Total hip replacement performed via a direct anterior approach: A comparison of the lateral and supine position

Abstract

Background

The direct anterior approach (DA) is a recognised approach for performing a total hip replacement (THR). Proponents cite improved recovery times, lower pain levels and improved patient satisfaction in the early post operative period. The procedure can be performed in the supine or lateral position. We wanted to compare the direct anterior approach in lateral decubitus (LD) position and supine (SU) position.

Methods

Single site, non-randomised, multiple surgeon retrospective cohort study between 2014 and 2021 to compare outcomes, complications and implant position for patients undergoing DAA THR in the SU or LD position.

Results

A total of 39 patients (22 lateral/17 supine) were identified. Patients had an average follow up of 45 months (17–81). 95% of the cohort were ASA 1 or 2. The majority of cases were uncemented (95%). A greater implant selection was used in the lateral group and the supine group used mainly implants associated with the supine table. No significant differences were found in post operative oxford hip scores, haemoglobin, length of stay, operative time, cup inclination, offset or post operative leg length. There were no recorded intra-operative complications in the LA group and two in the SU group – one calcar fracture and one canal perforation. No patient in either group has undergone a revision procedure.

Conclusion

Both the supine and lateral position have resulted in satisfactory patient and radiological outcomes. We identified a higher rate of intra-operative fracture in the supine group which is comparable with existing literature. Given the similar outcomes between both groups we would suggest that surgeons wishing to consider the DAA may consider performing this in the lateral position as this will be more familiar to them, they will be able to use their existing implants and do not need a specialised operating table.

1. Background

Total Hip Replacement (THR) is a successful surgical procedure used to treat end-stage osteoarthritis. In the United Kingdom the operation is performed mostly via a posterior or direct lateral approach with less than 1% of arthroplasties performed via the direct anterior approach in 2019¹ In contrast to the posterior or lateral approach the Direct Anterior Approach (DAA) uses a plane between sartorius and tensor fascia lata² It has been suggested to have several advantages compared to other popular approaches through its use of a natural intramuscular and intra-nervous interval. Proponents of this approach cite improved recovery times, lower pain levels, improved patient satisfaction in the early -operative period.3, 4, 5

Recent emphasis on enhanced recovery and the increase in popularity of day case joint replacements⁶ makes the use of an approach that improves immediate post-operative recovery attractive. As such several variations of the procedure have been described and many authors have published their experiences and technical keys to successfully accomplishing this procedure.⁷

Described techniques include in the supine position using specifically designed instruments, a specific operating table and intra-operative fluoroscopy.⁸ An alternative approach would be to perform the procedure using a regular table in the lateral position and without intraoperative imaging. In the United Kingdom at present around 75% of DAA THR are performed in the supine position⁹ and to date there is a very little published literature comparing the lateral and supine positions for anterior approach total hip replacement surgery.¹⁰

We conducted a single site, non-randomised, multiple surgeon retrospective cohort study between 2014 and 2021 to compare outcomes, complications, and cost of performing DAA THR in the supine position using a specific DAA table and image intensifier vs. a standard table without x ray and with the patient in the lateral position.

2. Methods

2.1. Research methods

Single site, non-randomised, multiple surgeon retrospective cohort study. Patients were identified using electronic theatre records and included provided they had undergone a total hip replacement via an anterior approach. Between August 2014 and August 2021, a total of 39 patients were identified. Ethical approval and informed consent not required for this study. This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

2.2. Surgical methods

2.2.1. Pre op

Patients underwent routine pre-operative radiographs prior to surgery. Intravenous antibiotics and tranexamic acid was used pre-operatively prior to the incision.

2.2.2. Position

For patients in the supine position the Medacta™ traction table was used in conjunction with intraoperative c-arm image intensifier. The traction table, using a member of un-scrubbed staff, allows the leg to be manipulated to facilitate access to both the femur and acetabulum using a hydraulic arm and traction post.

In the lateral position patients were on a standard operating table and routine lateral decubitus clamps were used to hold the position on the ASIS and sacrum similar to positioning as for a posterior approach with the exception that the hip is centred more to the posterior edge of the table. This allows for the leg to be dropped below the table with extension, adduction and external rotation of the hip during the femoral canal prep. The surgeon stands in front of the patient during the procedure (opposite to what one would do in a posterior approach). An additional pair of hands (limb holder or second assistant) is needed to control the limb which is positioned by the surgeon during the procedure.

