- Email: [email protected]
Learning Curves in Arthroplasty in Orthopedic Trainees
ORIGINAL REPORTS
Learning Curves in Arthroplasty in Orthopedic Trainees Obinna Nzeako, MB, ChB, and Diane Back, MBBS, BSc Department of Orthopaedics, St Thomas’ Hospital, London, UK The NHS is adapting to a changing environment, in which economical constraints have forced theatres to maximise efficiency. An environment in which working hours and surgical exposure has been reduced and outcomes are being published. Litigation is high, and patients are living longer with higher demands. We ask, will traditional methods of apprentiship type training suffice in producing competent arthroplasty surgeons when hands on experience is falling. We review learning curves and assessment tools available to accurately assess competency and support trainee orthopaedic surgeons in their acquisition of surgical proficiency. C 2016 Published ( J Surg Ed ]:]]]-]]]. Crown Copyright J by Elsevier Inc. on behalf of the Association of Program Directors in Surgery. All rights reserved. ) KEY WORDS: hip replacement, knee replacement, arthroplasty, learning curve, trainee COMPETENCIES: Medical Knowledge, Practice Based
Learning and Improvement, Patient Care
INTRODUCTION As the National Health Service adapts to a changing world, where people are living longer with higher expectations; but where economical constraints have forced hospitals to reexamine processes to maximize efficiency—much attention is on orthopedic training. The world health organization projects that by 2020 osteoarthritis would be the fourth leading cause of disability.1 Unsurprisingly, the number of performed total hip replacements (THRs) and total knee replacements (TKRs) are placing increasing pressures on hospitals, in their challenge to manage waiting times.2 Thus, consultant orthopedic surgeons are increasingly finding themselves torn between achieving targets and providing adequate training and sufficient load exposure to produce competent surgeons. Traditionally, orthopedic surgeons undertook an apprenticeship type training program, in which they would start by Correspondence: Inquiries to Obinna Nzeako, MB, ChB, Department of Orthopaedics, St Thomas’ Hospital, Westminster Bridge Rd, London SE1 7EH, UK; e-mail: [email protected]
assisting, then would be observed and supervised until they were competent to perform arthroplasty surgery unsupervised. It was assumed that the trainee would become competent in hip and knee arthroplasty by completion of a predetermined length of training. The concern with this model of training is that the patient may not receive the highest level of surgery from a relatively inexperienced surgeon, which may translate to poorer clinical outcomes. This has become even more relevant with the data on individual consultant outcomes now being published. Naturally, the result would be more consultant performed procedures, and less hands-on experience during specialty training, especially during the early years. Together with the implementation of the European working time directive,3 National inquiry into Patient Outcomes and Deaths,4 Modernizing Medical Careers (MMC), 5 and the Joint Committee in Surgical Training surgical education in the United Kingdom has been revisited. Similar changes have been reported in the United States, and a recent study has revealed that surgical trainees report having very few supervised operative experiences in basic procedures deemed essential by their program directors.6 With the introduction of the working time directive, the reduction in out of hours operating, the development of MMC, and development of the new specialist training programs (ST 1-7) there is marked concern that surgeons are entering consultant positions with significantly less experience than their predecessors. Analysis of log books from 2004 suggested that MMC reduced the number of primary THRs and TKRs performed by trainees from 37 and 44 to 22 and 24, respectively.7 A further study published in Ann R Coll Surg Eng showed that SpRs performed almost 35% less TKRs after the Independent Sector Treatment Centres were operational.8 With training moving toward a competency based framework and a modern orthopedic curriculum that involves significantly less hands-on experience, a better understanding of learning, and skill acquisition is essential. What is more there a need to fully use alternative techniques to aid in learning and assessment to ensure these competencies are truly achieved.
Journal of Surgical Education Crown Copyright & 2016 Published by Elsevier Inc. on behalf of the Association of Program Directors in Surgery. All rights reserved. 1931-7204/$30.00 http://dx.doi.org/10.1016/j.jsurg.2016.02.006
1
Simulation appears to be emerging as a useful tool, but has been critized for its poor transferability to the operative setting. This article reviews current available evidence on learning curves of junior orthopedic surgeons embarking on hip and knee arthroplasty surgery. We would clarify the learning curve and review its reported effect on surgical outcomes of orthopedic trainees when compared with consultant surgeons. We would also review tools and methods of assessing competency.
