DOI: 10.5301/hipint.5000108
Hip Int 2014; 24 ( 3): 277-283
Original Article
Anterior minimally invasive approach for total hip replacement: five-year survivorship and learning curve Daniel A. Müller, Patrick O. Zingg, Claudio Dora Department of Orthopaedics, University of Zürich, Balgrist Hospital, Zürich - Switzerland
Opponents associate minimally invasive total hip replacement (THR) with additional risks, potentially resulting in increased implant failure rates. The purpose was to document complications, quality of implant positioning and five-year survivorship of THR using the AMIS approach and to test the hypothesis that eventual high complication and revision rates would be limited to an early series and be avoided by junior surgeons who get trained by a senior surgeon. A consecutive series of 150 primary THR implanted during the introduction of the AMIS technique in the department was retrospectively analysed for complications, implant positioning and implant survival after a minimum of five years. Survivorship curves of implants were compared between different surgeons and time periods. Due to implant revision for any reason the five-year survival rate was 94.6%, 78.9% for the first 20 and 96.8% for the following 130 AMIS procedures (p = 0.001). The hazard ratio for implant failure was 0.979 indicating a risk reduction of 2% every further case. The five-year implant survivorship of those procedures performed by two junior surgeons was 97.7%. We conclude that adoption of AMIS temporarily exposed patients to a higher risk of implant revisions, which normalised after the first 20 cases and that experience from a single surgeon’s learning curve could effectively be taught to junior surgeons. Keywords: Learning curve, Primary total hip replacement, Anterior minimally invasive surgery, Survivorship Accepted: September 10, 2013
INTRODUCTION Despite increasing use of minimally invasive surgical (MIS) techniques for total hip replacement (THR), current literature fails to demonstrate more than only short term advantages and risks versus benefits are still an ongoing debate (1). Opponents associate MIS with additional risks such as infection, neurovascular injury, femoral fracture and component mal-positioning potentially resulting in increased implant failure rates (2, 3). Additionally, new implant designs without long-term documentation are introduced to adopt the needs of MIS techniques. One of the goals of minimally invasive THR is to minimise trauma to underlying muscles and tendons. Having been trained using the direct lateral approach (4), the senior author used this approach until 2004. Dissatisfaction regarding this
approach aroused due to active patients claiming residual trochanteric pain and limping when trying to restart physical activities beyond those of daily living. A study comparing magnetic resonance imaging (MRI) findings on tendons and muscles between active patients claiming residual pain and limping and asymptomatic patients after THR revealed a significant association between residual pain and soft tissue damage (5). This was the author’s rationale to look for a “less invasive” approach for THR. From an anatomical point of view the anterior approach to the hip (6) was expected to ideally fulfill the goal of minimising soft tissue trauma because of the inter-muscular and inter-nervous plane explored and because reports using this approach were promising (7). Thus, an anterior minimally invasive surgical (AMIS) approach using a leg holder and relatively new implants (Quadra®-H stem and Versafit®-CC cup, Medacta, Castel
© 2014 Wichtig Publishing - ISSN 1120-7000
277
AMIS THR: survivorship and learning curve
San Pietro, Switzerland) was introduced in our department in 2005. Since figures from the 2010 and 2011 annual report of the Australian implant registry on the same implants showed high one-year cumulative revision rates of 2.8% and 2.1%, respectively, a retrospective analysis of our patient cohort was undertaken. The aim was to document complications, quality of implant positioning and five years survivorship of the Quadra®-H stem and Versafit®-CC cup implanted by AMIS. Our hypothesis was that: 1) eventual high complication and revision rates would be limited to an early series; and 2) mistakes leading to high complication and revision rates within an early series would be avoided by juniors trained for AMIS by a senior surgeon.
MATERIAL AND METHODS
protecting the TFL muscle was performed and a soft tissue retractor aimed to be anchored within the joint capsule used. An offset reamer and cup impactor was used for acetabular preparation and cup placement. For femoral broaching the leg was positioned in about 20° to 30° of extension, 90° of external rotation and 20° of adduction using the leg positioner and straight Judet type broach handles without offset. All surgeries were performed under a laminar air flow system and three doses of cefuroxime 1.5 g in eight hourly intervals starting 30 minutes before incision were administered and enoxaparin natrium daily for six weeks were given to all patients. Weight bearing as tolerated on two crutches was recommended for two weeks; afterward full weight bearing was encouraged.
