20° anteversion (°) mean (SD)	0.9 (5.9)	0.7 (3.8)	0.643
95% CI	−3.1 to 1.2	−2.1 to 0.6
range	−20 to 8	−8 to 8
Safe zone outliers n (%)	13 (40.6)	3 (9.4)	0.008

SD, standard deviation; CI, confidence interval.

Figure 3.

Radiographic inclination (A) and anteversion (B) angles of the acetabular component.

The deviation of the postoperatively measured angles from the target position (40° RI, 20° RA) for the inclination was −0.7° ± 3.8° (range −7 to 9°) in the smartphone group and 0.3° ± 7.9° (range −14 to 17°) in the conventional group (p = 0.494). The value of deviation for the anteversion was 0.7° ± 3.8° (range −8 to 8°) in the smartphone group and 0.9° ± 5.9° (range −20 to 8°) in the conventional group (p = 0.643). Likewise, the SD values of these deviated angles in the smartphone group were significantly lower for inclination (p < 0.001) and anteversion (p = 0.016).

There were 3 safe zone outliers (9.4%) in the smartphone group, but 13 outliers (40.6%) in the conventional group (p = 0.008) (Figure 4). The risk difference was −0.31 (95% CI, –0.51 to −0.11). The risk ratio was 0.23 (95% CI, 0.07–0.73). The ICC for intra- and inter-observer reliability of inclination measurements were 0.87 and 0.93 respectively, and those of anteversion measurements were 0.76 and 0.80 respectively. The mean operative time was 14 minutes longer in the smartphone group than the conventional group (136 ± 34 vs 122 ± 28 minutes, p = 0.069). The average duration of follow-up was 29 ± 6 months (range 16–40). 1 hip in the conventional group, but none in the smartphone group, had posterior dislocation (p = 1.000). This hip had cup anteversion of 3° and inclination of 38°. The stem anteversion, measured by postoperative CT scan, was 23°. Closed reduction was performed and no recurrent dislocation occurred. There was no surgical site infection during the follow-up period.

Figure 4.

Distribution of inclination and anteversion angles measured in the postoperative radiographs.

Discussion

This study demonstrated no differences between the 2 groups with regard to the mean inclination, anteversion and deviation from the desired target position. These results represent similar accuracies of both implanting methods. Nevertheless, there were significantly lower SD values of all measured angles in the smartphone group. This indicates the superiority of this technique in terms of precision or a smaller variation in the positioning of the acetabular component. There were 9.4% outliers in the smartphone-assisted THAs, but 40.6% in the conventional group. The risk of outliers was significantly reduced by 77% and the absolute difference was –31%. Using this technique in an estimated 3 THAs would prevent 1 outlier of cup placement.

The significant benefit found in this study affirmed the same finding of our previous clinical trial. In a previous non-randomsed, historical controlled study,¹⁰ we found the smartphone and our device could improve the precision of cup positioning inside Lewinnek’s safe zone. However, it was not a concurrent study and might be biased by a period effect or surgeon experience. The target operative inclination was 36° in the smartphone group and 40° in the conventional group. These resulted in significantly lower radiographic inclination angles in the smartphone group than in the conventional group.¹⁰ The aim of the current study was to evaluate whether this significant improvement would remain in a randomised controlled trial design. The operative inclination was similarly aimed at 37° in both groups and we found no significant difference of the mean radiographic inclination angles between the 2 groups. Although the percentage of outliers in the conventional group reduced from 43.9% in the previous study to 40.6% in the current study, the safe zone outliers were similarly about 10% in the smartphone group.

Malposition of the acetabular component mainly results from the errors of preoperative pelvic positioning, intraoperative pelvic motion, and manual errors during the surgery.¹¹ Repositioning of the operating table by using the cross-table fluoroscopy can correct the tilting errors of the pelvis before surgery both in yaw and roll axes. The pitch axis could be achieved by rotating the C-arm horizontally and marking it with tape on the floor to be visible and used to determine the cup anteversion during the surgery.¹¹ Likewise, we verified this plane by adjusting the inclinometer axis to be parallel with the laser line perpendicularly projected from the C-arm. Aiming the MAG reference to the border of the smartphone was more practical than marking on the floor with tape.