2.2.3. Anaesthesia

Patients were anaesthetised with general anaesthesia unless contraindicated to enable early mobilisation post op and day case procedure.

2.2.4. Approach

All patients had the same approach using Hueter's interval. An incision was made 2 cm distally and laterally to the ASIS and extended distally towards the fibula head around 10 cm to minimise injury to lateral femoral cutaneous neve the fascia overlying the TFL muscle is incised and the muscle is bluntly mobilised off the fascia laterally.]. Deep to the muscle the investing fascia [ fascia innominata] is incided to reveal the lateral circumflex vessels. Deep to this the fat overlying the hip capsule is dissected away with a COBB elevator to reveal the reflected head of the rectus femoris. This serves as a landmark to the anterior edge of the acetabulum and is preserved in most cases. Before performing a capsulotomy it is important to release adhesions between the capsule and the under surface of the gluteus minimus, iliocapsularis and obturator internus tendon especially around the piriform fossa of the femur. The will allow access capsular release at later stage to enable easier elevation of the femur anteriorly for femoral canal prep. Following this a H shaped capsulotomy is performed. The superolateral leaf of capsule is released as far as necessary and up to the obturator internus tendon to facilitate elevation of the femur for canal prep later. Sometimes a section of the superolateral capsule can be resected especially if hypertrophied to give clear access to the femoral neck. Retractors are placed superiorly and inferiorly to the neck and an in-situ neck cut is made with a reciprocating saw. The femoral head is removed with a corkscrew. The acetabulum is exposed and reaming is performed parallel to TAL. Following cup implantation Femoral access can be difficult and can be facilitated with the releases described above.

2.2.5. Placement of implants

In the supine position implants are placed in accordance with pre-operative templating and checked intra-operatively using both manual means and the using intra-operative fluoroscopy during implant seating.

In the lateral decubitus DAA there is no different positioning of the implants to a posterior approach. leg length and offset are checked intraoperatively using manual means without xray and correlating with preop planning which is essential in all cases. The cup anteversion is parallel to TAL and inclination such that the inferior edge of the cup lies inside the TAL. This automatically gives the appropriate anteversion [ more neutral than posterior approach] and inclination of the prosthesis. The femoral entry point is critical to prevent anteroposterior alignment of the stem on the lateral xray. The entry point for broaching should be as posterior and lateral in the piriformis fossa of the femur. Accessing this is facilitated by the releases mentioned before and using offset broach handles.

2.2.6. Post operatively

Post-operatively patients were advised to weight bare as soon as able and no post-operative restrictions were placed on the patient. Prophylaxis against venous thrombo-embolism was with clexane dosed per patient weight for 28 days. Simple oral analgesia was provided. Radiographs are obtained post-operatively prior to discharge for all patients who did not undergo intraoperative fluoroscopy.

3. Results

A total of 39 patients (22 lateral/17 supine) were identified. The majority (31) were female. The mean age of the cohort was 63 years (46–85). Patients had an average follow up of 45 months (17–81). 95% of the cohort were ASA 1 or 2 [Fig. 1].

Fig. 1.

Participant cohort ASA grade distribution.

The majority of cases were uncemented (95%) however two patients in the LA group had a hybrid THR. A greater selection of implants were used in the lateral group and the SU group used mainly implants associated with the supine operating traction table [Fig. 2].

Fig. 2.

Lateral vs. Supine group implant selection.

No significant differences were found in post operative length of stay with LA mean LOS 1.8 days (1–8) SU mean LOS 1.2 (1–10) which was not significant on t-testing (p = 0.3).[Fig. 3].

Fig. 3.

Lateral vs. supine group inpatient length of stay.

Overall 1/3 of the cohort were performed as day case procedures and 2/3 had been discharged after a single overnight stay as an inpatient.

Blood loss was similar in both groups LA mean 23.57 (5–48) and SU 19.19 (11–33) with no statistically significant difference between the groups (p = 0.155) and no post operative transfusions required in any patient.

Operative time was the same regardless of position [Fig. 4] and did not reach significance (p = 0.861). The mean duration of operation was 1 h 31 min across both groups.