Data acquisition
All THRs implanted by AMIS technique performed between January 2005 and April 2006 were retrospectively included and recalled after a minimum follow-up of five years in April 2011. During this time period the senior surgeon used AMIS for every primary THR unless trochanteric advancement was deemed necessary to achieve true hip reconstruction or hardware removal required the use of a lateral or posterior approach. Trochanteric advancement was indicated in hips with sequelae of childhood hip disease having had prior surgery, which resulted in a highly distorted anatomy of the proximal femur. The senior author performed the first 22 AMIS before two junior surgeons were consecutively involved. A Quadra®-H stem and Versafit®-CC cup (Medacta, Castel San Pietro, Switzerland) together with a 28 mm CoCr ball and a highly cross-linked polyethylene inlay was used in all patients of this consecutive series.
Demographic parameters, peri-operative and late complications were recorded from electronic patient’s charts. Additionally, a standardised interview asking for all medico-surgical interventions concerning the affected hip performed at a follow-up visit appointed for the purpose of this study after a minimum of five years was performed. Anteroposterior pelvic x-rays acquired in the operation room immediately after surgery were analysed for; stem alignment, leg length discrepancy, cup inclination and version were measured (9) by one independent orthopedic surgeon (D.A.M.). The cumulative five years survival of the implants was estimated using Kaplan Meyer survival analysis with implant revision for any reason as endpoint. Subjective and clinical outcome was assessed using the WOMAC score (10), the subjective hip value (in accordance to the subjective shoulder value (11)) and the Harris Hip Score (12).
Surgical technique and treatment protocol
Statistics
A modified Hueter approach (6) in the supine position on a leg positioner (AMIS® mobile leg positioner, Medacta, Castel San Pietro, Switzerland) approximating Judet’s description (8) was performed as follows. In order to avoid lesions to the lateral cutaneous nerve of the thigh, the skin incision was centered on the top of the tensor fascia latae muscle (TFL) and its facial sheath entered before blunt dissection along its medial border. After ligature of the ascending branch of the lateral circumflex femoral artery a U-shaped capsulotomy aiming for a laterally based flap
A biostatistician performed all calculations. SPSS for Windows was used for all analysis. Distribution of continuous variables was assessed graphically. Normally distributed data were presented as means ± SD or mean and confidence intervals and non-parametric data with medians and ranges. Implant survival curves of different surgeons and of different chronological periods were compared using a Log rank test and Cox regression, respectively. The paired Wilcoxon signed ranks test was used to compare continuous non-parametric values at the time of diagnosis with
278
© 2014 Wichtig Publishing - ISSN 1120-7000
Müller et al
those at the time of follow-up. A Mann-Whitney U test was used to compare continuous non-parametric postoperative outcome parameters of different groups. Significance level was set at p
Hip Int 2014; 24 ( 3): 277-283
Original Article
Anterior minimally invasive approach for total hip replacement: five-year survivorship and learning curve Daniel A. Müller, Patrick O. Zingg, Claudio Dora Department of Orthopaedics, University of Zürich, Balgrist Hospital, Zürich - Switzerland
Opponents associate minimally invasive total hip replacement (THR) with additional risks, potentially resulting in increased implant failure rates. The purpose was to document complications, quality of implant positioning and five-year survivorship of THR using the AMIS approach and to test the hypothesis that eventual high complication and revision rates would be limited to an early series and be avoided by junior surgeons who get trained by a senior surgeon. A consecutive series of 150 primary THR implanted during the introduction of the AMIS technique in the department was retrospectively analysed for complications, implant positioning and implant survival after a minimum of five years. Survivorship curves of implants were compared between different surgeons and time periods. Due to implant revision for any reason the five-year survival rate was 94.6%, 78.9% for the first 20 and 96.8% for the following 130 AMIS procedures (p = 0.001). The hazard ratio for implant failure was 0.979 indicating a risk reduction of 2% every further case. The five-year implant survivorship of those procedures performed by two junior surgeons was 97.7%. We conclude that adoption of AMIS temporarily exposed patients to a higher risk of implant revisions, which normalised after the first 20 cases and that experience from a single surgeon’s learning curve could effectively be taught to junior surgeons. Keywords: Learning curve, Primary total hip replacement, Anterior minimally invasive surgery, Survivorship Accepted: September 10, 2013
INTRODUCTION Despite increasing use of minimally invasive surgical (MIS) techniques for total hip replacement (THR), current literature fails to demonstrate more than only short term advantages and risks versus benefits are still an ongoing debate (1). Opponents associate MIS with additional risks such as infection, neurovascular injury, femoral fracture and component mal-positioning potentially resulting in increased implant failure rates (2, 3). Additionally, new implant designs without long-term documentation are introduced to adopt the needs of MIS techniques. One of the goals of minimally invasive THR is to minimise trauma to underlying muscles and tendons. Having been trained using the direct lateral approach (4), the senior author used this approach until 2004. Dissatisfaction regarding this
approach aroused due to active patients claiming residual trochanteric pain and limping when trying to restart physical activities beyond those of daily living. A study comparing magnetic resonance imaging (MRI) findings on tendons and muscles between active patients claiming residual pain and limping and asymptomatic patients after THR revealed a significant association between residual pain and soft tissue damage (5). This was the author’s rationale to look for a “less invasive” approach for THR. From an anatomical point of view the anterior approach to the hip (6) was expected to ideally fulfill the goal of minimising soft tissue trauma because of the inter-muscular and inter-nervous plane explored and because reports using this approach were promising (7). Thus, an anterior minimally invasive surgical (AMIS) approach using a leg holder and relatively new implants (Quadra®-H stem and Versafit®-CC cup, Medacta, Castel
© 2014 Wichtig Publishing - ISSN 1120-7000
277
AMIS THR: survivorship and learning curve
San Pietro, Switzerland) was introduced in our department in 2005. Since figures from the 2010 and 2011 annual report of the Australian implant registry on the same implants showed high one-year cumulative revision rates of 2.8% and 2.1%, respectively, a retrospective analysis of our patient cohort was undertaken. The aim was to document complications, quality of implant positioning and five years survivorship of the Quadra®-H stem and Versafit®-CC cup implanted by AMIS. Our hypothesis was that: 1) eventual high complication and revision rates would be limited to an early series; and 2) mistakes leading to high complication and revision rates within an early series would be avoided by juniors trained for AMIS by a senior surgeon.
