Effect of body mass index on the conventional method of templating in uncemented total hip arthroplasty: A prospective study

Introduction

Total hip arthroplasty (THA) is one of the most satisfying and effective surgeries performed in orthopedics. Results of THA are very successful in patients with hip arthritis complaining of persistent pain and disability.^[1] Preoperative templating encompasses anticipation of type of implants to be used, size and position, determining offset, calculation of limb length inequality and any potential intra-op problem, using preoperative radiographs to restore biomechanical forces around the hip.^[2,3,4] Periprosthetic fractures as a result of overestimation of implant size as well as component loosening due to underestimation of implant size can be prevented by accurate implant size selection.^[5] Obesity prevalence is increasing in our society due to modern lifestyle.^[6] It has been proposed that the thickness of subcutaneous fat may obscure bony landmarks which deteriorate optimal implant positioning and prolong operative time. The effect of body mass index (BMI) on THA functional outcome, quality of life, and complication rate has been investigated in many clinical studies.^[7] Studies have evaluated the effect of BMI in results and functional outcome of THA. However, few studies have assessed the effect of BMI in templating.^[8] With increasing prevalence of obesity, the effect of BMI on preoperative templating is becoming important for consideration. Studies have shown gender-specific variation in various pelvic parameters.^[9] The effect of gender in templating results are not known. The hypothesis of our study was that BMI and gender have no effect on templating in THA.

Material and Methods

Patients undergoing primary uncemented hip arthroplasty were included in the study. Patients who underwent cemented or hybrid THA or patients with fixed rotational deformity were excluded from the study.

Patient parameters including BMI were documented and recorded.

Well centered and exposed radiographs were taken pre-operatively.^[10,11] As compared with normal anteroposterior (AP) view of pelvis, the AP view was taken with centering on the pubic symphysis. For templating purposes, a wider radiograph was taken with the proximal extent up to antero-superior iliac spine (ASIS) and distal extending to mid shaft of femur. To calculate the offset of the hip accurately on radiographs, the anteversion of hip was also considered. By internally rotating both lower limbs, femoral neck was brought parallel to the film, so that the full length of neck was seen, and offset was calculated properly. To determine the rotation on radiograph the lesser trochanter was assessed. In normal resting position of limb with 15°–20° external rotation, lesser trochanter is more prominent and usually measures more than 5 mm from femoral cortex. However, in 15°–20° of internal rotation, the value lies between 2–5 mm and will not be seen or less than 2 mm in case of excessive internal rotation. All the X-rays were obtained with 100 cm of distance from the tube to the X-ray plate. We controlled magnification by taking a standard object, metallic femoral head of 36 mm in between both thighs near pubis in plane of greater trochanter. Once X-rays were found to be of acceptable magnification and well centered, anatomical and mechanical landmarks were marked on X-ray, as described by Scherlink et al.^[11] Templated size of acetabular and femoral component, limb length difference, and femoral offset were documented.

Results

There were 60 participants in this study with mean age of 55.95 (±9.82) years. The sample comprised of 36 male and 24 female patients. The cases were further divided into three groups based on their BMI as underweight, normal, and overweight. Other demographic features of the study participants are shown in Table 1.

The association of BMI [Table 2] and gender [Table 3] with various preoperative predictions and surgical outcomes were explored but no significant association was found.

Postoperative association of BMI and restoration of offset, limb length discrepancy were also explored [Table 4] and significant association between high BMI and poor restoration of horizontal offset and limb length was found.

Postoperative radiographs were evaluated to assess inclination and anteversion of acetabulum template in 45º ± 10º and 15º ± 10º. All patients were found to have acetabular inclination in safe range with average of 43.33º ± 0.429º and anteversion of 18º.

Discussion

The goal of a successful THA is to achieve a painless and functional hip replacement. Accurate templating in various studies has shown increased precision during surgery, decreased operative time, reduced revision rates, and reduced overall complication rate.^[2,12] Presently, with the advent of new software and 3D templating, more accurate prediction of component and potential intraoperative problems can be anticipated. However, one of the limiting factors in the use of advance templating methods is the affordability of expensive software. Hence, the traditional method of templating still finds its place in the setting where affordability of these software is an issue due to financial constraints.

