Numerous approaches to minimally invasive total hip arthroplasty (THA) have developed over the decades, with the direct anterior and mini-posterior techniques being some of the most widespread.1–10 Other than a smaller scar, minimally invasive THA has demonstrated shorter hospital stays, less intraoperative blood loss, and faster recovery when compared with traditional techniques.10–12 However, the reduced visualization of the operative field has downsides, including increased operative time, poor component positioning, and increased technical difficulty resulting in, what may be the most concerning, a prolonged learning curve.13–16
The mini-posterior approach is currently one of the most widely used minimally invasive approaches, because it offers an optimal view of the acetabulum and femoral head but at the cost of releasing the short external rotators of the hip.17 The direct superior (DS) approach is a Food and Drug Administration–approved modification of the mini-posterior approach that spares the iliotibial band and several of the short external rotators, mainly the obturator externus tendon and quadratus femoris muscle.18,19 Both techniques have an incision length of approximately 8 to 12 cm, with the mini-posterior approach having the majority of the incision distal to the tip of the greater trochanter and the DS approach having the majority of the incision proximal to the tip of the greater trochanter.19,20 Although some preliminary studies show DS-THA has a low dislocation rate and limited inadvertent muscle damage, there are currently no studies regarding the learning curve associated with the DS approach.21,22 Learning curve studies are based on the concept that surgical skills take time to develop and surgeons require a certain number of repetitions to reach proficiency.23,24 Examining the learning curve of surgical procedures is important in making sure orthopedic trainees gain enough experience and in evaluating the opportunity costs associated with surgeons switching to newer techniques. Evaluating the learning curve can be done by examining sequential groups of patients operated on by a surgeon and quantifying the safety, clinical outcomes, and complication rate.25
The purpose of this study was to quantify the learning curve for primary DS-THA and compare it with a gold standard minimally invasive approach to THA—mini-posterior—with a short learning curve. The direct anterior approach, in which the surgeon dissects the internervous plane between the sartorius and tensor fascia lata muscles on a supine patient, is another popular technique with favorable outcomes.17,26 Previous studies on the direct anterior approach have suggested a steep curve, with 50 to 100 cases being needed before the rate of complications and revision surgeries stabilizes.27
Materials and Methods
This study was a retrospective case series evaluating the surgical outcomes and safety that establish the learning curve associated with the DS approach. The THA cases analyzed were from a single surgeon (D.F.A.) who completed a fellowship in joint replacement/adult reconstruction in 2014 and was primarily trained on the posterior approach. Ninety-day data were used, because this is becoming an increasingly common time point for evaluating complications and outcomes for bundled-payment models of reimbursements.28,29 This research received institutional review board approval. For the first year of practice, the surgeon conducted THAs using the mini-posterior approach, and in 2015 he conducted his first training experience on the DS approach. Starting in 2015, THAs were conducted using the DS approach, with the contraindications being the patient having a body mass index above 40 kg/m2 or the reason for surgery being a femoral neck fracture, in which case mini-posterior surgery was done instead. The DS cases presented in this study were the first 40 consecutive primary DS-THAs conducted between 2015 and 2017. The mini-posterior cases used as a comparison group of baseline performance were the first 20 consecutive cases following the completion of the fellowship in 2014.
The DS cohort was divided into 2 cohorts of 20 each to allow the authors to examine whether the learning curve exceeds 40 patients by retroactively comparing the safety, outcomes, and complications of both groups. To gain better insight into the safety of the DS approach during the early phases of learning, the authors compared the outcomes and safety of the first 20 DS cases with those of the first 20 mini-posterior cases (Table 1). Age and duration of follow-up were statistically significantly different (P=.001 and P=.013, respectively); however, the authors believe both do not represent clinically relevant differences. There were no major differences noted between the patient groups regarding medical comorbidities (Table 2).