UNDERSTANDING THE LEARNING CURVE When learning a new procedure, performance usually improves with experience. Graphically plotting performance against experience produces a learning curve. Initially, the gradient of ascent indicates how quickly the clinician is improving. As the degree of improvement attained with each case reduces as technique is refined, overall performance begins to tail off. A point is then reached when the individual is able to perform the procedure independently and still achieve the same good outcomes. Further experience still improves performance by small amounts until a plateau is reached.9 In surgery, there are potentially dramatic implications of misjudging a surgeon or units position on the learning curve. Recent examples include the Oxford Paediatric Cardiac death inquiry and the General Medical Council inquiry into the Bristol Paediatric Surgical Unit, who stated that “patients should not be exposed to Surgeons at the start of their learning curves.”10 Measures of learning curves in arthroplasty can be broadly divided into measures of the surgical process and measures of patient outcomes, for example, length of stay, transfusion requirements, functional outcomes, and morbidity An extensive review of the literature revealed a very few published data specifically plotting the learning curves of trainee surgeons in arthroplasty surgery. There is also little evidence to support the use of one of these outcome measures over another in the assessment of learning. Within training, the challenge is providing enough time and exposure to facilitate universal achievement of competency. A variety of teaching and assessment strategies must be employed to support this process. As supervised training opportunities dwindle newly appointed consultants must recognize their position on the learning curve and the need for further training, structured appraisal and senior mentorship.
THE LEARNING CURVE AND CLINICAL OUTCOMES Several recent studies have compared clinical and radiographic outcomes of THRs and TKRs performed either by a consultant or a trainee (supervised and unsupervised). A commonly cited prospective randomized trial of THRs 2
using the Stanmore and Charnley prostheses showed that unsupervised trainees had higher revision rates than those completed by an orthopedic consultant.11 In a study of 2906 hip hemiarthroplasties performed by senior registrars, junior registrars, or senior house officers (SHO), Unwin and Thomas12 found higher dislocation rates for junior registrars, and highest in SHOs. However, there is good recent evidence that supervised trainees can achieve similar surgical results when compared with consultant orthopedic surgeons. Mahaluxmivala et al.13 looked at 673 press fit condylar posterior cruciate sacrificing TKRs. They assessed surgeon experience against radiographic parameters for component positioning. They concluded that there was no difference between supervised trainees and consultants. In addition, Moran et al. reported no difference in the Harris Hip scores for THRs performed by specialist registrars vs. consultant at up to 18 months of follow-up.14 A multicentre prospective study of 1501 THRs performed either by a specialist registrar or consultant published in the JBJS(Br) concluded that there was no difference in functional outcome scores at up to 5 years follow-up.15 They also reported no difference in complications including dislocation rates. They did, however, report a statistically significant difference (p o 0.001) in operating time between the 2 groups, with a mean difference of 28 minutes (shorter duration in consultant group), and the length of hospital stay (p ¼ 0.22), 9.8 days in the consultant group, and 11 days in the registrar group. Interestingly, they also divided the registrar group into senior and junior registrars, and reported an Oxford hip score of 25.2 at 5 years follow-up in the senior registrar group that was significantly higher than (p ¼ 0.001) than that in the junior registrar group (Oxford hip score ¼ 21.8). However, there was no difference in complication or revision rate.15 These findings are reassuring to the public and to trainers. Supervised trainees are consistently found to be achieving comparable clinical results to consultants when performing arthroplasty surgery This suggests that the learning curve of trainee surgeons should not compromise the quality of the surgery provided. There has previously been some concern regarding the greater length of time required for trainee surgeons. A total of 31 Norwegian study based on 745 THRs showed that the mean operating time for cemented hip arthroplasty was 96 minutes.16 Operating times greater than 96 minutes were associated with greater revision rates for aseptic loosening, and those greater than 150 minutes were associated with increased rates of infection. However, this was not replicated by Palan et al.,15 a multicentre prospective study of 1501 THAs. What is clear is the key is, supervision appropriate to the traineeʼs level of experience. Older studies have shown greater complication rates and inferior clinical outcomes when joint replacement surgery is performed by unsupervised trainees still on the learning curve.12,17 There is little consensus on how many times a particular procedure must be performed to achieve competency for Journal of Surgical Education Volume ]/Number ] ] ]]]]
independent practice. A more focused training program is crucial. We must better understand the learning process and utilize available resources to maximize on training experiences.