The intraoperative pelvic motion can be caused by an inadequate stabilsation of the pelvis on the table, rolling of the leg during surgery, and movement generated by the surgical procedure. This problem can be handled by inserting a K-wire into the iliac crest and parallel to a line linking the left and right ASIS. Pelvic AP radiographs were taken to confirm its parallelism with the inter-teardrop line. When positioning the cup, the vertical arm of the MAG was aligned to the K-wire.¹² Some investigators inserted a Steinmann pin and connected it with an adjustable bubble-level.^13,14 The level was adjusted until the bubble was centred before starting the surgery. During the cup placement, the surgeon could tilt the table to correct the roll of the patient when the bubble was re-centred. Echeverri et al.³ used a bull’s-eye bubble level connected to a Schanz pin to identify yaw and roll pelvic motions. Asayama et al.¹⁵ invented a special device using 2 mechanical goniometers and a pendulum. It was fixed with a threaded Steinmann pin to assess yaw and roll motions. A digital compass was incorporated into the device to measure the pitch motion. Our devised inclinometer can present yaw and roll motions in digital numbers and can alert the surgeon when the 0° are shown in both axes of the application. It is less complicated, smaller and lighter than those instruments used in the previous studies. To assess the pitch motion, the lateral border of the smartphone was parallel with the coronal plane of the patient after verification by a laser line and did not change throughout the surgery because of a secure fixation with screws. It was the reference for aiming the horizontal arm of the MAG during the cup positioning.

Manual errors during surgery can be reduced by several techniques, including the use of computer navigation, robotic-assisted navigation and other devices that enable the surgeon to aim for a planned target. Computer navigation can increase the percentage of placements in the safe zone (80–96%)¹⁶ compared with conventional insertion (26–71%),² whereas 100% within the safe zone can be achieved by robotic-assisted navigation.¹⁷ The percentage of outliers for inclination angle can be significantly minimised by using a bubble level,^3,14,18 a digital protractor,⁴ a digital inclinometer or a smartphone with the Angle application.^5,6,19 Our results confirmed similar benefits of a smartphone with the Clinometer application. These devices are positioned on the handle of the introducer prior to impaction and can reduce this manual error because the inclination angle usually declines by almost 3° during impaction of a cementless cup.¹⁹

There are several limitations of this study. Firstly, we used intraoperative fluoroscopy to adjust pelvic position only in the smartphone group. It is possible that the improved cup orientation came from using fluoroscopy. However, its positive effect was eliminated by the return of the operating table to its original position before starting the surgery. Secondly, our target RA was 20° because we set the goal for femoral component anteversion at 15–20° and aimed to achieve a combined anteversion of 25–50°. The percentage of outliers might differ if we set the target RA at 15°. Thirdly, we used the Widmer method to measure the anteversion which might have some variation of measurement.⁹ Ideally, a reformatted CT scan should be used as a reference standard in determination of cup anteversion,²⁰ but its special software is not available in our hospital. Nevertheless, measurements with the Widmer method are very close to those using reformatted CT, with a mean difference of −0.9°.²⁰ A good level of ICC in this study indicated the high reliability of anteversion measurements. Finally, we could not detect significant differences with regard to dislocation or infection because of the small sample size. To the best of our knowledge, this is the first randomised controlled study that uses a smartphone as a pelvic inclinometer.

Conclusion

The smartphone-assisted technique can improve the precision of cup placement in THA at a desired target in Lewinnek’s safe zone. It can correct the main causes of acetabular component malposition and provides an optional tool for the surgeon to achieve a better cup position.

Footnotes

Acknowledgements

The authors wish to thank Andrew Sherratt for his help in approving the English language.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship and/or publication of this article.