Demographics and Diagnosis
Relevant Medical Comorbidities
All patients were ambulated within 24 hours of their surgery with the guidance of a physical therapist. Timed-up-and-go (TUG) trials were measured as the time required for a patient to get up from a standard armchair, walk 10 feet away and toward the chair, and sit back down. All patients had an intraoperative anterior-to-posterior pelvis radiograph and a standing anterior-to-posterior pelvis radiograph at 3-month follow-up. Acetabular component inclination and anteversion (Martell method), Harris Hip Score (HHS), estimated blood loss, operative time, length of hospital stay, and complications were determined for each patient. The Lewinnek safe zone was defined as 15°±10° for anteversion and 40°±10° for cup inclination.30 A titanium acetabular component and Accolade II femoral component (Stryker, Mahwah, New Jersey) with a 36-mm ceramic head and highly cross-linked X3 polyethylene were used in each case. All of the patients received 1 g of tranexamic acid intravenously on incision and during closure. None of the patients received a postoperative blood transfusion. All of the surgeries were conducted at the same institution, and patients were sent to the same arthroplasty hospital unit for recovery.
Continuous variables were displayed as mean and standard deviation, whereas categorical variables were presented as number and percentage. Statistical analyses were conducted using an analysis of variance or t test. Chi-square or Fisher's exact tests were used for categorical data. Differences were considered significant at P<.05.
When directly comparing the first and second sets of DS patients (Table 3), there were no statistically significant differences in intraoperative blood loss, length of hospital stay, distance ambulated at discharge, TUG result, discharge location, number of patients receiving narcotics, or mean HHS at 90-day follow-up (P>.05). The only statistically significant difference was the mean operative time, which decreased from 130±24 to 104±20 minutes, respectively (P<.001). There were 5 cases (25%) in the first DS group in which the incision was superficially extended for better positioning of the reamer in abduction; no additional quadratus femoris muscle was released. Only 1 (5%) case in the second DS group had the incision extended. There was a trend toward decreasing frequency of incisions extension (P=.077) over the learning curve, but there was no difference in final incision length between the first (10.0±1.9 cm) and second (9.2±1.2 cm; P=.131).
Operative Parameters and Outcomes at 90 Days Postoperatively
The first 20 DS and mini-posterior cases (Table 3) had no statistically significant differences in mean operative time, intraoperative blood loss, TUG results, discharge location, number of patients receiving narcotics, or mean HHS at 90-day follow-up (P>.05). The mean length of stay for the initial DS-THAs (48±16 hours) was shorter and more consistent when compared with the initial mini-posterior THAs (72±33 hours; P=.007). The mean distance ambulated at discharge in the initial DS-THAs (111±111 ft) was greater than the initial mini-posterior THAs (34±33 ft; P=.007). Mini-posterior patients (12.9±1.8 cm) had incisions that were approximately 3 cm longer than the DS patients (P<.001). There were no episodes of acute hip instability, periprosthetic fracture, peri-prosthetic joint infection, or other postoperative surgical complications at 90 days for any of the patients.
The distribution of femoral head sizes for all patients are presented in Table 4. A chi-square test comparing head sizes 32 mm or smaller and 36 mm or larger for the 3 groups was not statistically significant (P=.344).
Distribution of Femoral Head Sizes
An important element in evaluating component position and potential risk for dislocations is measuring acetabular component inclination and anteversion. When examining the Lewinnek plot, there were no cases outside of the safe zone for any DS patients, whereas there were 2 (10%) outside of the safe zone for the mini-posterior technique (Figure 1). The mean inclination for the first 20 DS hips (39°±6°) was not significantly different from either the second 20 DS hips (39°±5°; P=.827) or the mini-posterior hips (38°±6°; P=.817). The mean acetabular anteversion for the first 20 DS hips (12°±3°) was not significantly different from the second 20 DS hips (13°±4°; P=.394). The mean acetabular anteversion was found to be statistically different between the first set of DS (12°±3°) and mini-posterior hips (16°±4°; P=.006), with mini-posterior hips tending to have a slightly more acetabular anteversion (Figure 2).
Lewinnek plot showing inclination and anteversion of each acetabular component in the series (•, direct superior cases 1–20; +, direct superior cases 21–40; ∆, mini-posterior cases 1–20).
Mean acetabular component inclination (A) and anteversion (B) as measured from postoperative anterior-to-posterior pelvis radiographs (DS 1–20, direct superior cases 1–20; DS 21–40, direct superior cases 21–40; MP 1–20, mini-posterior cases 1–20). *P<.05.