ORTHOPEDIC TRAINING AND THE FUTURE Modern UK orthopedic training is based on the achievement of multiple core-competencies set out by OCAP.18 There are, however, few proven assessment tools specific to Arthroplasty. OCAP necessitates the use of work based assessments (WBAs) to evaluate trainee performance. In arthroplasty, procedure based assessments are used to assess surgical skill against a predetermined set of criteria.19 This allows for immediate formative feedback and facilitates the assessment of competency within a clinical setting. This forces the trainee to reflect on each stage of the procedure. These have, however, received criticism in recent times. Many trainers and trainees have felt them to be time consuming tick box exercises. The usefulness of each WBA is largely dependent on the willingness of the trainer and trainee to fully engage with the process. Within Europe and Canada another tool uses video recordings of the surgical procedure to objectively analyze the learning curve and evaluate surgical performance.20 This is referred to as time action analysis (TAA). The whole surgical procedure is divided into the following 5 different goal oriented phases: the incision phase, the femoral phase, the acetabulum phase, the stem phase, and the closure phase. Each goal oriented phases is subdivided in goal oriented actions and each goal oriented actions is subdivided in separate actions thereby defining all the necessary actions to complete the procedure.21 By analyzing the number and duration of the actions needed for a surgeon to achieve his or her goal, and the efficiency of each action, individual learning curves can be plotted. Not only does it allow assessment of competency but also objective comparison of multiple surgeons and trainees. By comparing experienced and inexperienced surgeons, it is possible to identify particularly difficult steps within a procedure that need further consideration during the learning curve.21 This tool has been well reported in the assessment of learning curves in laparoscopic procedures and shoulder arthroplasty.22,20,23 The learning curve cumulative sums method (CUSUM) is another tool developed from the CUSUM used for the long-term assessment of surgeon performance.24 Recently, it has been increasingly used to evaluate the learning curves in several medical procedures. Each time a procedure is performed it is scored against a predefined set of criteria. A method of statistical process control is employed to detect the adequacy of performance and determine when proficiency has been achieved. Lee et al.,25 used LC-CUSUM to show that there is a substantial learning period necessary in the optimal positioning of an acetabular cup during THR. Journal of Surgical Education Volume ]/Number ] ] ]]]]
The CUSUM technique too has received criticism for occasionally generating false-negatives; classifying adequate procedures as inadequate.24 Assessment and support of individual learning curves have not traditionally been a part of UK specialty training. However, with the development of these tools, the changing hospital dynamics and the reduction in individual operating numbers there is a recognized need for the inclusion of more comprehensive strategies.