References

Lewinnek

Lewis

Tarr

, et al. Dislocations after total hip-replacement arthroplasties. J Bone Joint Surg Am 1978; 60: 217–220.

Callanan

Jarrett

Bragdon

, et al. The John Charnley Award: risk factors for cup malpositioning: quality improvement through a joint registry at a tertiary hospital. Clin Orthop Relat Res 2011; 469: 319–329.

Echeverri

Leyvraz

Zambelli

, et al. Reliable acetabular cup orientation with a new gravity-assisted guidance system. J Arthroplasty 2006; 21: 413–419.

Meermans

Goetheer-Smits

Lim

, et al. The difference between the radiographic and the operative angle of inclination of the acetabular component in total hip arthroplasty: use of a digital protractor and the circumference of the hip to improve orientation. Bone Joint J 2015; 97-B: 603–610.

Peters

Greeff

Goldstein

, et al. Improving acetabular cup orientation in total hip arthroplasty by using smartphone technology. J Arthroplasty 2012; 27: 1324–1330.

Kurosaka

Fukunishi

Fukui

, et al. Assessment of accuracy and reliability in acetabular cup placement using an iPhone/iPad system. Orthopedics 2016; 39: e621–e626.

Murray

DW.

The definition and measurement of acetabular orientation. J Bone Joint Surg Br 1993; 75: 228–232.

Kalteis

Handel

Bäthis

, et al. Imageless navigation for insertion of the acetabular component in total hip arthroplasty. Is it as accurate as CT-based navigation? J Bone Joint Surg Br 2006; 88: 163–167.

Widmer

KH.

A simplified method to determine acetabular cup anteversion from plain radiographs. J Arthroplasty 2004; 19: 387–390.

10.

Pongkunakorn

Chatmaitri

Diewwattanawiwat

Use of smartphone to improve acetabular component positioning in total hip arthroplasty: a comparative clinical study. J Orthop Surg 2019; 27: 1–8.

11.

Nishikubo

Fujioka

Ueshima

, et al. Preoperative fluoroscopic imaging reduces variability of acetabular component positioning. J Arthroplasty 2011; 26: 1088–1094.

12.

Ishidou

Hirotsu

Setoguchi

, et al. A Kirschner wire as a transverse-axis guide to improve acetabular cup positioning. J Orthop Surg (Hong Kong) 2016; 24: 22–26.

13.

Goldstein

Jimenez

Gordon

, et al. Use of a sentinel pin as a guide to acetabular component anteversion in total hip arthroplasty. J Bone Joint Surg Am 2006; 88(Suppl. 4): 97–100.

14.

Leone

Nevelos

Fonti

, et al. A new surgical instrument and technique designed to achieve more accurate component placement in total hip replacement. Tech Orthop 2011; 26: 217–221.

15.

Asayama

Akiyoshi

Naito

, et al. Intraoperative pelvic motion in total hip arthroplasty. J Arthroplasty 2004; 19: 992–927.

16.

Zhang

, et al. Computer navigation in total hip arthroplasty: a meta-analysis of randomized controlled trials. Int J Surg 2014; 12: 528–533.

17.

Domb

El Bitar

Sadik

, et al. Comparison of robotic assisted and conventional acetabular cup placement in THA: a matched-pair controlled study. Clin Orthop Relat Res 2014; 472: 329–336.

18.

Darrith

Bell

Culvern

, et al. Can the use of an inclinometer improve the positioning of the acetabular component in total hip arthroplasty? Bone Joint J 2018; 100-B: 862–866.

19.

Beverland

O’Neill

Rutherford

, et al. Placement of the acetabular component. Bone Joint J 2016; 98-B(Suppl. A): 37–43.

20.

Nomura

Naito

Nakamura

, et al. An analysis of the best method for evaluating anteversion of the acetabular component after total hip replacement on plain radiographs. Bone Joint J 2014; 96-B: 597–603.