Examining the DS demographics, it is apparent that patient body mass index, comorbidities, sex, hip operation side, and diagnoses were similar. The second set of DS patients tended to be younger, but this statistically significant variation was not found to be clinically relevant. Based on the operative data and the similarity in patient demographics, it can be argued that the DS learning curve is short, perhaps fewer than 20 cases. The absence of any statistically significant differences for DS patients in terms of length of hospital stay, estimated blood loss, HHS, and presence of perioperative complications supports this finding. If the authors were expecting a learning curve similar to that of the direct anterior approach, shown by de Steiger et al27 to be approximately 50 to 100 cases, then the second set of DS patients should have had statistically significant differences in the perioperative parameters and complications, indicating room for improvement. It is clear, however, that experience was gained over time as surgical time decreased by a mean of 24 minutes. Gaining a better understanding of exactly when operative time equilibrates would require tracking a greater number of cases. Fewer cases required incision extension for reamer positioning in the second set of DS patients. Although not statistically significant, this trend does indicate an improvement in surgical technique over time.
One of the potential limitations cited for minimally invasive THA is poor acetabular component positioning.14 Although acetabular component positioning alone is not entirely predictive of dislocations and long-term complications, it is a strong factor in determining risk of hip instability. The values for acetabular component inclination and anteversion for the second set of DS cases were within 1° of the first set, with all values falling within the Lewinnek safe zone. This indicates that the minimally invasive nature of the DS approach does not compromise the anatomical positioning of the acetabular component and that the learning curve is likely less than 20 patients. Although 90-day data are not sufficient in completely evaluating the safety profile of the DS approach given the long-term prevalence of hip instability, 58% of hip dislocations occurred within 3 months of THA, making the data presented here encouraging.31 However, 90-day follow-up is not comprehensive. Longer term follow-up on DS-THA is required, especially related to instability and infection.
Although comparing patients within the DS group is useful in observing the relative trajectory of outcomes as the surgeon gains experience, it does not offer answers on the absolute performance of the DS approach as a minimally invasive technique. The only major differences between the DS and mini-posterior approach were in length of hospital stay and distance ambulated at discharge. The DS patients left the hospital 27 hours earlier and ambulated more than twice the mean distance as the mini-posterior patients. Both of these results are encouraging examples of the potential advantages of the DS approach as a muscle-sparing technique being more in line with the potential benefits of minimally invasive THA. Mean acetabular component inclinations of the first 20 DS and mini-posterior cases were within 1° with no statistically significant differences. However, mini-posterior patients tended to have more acetabular anteversion, a difference that could be attributed to the anatomical access plane into the hip joint. This difference did not affect the ability of the surgeon to place the component within the Lewinnek safe zone and did not increase the instability rate in this small retrospective series. Two (10%) of the acetabular components from the mini-posterior cases fell outside the Lewinnek safe zone, as compared with none (0%) for the DS cases.
Data at 90 days are becoming increasingly relevant to joint replacement surgeons as bundled-payment models become a more popular form of reimbursement. Although 90-day data are not sufficient to fully evaluate hip reconstructive surgeries, they are currently being heavily relied on for assessing health care providers and adjusting their compensation.28,29 The data presented therefore provide insight into how the DS approach would perform under such reimbursement plans but are not comprehensive, and longer term follow-up is required.
Because the duration of the learning curve is a major part of minimally invasive THA, it is vital for arthroplasty surgeons to evaluate the number of cases required to reach proficiency. The results of this study indicate that the learning curve associated with the DS approach appears to be short. There were no significant differences in perioperative outcome parameters or complications with DS-THA. The data do suggest, however, that DS operative times decline after the first 20 patients. Future studies evaluating more cases are warranted to evaluate if operative time will continue to decrease as surgical volume increases. There were also no striking differences between the first 20 DS and mini-posterior patients regarding perioperative outcome parameters or complications. However, the mean length of hospital stay was shorter and the mean distance ambulated was longer for the DS patients when compared with the mini-posterior patients. All of the patients had the surgeries conducted at the same institution within a 3-year period, which minimizes variability in postsurgical management.
Although this study shows that the DS approach had a more favorable learning curve when compared with the historical learning curve of the direct anterior approach, it is based on the experience of a single surgeon. Moreover, the surgeon learned the DS technique after practicing the posterior approach for approximately 1 year. Although this skews the data favorably toward the DS approach, it offers insight into the learning curve of DSTHA for surgeons who already practice with the posterior approach. Because the posterior approach is the most common worldwide, DS-THA could offer an alternative to the direct anterior approach for a large number of surgeons who might be considering minimally invasive THA but do not want to endure a prolonged learning curve.