SIMULATION AND COMPUTER NAVIGATION Simulation is quickly becoming an integral part of modern training. It provides conceptual scaffolding, promotes collaborative learning and reinforces motor learning, all within a safe nonclinical environment. There is an abundance of evidence showing that simulation can enhance technical ability and decision making in trainees in other specialties such as general surgery, anesthetics and obstetrics, and gynecology.26-31 The benefits have been well described for laparoscopy and arthroscopy, however, there is a lack of evidence in the literature detailing the advances of simulation technology in the training of open surgical skills. Some authors showed that after using computer navigation in the operating department, even experienced surgeons demonstrated improved accuracy in freehand component positioning.32 Given its potential to improve understanding of the operative field and the individual steps, it has been assumed to be a valuable educational resource. Similarly, other authors have shown that computer navigation technology can reduce the length of the learning curve in hip resurfacing.33 Perhaps, unsurprisingly within the United States it is being increasingly utilized with surgical trainees.34 There has, however, been a concern of the development of operator dependence on feedback from navigation technology with prolonged use. Motor learning studies have repeatedly demonstrated that extrinsic feedback (e.g., from a trainer or navigation technology) needs to be appropriately diminished over time to avoid the development of learner dependence. As a result its role within orthopedic training internationally is yet to be clarified. In a randomized study, looking at the effect of computer navigation on trainee learning of surgical skills Gofton et al.34 concluded that computer assisted orthopedic surgery resulted in improved early performance and equivalent learning. Each Specialist Advisory Committee has appointed a lead for simulation curriculum development. There are, however, logistical challenges to national expansion, such as funding for purchasing and maintaining the technology, availability of facilities for delivering training, manpower for administration 3
and training, and providing equal availability throughout the country. Despite this, many feel that it is in the best interest of the public to make this investment. There is evidence to suggest that time spent on simulators can reduce learning curves. While simulation in arthroscopy is becoming commonplace, there remains a paucity of information regarding simulation technology in arthroplasty.
4. Burke M, Callum KG, Gray AJG, et al. The 2001
report of the national confidential enquiry into perioperative deaths. London; 2001. 5. Kang S-N, Sanghrajka A, Amin A, Lee J, Briggs T.
Modernising medical careers: orthopaedic traineesʼ perspectives. Ann R Coll Surg Engl. 2005;87(suppl): S310-S312. 6. Bell RH, Biester TW, Tabuenca A. Operative experi-
CONCLUSION In 1979 Luft et al.35 reported a positive relationship between clinical outcomes and volume of procedures. This has been supported by several other studies, but despite this there is a lack of high-level data available to make an evidenced-based decision on number of procedures required to achieve competency. Sound systems for the recording and analysis of learning curves have emerged. These include LC-CUSUM, TAA, and perhaps even WBAs. The great benefit is the collection of numerical data and statistical analysis would allow for comparison of individuals, units, and educational establishments. A small step toward greater understanding of learning within the United Kingdom, could be to have trainees collect and keep their own data pertaining to their learning curve. Published evidence indicates that trainees are able to achieve good clinical outcomes with no difference in complication rates when supervised appropriately. However, there is a consistently reported longer duration of surgery, which is to be expected, and has obvious economic repercussions. However, despite this, if we do not make the necessary adjustments to accommodate for training and the learning curve, how would we produce proficient and skilled orthopedic surgeons? What certain is, more research is required to identify accurate methods of assessing and supporting ones learning curve. Perhaps this could be used to determine competency rather than obligatory numbers for each procedure. For this it seems technology may hold the key, in the form of simulation, computer navigation, or TAA.
REFERENCES
7. Sher JL, Reed MR, Calvert P, Wallace WA, Lamb A.
Influencing the national training agenda: the UK and Ireland orthopaedic E-logbook. J Bone Joint Surg Br. 2005;87(9):1182-1186. 8. Clamp J, Baiju D, Copas D, Hutchinson J, Rowles J.
Do independent sector treatment centres (ISTC) impact on specialist registrar training in primary hip and knee arthroplasty? Ann R Coll Surg Engl. 2008;90 (6):492-496. 9. Hopper A, Jamison M, Lewis G. Learning curves in
surgical practice. Postgrad Med J. 2007;83(986): 777-779. 10. Jackson.B. The Bristol enquiry report. Available at: 〈www.
bristol-inquiry.org.uk/final_report/annex_a/chapter_14_ 15.htm〉 Accessed 15.03.16. 11. Marston RA, Cobb AG, Bentley G. Stanmore com-
pared with Charnley total hip replacement: a prospective study of 413 arthroplasties. J Bone Joint Surg Br. 1996;78(2):178-184. 12. Unwin AJ, Thomas M. Dislocation after hemiarthro-
plasty of the hip: a comparison of the dislocation rate after posterior and lateral approaches to the hip. R Coll Surg Eng. 1994;76(5):327-329. 13. Mahaluxmivala J, Bankes MJ, Nicolai P, et al. The
effect of surgeon experience on component positioning in 673 pres fit condylar posterior cruciate sacrificing total knee arthroplasties. J Arthroplasty. 2001;16 (5):635-640.