MATERIAL AND METHODS
protecting the TFL muscle was performed and a soft tissue retractor aimed to be anchored within the joint capsule used. An offset reamer and cup impactor was used for acetabular preparation and cup placement. For femoral broaching the leg was positioned in about 20° to 30° of extension, 90° of external rotation and 20° of adduction using the leg positioner and straight Judet type broach handles without offset. All surgeries were performed under a laminar air flow system and three doses of cefuroxime 1.5 g in eight hourly intervals starting 30 minutes before incision were administered and enoxaparin natrium daily for six weeks were given to all patients. Weight bearing as tolerated on two crutches was recommended for two weeks; afterward full weight bearing was encouraged.
Data acquisition
All THRs implanted by AMIS technique performed between January 2005 and April 2006 were retrospectively included and recalled after a minimum follow-up of five years in April 2011. During this time period the senior surgeon used AMIS for every primary THR unless trochanteric advancement was deemed necessary to achieve true hip reconstruction or hardware removal required the use of a lateral or posterior approach. Trochanteric advancement was indicated in hips with sequelae of childhood hip disease having had prior surgery, which resulted in a highly distorted anatomy of the proximal femur. The senior author performed the first 22 AMIS before two junior surgeons were consecutively involved. A Quadra®-H stem and Versafit®-CC cup (Medacta, Castel San Pietro, Switzerland) together with a 28 mm CoCr ball and a highly cross-linked polyethylene inlay was used in all patients of this consecutive series.
Demographic parameters, peri-operative and late complications were recorded from electronic patient’s charts. Additionally, a standardised interview asking for all medico-surgical interventions concerning the affected hip performed at a follow-up visit appointed for the purpose of this study after a minimum of five years was performed. Anteroposterior pelvic x-rays acquired in the operation room immediately after surgery were analysed for; stem alignment, leg length discrepancy, cup inclination and version were measured (9) by one independent orthopedic surgeon (D.A.M.). The cumulative five years survival of the implants was estimated using Kaplan Meyer survival analysis with implant revision for any reason as endpoint. Subjective and clinical outcome was assessed using the WOMAC score (10), the subjective hip value (in accordance to the subjective shoulder value (11)) and the Harris Hip Score (12).
Surgical technique and treatment protocol
Statistics
A modified Hueter approach (6) in the supine position on a leg positioner (AMIS® mobile leg positioner, Medacta, Castel San Pietro, Switzerland) approximating Judet’s description (8) was performed as follows. In order to avoid lesions to the lateral cutaneous nerve of the thigh, the skin incision was centered on the top of the tensor fascia latae muscle (TFL) and its facial sheath entered before blunt dissection along its medial border. After ligature of the ascending branch of the lateral circumflex femoral artery a U-shaped capsulotomy aiming for a laterally based flap
A biostatistician performed all calculations. SPSS for Windows was used for all analysis. Distribution of continuous variables was assessed graphically. Normally distributed data were presented as means ± SD or mean and confidence intervals and non-parametric data with medians and ranges. Implant survival curves of different surgeons and of different chronological periods were compared using a Log rank test and Cox regression, respectively. The paired Wilcoxon signed ranks test was used to compare continuous non-parametric values at the time of diagnosis with
278
© 2014 Wichtig Publishing - ISSN 1120-7000
Müller et al
those at the time of follow-up. A Mann-Whitney U test was used to compare continuous non-parametric postoperative outcome parameters of different groups. Significance level was set at p