This study has provided insight into the DS-THA learning curve, especially for surgeons who already practice the posterior approach. At 90-day follow-up, DS-THA appears to be safe and outcomes appear to be comparable to those of the mini-posterior approach. The lack of major differences in operative parameters, complications, and patient outcomes in both DS groups indicates that the learning curve is less than 20 patients. However, more long-term data are required to reach definitive conclusions on the complications, dislocations, and outcome scores to fully evaluate the safety of this procedure.
- Amlie E, Havelin LI, Furnes O, et al. Worse patient-reported outcome after lateral approach than after anterior and posterolateral approach in primary hip arthroplasty: a cross-sectional questionnaire study of 1,476 patients 1–3 years after surgery. Acta Orthop. 2014;85(5):463–469. doi:10.3109/17453674.2014.934183 [CrossRef] PMID:24954494
- Browne JA, Pagnano MW. Surgical technique: a simple soft-tissue-only repair of the capsule and external rotators in posterior-approach THA. Clin Orthop Relat Res. 2012;470(2):511–515. doi:10.1007/s11999-011-2113-6 [CrossRef] PMID:21989783
- Chechik O, Khashan M, Lador R, Salai M, Amar E. Surgical approach and prosthesis fixation in hip arthroplasty world wide. Arch Orthop Trauma Surg. 2013;133(11): 1595–1600. doi:10.1007/s00402-013-1828-0 [CrossRef] PMID:23912418
- Chimento GF, Pavone V, Sharrock N, Kahn B, Cahill J, Sculco TP. Minimally invasive total hip arthroplasty: a prospective randomized study. J Arthroplasty. 2005;20(2):139–144. doi:10.1016/j.arth.2004.09.061 [CrossRef] PMID:15902851
- Goldstein WM, Branson JJ, Berland KA, Gordon AC. Minimal-incision total hip arthroplasty. J Bone Joint Surg Am. 2003;85(suppl 4):33–38. doi:10.2106/00004623-200300004-00004 [CrossRef] PMID:14652391
- Hananouchi T, Takao M, Nishii T, et al. Comparison of navigation accuracy in THA between the mini-anterior and -posterior approaches. Int J Med Robot. 2009;5(1):20–25. doi:10.1002/rcs.226 [CrossRef] PMID:19107818
- Martin CT, Pugely AJ, Gao Y, Clark CR. A comparison of hospital length of stay and short-term morbidity between the anterior and the posterior approaches to total hip arthroplasty. J Arthroplasty. 2013;28(5):849–854. doi:10.1016/j.arth.2012.10.029 [CrossRef] PMID:23489731
- Nakamura S, Matsuda K, Arai N, Wakimoto N, Matsushita T. Mini-incision posterior approach for total hip arthroplasty. Int Orthop. 2004;28(4):214–217. doi:10.1007/s00264-004-0570-1 [CrossRef] PMID:15168084
- Nakata K, Nishikawa M, Yamamoto K, Hirota S, Yoshikawa H. A clinical comparative study of the direct anterior with mini-posterior approach: two consecutive series. J Arthroplasty. 2009;24(5):698–704. doi:10.1016/j.arth.2008.04.012 [CrossRef] PMID:18555653
- Wright JM, Crockett HC, Delgado S, Lyman S, Madsen M, Sculco TP. Mini-incision for total hip arthroplasty: a prospective, controlled investigation with 5-year follow-up evaluation. J Arthroplasty. 2004;19(5):538–545. doi:10.1016/j.arth.2003.12.070 [CrossRef] PMID:15284972
- Berger RA, Jacobs JJ, Meneghini RM, Della Valle C, Paprosky W, Rosenberg AG. Rapid rehabilitation and recovery with minimally invasive total hip arthroplasty. Clin Orthop Relat Res. 2004;429:239–247. doi:10.1097/01.blo.0000150127.80647.80 [CrossRef] PMID:15577494