1. Wadey V, Maloney W, Dev P, Walker D. Orthopae-
14. Moran M, Yap SL, Walmsley P, Brenkel IJ. Clinical
dic surgery core curriculum hip and knee reconstruction. Can J Surg. 2008;51(2):135-141.
and radiographic outcome of of total hip arthroplasty performed by trainee compared with consultant orthopaedic surgeons. J Arthroplasty. 2004;19(7):853-857.
2. Frankel S, Eachus J, Pearson N. Population require-
ment for primary hip replacement surgery: a cross sectional study. Lancet. 1999;353(9161):1304-1309. 3. Lim E, Tsui S. Impact of the European working time
directive on exposure to operative cardiac surgical training. Eur J Cardiothorac Surg. 2006;30(4): 574-577. 4
ence of residents in US general surgery programs: a gap between expectation and experience. Ann Surg. 2009;249(5):719-724.
15. Palan J, Gulati A, Andrew J, Murray DW, Beard DJ.
The trainer, the trainee and the surgeons assistant— clinical outcomes following total hip replacement. J Bone Joint Surg Br. 2009;91:928-934. 16. Smabrekke A, Espehaug B, Havelin LI, Furnes O.
Operating time and survival of primary total hip Journal of Surgical Education Volume ]/Number ] ] ]]]]
replacements: an analysis of 31,745 primary cemented and uncemented total hip replacements from local hospitals reported to the Norwegian Arthroplasty Register. Acta Orthop Scand. 2004;75(5):524-532.
26. Temple J. A review of the impact of the European
17. Sosa JA, Bowman HM, Tielsch JM, et al. The
27. Howells N, Brinsden MD, Gill RS, Carr AJ, Rees JL.
importance of surgeon experience for clinical and economic outcomes from thyroidectomy. Ann Surg. 1998; 228(3):320-330.
Motion analysis: a validated method for showing skill levels in arthroscopy. Arthroscopy. 2008;24(3): 335-342.
18. Pitts D, Rowley DI, Sher JL. Assessment of perform-
28. Howells NR, Gill RS, Carr AJ, Price AJ, Rees JL.
ance in orthopaedic training. J Bone Joint Surg [Br]. 2005;87(9):1187-1191. 19. Marriott J, Purdie H, Crossley J, Beard JD. Evaluation of
procedure-based assessment for assessing traineesʼ skills in the operating theatre. Br J Surg. 2011;98(3):450-457. 20. Minekus JP, Rozing PM, Valstar ER, Dankelman J.
Evaluation of humeral head replacements using timeaction analysis. J Shoulder Elbow Surg. 2003;12:152-157. 21. Van Oldenrijk J, Schafroth M, Bhandari M, Runne W,
Poolman R. Time-action analysis (TAA) of the surgical technique implanting the collum femoris preserving (CFP) hip arthroplasty. TAASTIC trial Identifying pitfalls during the learning curve of surgeons participating in a subsequent randomized controlled trial. BMC Musculoskelet Disord. 2008;9(93):1-9. 22. Den Boer KT, Straatsburg IH, Schellinger AV, de Wit
LT, Dankelman J, Gouma DJ. Quantitative analysis of the functionality and efficiency of three surgical dissection techniques: a time-motion analysis. J Laparoendosc Adv Surg Tech A. 1999;9(5):389-395.
Working Time Directive on the quality of training. Retrieved from Time for Training. Available at: 〈www.mee.nhs.uk〉.
Transferring simulated arthroscoic skills to the operating theatre. A randomised blinded study. J Bone Joint Surgery Br. 2008;90(4):494-499. 29. Aggarwal R, Ward J, Balasundaram I, et al. Proving
the effectiveness of virtual reality simulation for training in laparoscopic surgery. Ann Surg. 2007;246 (5):771-779. 30. Aggarwal R, Grantcarov TP, Erikson JR, et al. An
evidence-based virtual reality training programme for novice laparoscopic surgeons. Ann Surg. 2006;244(2): 310-314. 31. Aggarwal R, Darzi A. Technical skills training in the
21st century. N Engl J Med. 2006;355:2695-2697. 32. Leenders T, Vandevele D, MAhieu G, Nuyts R.