Fig. 4.

Lateral vs. supine group operative time.

Implant position was assessed by two independent assessors who were blinded from each other and their values aggregated. Neither assessor was an operating surgeon in the series. The position was assessed on routine post operative radiographs and no significant difference was noted in cup inclination, leg length or offset between the groups [Fig. 5].

Fig. 5.

Lateral vs. supine group radiographic assessment of implant position.

There were no recorded intra-operative complications in the LA group and two in the SU group – one calcar fracture and one canal perforation although this was not deemed statistically significant with Fisher exact testing (p = 0.1835). No patient in either group had undergone a revision procedure or dislocation to date. Oxford hip scores were equitable in both groups with LA mean OHS 37.5 (28–48) and SU 43.6 (29–48). This was not statistically significant (p = 0.15).

Patients in the LA group required a third assistant in theatre to hold the leg externally rotated, patients in the SU group required an additional staff member to operate the table and use of a radiographer and image intensifier.

4. Discussion

In our study satisfactory restoration of offset, leg length and acetabular component orientation was achieved regardless of the position of the patient. Additionally operative time was comparable for both approaches. One third of patients were discharged on the day of surgery and an additional third discharged early the following day.

There was a higher rate of intraoperative complications in the SU group however it is worth noting that this complication occurred within the surgeons first 10 cases and may be related to the learning curve associated with the DA approach. The presence of a learning curve associated with the implementation of a new operative technique whether that be patient position, the operative approach or the choice of instrumentation and implants is well documented in the surrounding literature.¹¹

Yi et al., 2013. reported a high incidence of complications during the initial learning curve associated with the implementation of the DA approach in the supine position, this study showed an overall complication rate of 16.4% with 6.5% reoperation rate during the learning curve.¹² Previous analysis of the learning curve when first using the DA approach via the supine position has determined that, operative efficiency can be gained over a sequence of 15 cases however a slower improvement in estimated blood loss and complications due to process of learning new anatomy can be expected,^11,13 however whilst the components and consequences of the learning curve are fairly well established, the documented length of the learning curve is highly variable in the associated literature and varies from as low as 4 cases to as many as 200 cases.^14,15

The literature also demonstrates that it is not just complication rates and patient outcomes that are associated with a learning curve. Masonis 2008 and Rathod 2014 when performing analysis of post-operative radiographs recorded an improvement in acetabular cup inclination, stem alignment and leg length discrepancy after 100 cases.^16,17 This was not observed in our study and with all implants placed into an acceptable position and no post operative dislocations in the current follow up period.

Comparatively unlike the DA approach in the supine position, previous analysis of the DA approach in the lateral decubitus position has not been able to identify a learning curve which has been concluded secondary to the operating surgeon using a standard patient position alongside the familiarity of a routine operating table, positioning devices, and instrumentation.^18,19 In our series the intra-operative complication rate did not reach significance – however given the relatively low numbers in out series the risk of Type II error must be considered.

The study is limited by its low numbers however the authors feel it is representative of practice in the United Kingdom in which surgeons often perform the approach in selected cases and therefore are less familiar with the approach as a result. This is reflected in the National Joint Registry data with the majority of surgeons demonstrating a preference of the posterior approach and those that are performing DA approach hips do so in a supine position.

5. Conclusion

Both the supine and lateral position have resulted in satisfactory patient and radiological outcomes in our series. We identified a higher rate of intra-operative fracture in the supine group which is comparable with existing literature. Given the similar outcomes between both groups we would suggest that surgeons wishing to consider the DAA may consider performing this in the lateral position as it if more familiar, they may use their existing implants, do not require intraoperative fluoroscopy, require fewer staff in theatre and do not need a specialised operating table.

Funding

This research did not receive any specific grant from funding agencies in the public, commercial or not-for-profit sectors .

Authors contributions

Rogers Gareth: Methodology, Investigation, Formal Analysis, Writing – Original Draft, Writing – Review & Editing. Hoggett Lee: Conceptualization, Supervision, Methodology, Investigation, Formal Analysis, Writing – Original Draft, Writing – Review & Editing. Bhutta Aqeel: Supervision, Original Draft, Writing. Bonshahi Ardeshir: Supervision, Original Draft, Writing.

Declaration of competing interest

None.