- DiGioia AM III, Plakseychuk AY, Levison TJ, Jaramaz B. Mini-incision technique for total hip arthroplasty with navigation. J Arthroplasty. 2003;18(2):123–128. doi:10.1054/arth.2003.50025 [CrossRef] PMID:12629599
- Berry DJ, Berger RA, Callaghan JJ, et al. Minimally invasive total hip arthroplasty: development, early results, and a critical analysis. Presented at the Annual Meeting of the American Orthopaedic Association, Charleston, South Carolina, USA, June 14, 2003. J Bone Joint Surg Am. 2003;85(11):2235–2246. doi:10.2106/00004623-200311000-00029 [CrossRef] PMID:14630860
- Teet JS, Skinner HB, Khoury L. The effect of the “mini” incision in total hip arthroplasty on component position. J Arthroplasty. 2006;21(4):503–507. doi:10.1016/j.arth.2005.06.011 [CrossRef] PMID:16781401
- Woolson ST, Mow CS, Syquia JF, Lannin JV, Schurman DJ. Comparison of primary total hip replacements performed with a standard incision or a mini-incision. J Bone Joint Surg Am. 2004;86(7):1353–1358. doi:10.2106/00004623-200407000-00001 [CrossRef] PMID:15252080
- Ogonda L, Wilson R, Archbold P, et al. A minimal-incision technique in total hip arthroplasty does not improve early postoperative outcomes: a prospective, randomized, controlled trial. J Bone Joint Surg Am. 2005;87(4):701–710. doi:10.2106/00004623-200504000-00002 [CrossRef] PMID:15805196
- Palan J, Manktelow A. Surgical approaches for primary total hip replacement. Orthop Trauma. 2018;32(1):1–12. doi:10.1016/j.mporth.2017.11.003 [CrossRef]
- Penenberg BL, Bolling WS, Riley M. Percutaneously assisted total hip arthroplasty (PATH): a preliminary report. J Bone Joint Surg Am. 2008;90(suppl 4):209–220. doi:10.2106/JBJS.H.00673 [CrossRef] PMID:18984733
- Roger DJ, Hill D. Minimally invasive total hip arthroplasty using a transpiriformis approach: a preliminary report. Clin Orthop Relat Res. 2012;470(8):2227–2234. doi:10.1007/s11999-011-2225-z [CrossRef] PMID:22215476
- American Academy of Orthopaedic Surgeons. Mini-posterior approach. https://www.aaos.org/CustomTemplates/VideoGallery.aspx?id=6442454302&nav=1050&ssopc=1. Accessed March 27, 2019.
- Bono OJ, Damsgaard C, Robbins C, Aghazadeh M, Talmo CT, Bono JV. Influence of soft tissue preservation in total hip arthroplasty: a 16-year experience. Surg Technol Int. 2018;33:301–307. PMID:29985520
- Amanatullah DF, Masini MA, Roger DJ, Pagnano MW. Greater inadvertent muscle damage in direct anterior approach when compared with the direct superior approach for total hip arthroplasty. Bone Joint J. 2016;98-B(8):1036–1042. doi:10.1302/0301-620X.98B8.37178 [CrossRef] PMID:27482014
- Stone AH, Sibia US, Atkinson R, Turner TR, King PJ. Evaluation of the learning curve when transitioning from postero-lateral to direct anterior hip arthroplasty: a consecutive series of 1000 cases. J Arthroplasty. 2018;33(8):2530–2534. doi:10.1016/j.arth.2018.02.086 [CrossRef] PMID:29622494
- Archibeck MJ, White RE Jr, . Learning curve for the two-incision total hip replacement. Clin Orthop Relat Res. 2004;429:232–238. doi:10.1097/01.blo.0000150272.75831.2f [CrossRef] PMID:15577493
- Hopper AN, Jamison MH, Lewis WG. Learning curves in surgical practice. Postgrad Med J. 2007;83(986):777–779. doi:10.1136/pgmj.2007.057190 [CrossRef] PMID:18057179
- Barry JJ, Masonis JL, Mason JB. Recovery and outcomes of direct anterior approach total hip arthroplasty. Ann Joint. 2018;3:51. doi:10.21037/aoj.2018.04.09 [CrossRef]