Reduction in variability of acetabular cup abduction using computer assisted surgery: a prospective and randomized study. Comput Aided Surg. 2002;7(2): 99-106. 33. Seyler T, Lai L, Sprinkle D, Ward W, innah R. Does
Grimbergen CA. Comparison of efficiencies of three techniques for colon surgery. J Laparoendosc Adv Surg Tech A. 2000;10(1):47-53.
computer-assisted surgery improve accuracy and decrease the learning curve in hip resurfacing? A radiographic analysis J Bone Joint Surg Am. 2008;90(1):71-80.
24. Biau DJ, Porcher R, Salomon LJ. CUSUM: a tool for
34. Gofton W, Dubrowski A, Tabloie F, Backstein D. The
ongoing assessment of performance. Ultrasound Obstet Gynecol. 2008;31(3):252-255.
effect of computer navigation on trainee learning of surgical skills. J Bone Joint Surg Am. 2007;89(12): 2819-2827.
23. Sjoerdsma W, Meijer DW, Jansen A, den Boer KT,
25. Lee YK, Biau DJ, Yoon BH, Kim TY, Ha YC,
Koo KH. Learning curve of acetabular cup positioning in total hip arthroplasty using a cumulative summation test for learning curve (LC-CUSUM). J Arthroplasty. 2014;29(3):586-589.
Journal of Surgical Education Volume ]/Number ] ] ]]]]
35. Luft HS, Bunker JP, Enthoven AC. Should operations
be regionalized? The empirical relation between surgical volume and mortality N Engl J Med. 1979;301(25): 1364-1369.
5
Learning Curves in Arthroplasty in Orthopedic Trainees Obinna Nzeako, MB, ChB, and Diane Back, MBBS, BSc Department of Orthopaedics, St Thomas’ Hospital, London, UK The NHS is adapting to a changing environment, in which economical constraints have forced theatres to maximise efficiency. An environment in which working hours and surgical exposure has been reduced and outcomes are being published. Litigation is high, and patients are living longer with higher demands. We ask, will traditional methods of apprentiship type training suffice in producing competent arthroplasty surgeons when hands on experience is falling. We review learning curves and assessment tools available to accurately assess competency and support trainee orthopaedic surgeons in their acquisition of surgical proficiency. C 2016 Published ( J Surg Ed ]:]]]-]]]. Crown Copyright J by Elsevier Inc. on behalf of the Association of Program Directors in Surgery. All rights reserved. ) KEY WORDS: hip replacement, knee replacement, arthroplasty, learning curve, trainee COMPETENCIES: Medical Knowledge, Practice Based
Learning and Improvement, Patient Care
INTRODUCTION As the National Health Service adapts to a changing world, where people are living longer with higher expectations; but where economical constraints have forced hospitals to reexamine processes to maximize efficiency—much attention is on orthopedic training. The world health organization projects that by 2020 osteoarthritis would be the fourth leading cause of disability.1 Unsurprisingly, the number of performed total hip replacements (THRs) and total knee replacements (TKRs) are placing increasing pressures on hospitals, in their challenge to manage waiting times.2 Thus, consultant orthopedic surgeons are increasingly finding themselves torn between achieving targets and providing adequate training and sufficient load exposure to produce competent surgeons. Traditionally, orthopedic surgeons undertook an apprenticeship type training program, in which they would start by Correspondence: Inquiries to Obinna Nzeako, MB, ChB, Department of Orthopaedics, St Thomas’ Hospital, Westminster Bridge Rd, London SE1 7EH, UK; e-mail: [email protected]
assisting, then would be observed and supervised until they were competent to perform arthroplasty surgery unsupervised. It was assumed that the trainee would become competent in hip and knee arthroplasty by completion of a predetermined length of training. The concern with this model of training is that the patient may not receive the highest level of surgery from a relatively inexperienced surgeon, which may translate to poorer clinical outcomes. This has become even more relevant with the data on individual consultant outcomes now being published. Naturally, the result would be more consultant performed procedures, and less hands-on experience during specialty training, especially during the early years. Together with the implementation of the European working time directive,3 National inquiry into Patient Outcomes and Deaths,4 Modernizing Medical Careers (MMC), 5 and the Joint Committee in Surgical Training surgical education in the United Kingdom has been revisited. Similar changes have been reported in the United States, and a recent study has revealed that surgical trainees report having very few supervised operative experiences in basic procedures deemed essential by their program directors.6 With the introduction of the working time directive, the reduction in out of hours operating, the development of MMC, and development of the new specialist training programs (ST 1-7) there is marked concern that surgeons are entering consultant positions with significantly less experience than their predecessors. Analysis of log books from 2004 suggested that MMC reduced the number of primary THRs and TKRs performed by trainees from 37 and 44 to 22 and 24, respectively.7 A further study published in Ann R Coll Surg Eng showed that SpRs performed almost 35% less TKRs after the Independent Sector Treatment Centres were operational.8 With training moving toward a competency based framework and a modern orthopedic curriculum that involves significantly less hands-on experience, a better understanding of learning, and skill acquisition is essential. What is more there a need to fully use alternative techniques to aid in learning and assessment to ensure these competencies are truly achieved.