In our study, we found more accurate prediction of femoral component as compared with acetabular components. We successfully templated 76.6% femoral components, whereas our accuracy of prediction for acetabular component was 63.3%. Results from our study are comparable with results obtained by other authors. The study conducted by Eggli et al.^[3] of 100 THA performed reported predicting more than 90% of the components size accurately. In the study conducted by Untana et al.^[2] in 2009, exact size was estimated in 42.2% of acetabular components and 68.8% of femoral components on templating preoperative radiographs in patients undergoing uncemented THA. Eggli et al.^[3] suggested that decrease in accuracy in prediction was due to the peculiar feature of acetabular component. He attributed this decrease in prediction mainly to the variation in the intraoperative component used. The intraoperative component varied as it was difficult to control the pelvic fixation during surgery and with change in position the acetabulum component varied.

A few studies have postulated the effects of BMI on preoperative templating. The study conducted by Holzer et al.^[12] showed poor size estimation in patients with higher BMI as compared with patients with normal BMI. The decrease in prediction of femoral component in patients with high BMI is due to improper positioning of radio-opaque reference objects while performing X-rays and consequently resulting in magnification errors.^[12] However, results published by Sershon et al. found no significant effect of BMI on results of THA templating. In our study, there were 2 underweight, 42 normal weight, and 16 overweight patients. In our study, we were unable to find any statistically significant effect of BMI on prediction of hip arthroplasty components.

While assessing the effect of gender in templating, Sershon et al.^[8] and Holzer et al.^[12] found no gender-specific differences in prediction of size. Similarly, we had comparable results when gender was in consideration.

Restoration of femoral offset, placement of acetabulum in safe zone, and limb length discrepancy in acceptable limits are goals of successful arthroplasty. Horizontal offset is measured by distance from center of head to femoral axis. Al-Amiry, et al.^[13] in their study showed significant impact of BMI on offset restoration. They ascribed the mechanical difficulty in restoring the offset to the increased adipose tissue and lack of a reliable method of intraoperative assessment of offset restoration. Our study also showed more higher offset variation in the high BMI group of patients. Our offset variation was 37.5% in the overweight group which was found to be statistically significant.

Limb length discrepancy is one of the common complications after THA. Currently, most studies accept less than 10 mm of variation. Intraoperative measurement of limb length in lateral position is done by palpating bony landmarks—patella and heel level and assessing soft tissue tension. Palpation of these landmarks is obscured by drapes and more so difficult in the presence of increased adipose tissue. Our study had 10% of patients who had variation in limb length vis-a-vis 15% in the study by Al-Amiry, et al.^[13] Most of our limb length difference was found in the high BMI group, similar to the study by Al-Amiry, et al.^[13] Statistically analyzing variation of limb length in all BMI groups showed P-value of 0.01 which is significant.

It has been recommended to place acetabular component in safe zone by Lewinnek.^[14] In their study, to reduce the risk of postoperative dislocation, it was recommended to place a cup in 45º ± 10º of inclination and 15º ± 10º of anteversion. In our study, we were able to place all our cups in an acceptable range of inclination. Average inclination and anteversion obtained in our study were 43.33º and 18º respectively. Three of our components had excessive anteversion with no component in retroverted position.

Limitation of our study is in terms of small sample size. However, decrease in incidence of hip disease in the Indian population as compared with the western world explained this low number of hip arthroplasty performed. Presently, more sophisticated methods of templating are available; however, their high cost prevents their use.

Conclusion

We rejected our null hypothesis as there was no statistically significant effect of BMI and gender on component size prediction on conventional radiographs in patients undergoing THA. However, the conventional method of templating resulted in an accurate prediction of 63.3% acetabulum and 76.6% of femoral component in our study. Our study showed significant association between high BMI and limb length discrepancy and poor restoration of femoral offset. Thus, conventional method of templating still is an effective method of estimating the component size.