- de Steiger RN, Lorimer M, Solomon M. What is the learning curve for the anterior approach for total hip arthroplasty?Clin Orthop Relat Res. 2015;473(12):3860–3866. doi:10.1007/s11999-015-4565-6 [CrossRef] PMID:26394641
- Iorio R, Clair AJ, Inneh IA, Slover JD, Bosco JA, Zuckerman JD. Early results of Medicare's bundled payment initiative for a 90-day total joint arthroplasty episode of care. J Arthroplasty. 2016;31(2):343–350. doi:10.1016/j.arth.2015.09.004 [CrossRef] PMID:26427938
- Williams J, Kester BS, Bosco JA, Slover JD, Iorio R, Schwarzkopf R. The association between hospital length of stay and 90-day readmission risk within a total joint arthroplasty bundled payment initiative. J Arthroplasty. 2017;32(3):714–718. doi:10.1016/j.arth.2016.09.005 [CrossRef] PMID:27776899
- Lewinnek GE, Lewis JL, Tarr R, Compere CL, Zimmerman JR. Dislocations after total hip-replacement arthroplasties. J Bone Joint Surg Am. 1978;60(2):217–220. doi:10.2106/00004623-197860020-00014 [CrossRef] PMID:641088
- Yuan L, Shih C. Dislocation after total hip arthroplasty. Arch Orthop Trauma Surg. 1999;119(5–6):263–266. doi:10.1007/s004020050406 [CrossRef] PMID:10447619
Demographics and Diagnosis
|Parameter||Direct Superior (1–20)||Direct Superior (21–40)||Mini-Posterior (1–20)||Pa|
|Age, mean±SD, y||58±10||51±12||64±9||.001b|
|Body mass index, mean±SD, kg/m2||28±5||26±5||29±5||.137|
|Sex, male/female, %||65/35||50/50||65/35||.535|
|Hip, right/left, %||70/30||60/40||55/45||.610|
|Preoperative diagnosis, No.|
| Osteoarthritis||20 (100%)||20 (100%)||17 (90%)|
| Avascular necrosis||0 (0%)||0 (0%)||2 (10%)||NA|
| Femoral neck fracture||0 (0%)||0 (0%)||1 (5%)|
| Follow-up, mean±SD, d||108±28||94±11||91±13||.013b|
Relevant Medical Comorbidities
|Direct Superior (1–20)||Direct Superior (21–40)||Mini-Posterior (1–20)|
|Hypertension||9 (45%)||5 (25%)||9 (45%)||.324|
|Hypothyroidism||NA||4 (20%)||4 (20%)||1.000|
|Diabetes mellitus||2 (10%)||NA||1 (5%)||.548|
|Hyperlipidemia||5 (25%)||3 (15%)||4 (20%)||.732|
|Chronic obstructive pulmonary disease/asthma||1 (5%)||1 (5%)||1 (5%)||1.000|
|Coronary artery disease||3 (15%)||NA||2 (10%)||.633|
|Deep venous thrombosis||0 (0%)||2 (10%)||0 (0%)||NA|
|Autoimmune disorder||0 (0%)||4 (20%)||0 (0%)||NA|
Operative Parameters and Outcomes at 90 Days Postoperatively
|Parameter/Outcome||Direct Superior (1–20)||Direct Superior (21–40)||Mini-Posterior (1–20)||P DS/DS||P DS/MP|
|Discharge location, No.|
| Home||19 (95%)||19 (95%)||17 (95%)||1.000||.292|
| Skilled nursing facility||1 (5%)||1 (5%)||3 (5%)|
|Operative time, mean±SD, min||130±24||104±20||141±29||<.001a||.227|
|Estimated blood loss, mean±SD, mL||182±86||156±29||268±143||.219||.033|
|Length of stay, mean±SD, h||48±16||45±17||72±33||.492||.007b|
|Distance ambulated at discharge, mean±SD, ft||111±111||80±69||34±33||.299||.007b|
|TUG, mean±SD, s||10±4||11±4||12±4||.393||.139|
|Receiving narcotics, No.||3 (15%)||2 (10%)||1 (5%)||.633||.548|
| Intraoperative extension, No.||5||1||0||.077||NA|
| Incision length at 90 days, mean±SD, cm||10.0±1.9||9.2±1.2||12.9±1.8||.131||<.001|
Distribution of Femoral Head Sizes
|Group||Femoral Head Size, No.|
|32 mm||36 mm|
|Direct superior, 1–20||3 (15%)||17 (85%)|
|Direct superior, 21–40||5 (25%)||15 (75%)|
|Mini-posterior, 1–20||7 (35%)||13 (65%)|