Journal of Surgical Education Crown Copyright & 2016 Published by Elsevier Inc. on behalf of the Association of Program Directors in Surgery. All rights reserved. 1931-7204/$30.00 http://dx.doi.org/10.1016/j.jsurg.2016.02.006
1
Simulation appears to be emerging as a useful tool, but has been critized for its poor transferability to the operative setting. This article reviews current available evidence on learning curves of junior orthopedic surgeons embarking on hip and knee arthroplasty surgery. We would clarify the learning curve and review its reported effect on surgical outcomes of orthopedic trainees when compared with consultant surgeons. We would also review tools and methods of assessing competency.
UNDERSTANDING THE LEARNING CURVE When learning a new procedure, performance usually improves with experience. Graphically plotting performance against experience produces a learning curve. Initially, the gradient of ascent indicates how quickly the clinician is improving. As the degree of improvement attained with each case reduces as technique is refined, overall performance begins to tail off. A point is then reached when the individual is able to perform the procedure independently and still achieve the same good outcomes. Further experience still improves performance by small amounts until a plateau is reached.9 In surgery, there are potentially dramatic implications of misjudging a surgeon or units position on the learning curve. Recent examples include the Oxford Paediatric Cardiac death inquiry and the General Medical Council inquiry into the Bristol Paediatric Surgical Unit, who stated that “patients should not be exposed to Surgeons at the start of their learning curves.”10 Measures of learning curves in arthroplasty can be broadly divided into measures of the surgical process and measures of patient outcomes, for example, length of stay, transfusion requirements, functional outcomes, and morbidity An extensive review of the literature revealed a very few published data specifically plotting the learning curves of trainee surgeons in arthroplasty surgery. There is also little evidence to support the use of one of these outcome measures over another in the assessment of learning. Within training, the challenge is providing enough time and exposure to facilitate universal achievement of competency. A variety of teaching and assessment strategies must be employed to support this process. As supervised training opportunities dwindle newly appointed consultants must recognize their position on the learning curve and the need for further training, structured appraisal and senior mentorship.
THE LEARNING CURVE AND CLINICAL OUTCOMES Several recent studies have compared clinical and radiographic outcomes of THRs and TKRs performed either by a consultant or a trainee (supervised and unsupervised). A commonly cited prospective randomized trial of THRs 2
using the Stanmore and Charnley prostheses showed that unsupervised trainees had higher revision rates than those completed by an orthopedic consultant.11 In a study of 2906 hip hemiarthroplasties performed by senior registrars, junior registrars, or senior house officers (SHO), Unwin and Thomas12 found higher dislocation rates for junior registrars, and highest in SHOs. However, there is good recent evidence that supervised trainees can achieve similar surgical results when compared with consultant orthopedic surgeons. Mahaluxmivala et al.13 looked at 673 press fit condylar posterior cruciate sacrificing TKRs. They assessed surgeon experience against radiographic parameters for component positioning. They concluded that there was no difference between supervised trainees and consultants. In addition, Moran et al. reported no difference in the Harris Hip scores for THRs performed by specialist registrars vs. consultant at up to 18 months of follow-up.14 A multicentre prospective study of 1501 THRs performed either by a specialist registrar or consultant published in the JBJS(Br) concluded that there was no difference in functional outcome scores at up to 5 years follow-up.15 They also reported no difference in complications including dislocation rates. They did, however, report a statistically significant difference (p o 0.001) in operating time between the 2 groups, with a mean difference of 28 minutes (shorter duration in consultant group), and the length of hospital stay (p ¼ 0.22), 9.8 days in the consultant group, and 11 days in the registrar group. Interestingly, they also divided the registrar group into senior and junior registrars, and reported an Oxford hip score of 25.2 at 5 years follow-up in the senior registrar group that was significantly higher than (p ¼ 0.001) than that in the junior registrar group (Oxford hip score ¼ 21.8). However, there was no difference in complication or revision rate.15 These findings are reassuring to the public and to trainers. Supervised trainees are consistently found to be achieving comparable clinical results to consultants when performing arthroplasty surgery This suggests that the learning curve of trainee surgeons should not compromise the quality of the surgery provided. There has previously been some concern regarding the greater length of time required for trainee surgeons. A total of 31 Norwegian study based on 745 THRs showed that the mean operating time for cemented hip arthroplasty was 96 minutes.16 Operating times greater than 96 minutes were associated with greater revision rates for aseptic loosening, and those greater than 150 minutes were associated with increased rates of infection. However, this was not replicated by Palan et al.,15 a multicentre prospective study of 1501 THAs. What is clear is the key is, supervision appropriate to the traineeʼs level of experience. Older studies have shown greater complication rates and inferior clinical outcomes when joint replacement surgery is performed by unsupervised trainees still on the learning curve.12,17 There is little consensus on how many times a particular procedure must be performed to achieve competency for Journal of Surgical Education Volume ]/Number ] ] ]]]]
independent practice. A more focused training program is crucial. We must better understand the learning process and utilize available resources to maximize on training experiences.
ORTHOPEDIC TRAINING AND THE FUTURE Modern UK orthopedic training is based on the achievement of multiple core-competencies set out by OCAP.18 There are, however, few proven assessment tools specific to Arthroplasty. OCAP necessitates the use of work based assessments (WBAs) to evaluate trainee performance. In arthroplasty, procedure based assessments are used to assess surgical skill against a predetermined set of criteria.19 This allows for immediate formative feedback and facilitates the assessment of competency within a clinical setting. This forces the trainee to reflect on each stage of the procedure. These have, however, received criticism in recent times. Many trainers and trainees have felt them to be time consuming tick box exercises. The usefulness of each WBA is largely dependent on the willingness of the trainer and trainee to fully engage with the process. Within Europe and Canada another tool uses video recordings of the surgical procedure to objectively analyze the learning curve and evaluate surgical performance.20 This is referred to as time action analysis (TAA). The whole surgical procedure is divided into the following 5 different goal oriented phases: the incision phase, the femoral phase, the acetabulum phase, the stem phase, and the closure phase. Each goal oriented phases is subdivided in goal oriented actions and each goal oriented actions is subdivided in separate actions thereby defining all the necessary actions to complete the procedure.21 By analyzing the number and duration of the actions needed for a surgeon to achieve his or her goal, and the efficiency of each action, individual learning curves can be plotted. Not only does it allow assessment of competency but also objective comparison of multiple surgeons and trainees. By comparing experienced and inexperienced surgeons, it is possible to identify particularly difficult steps within a procedure that need further consideration during the learning curve.21 This tool has been well reported in the assessment of learning curves in laparoscopic procedures and shoulder arthroplasty.22,20,23 The learning curve cumulative sums method (CUSUM) is another tool developed from the CUSUM used for the long-term assessment of surgeon performance.24 Recently, it has been increasingly used to evaluate the learning curves in several medical procedures. Each time a procedure is performed it is scored against a predefined set of criteria. A method of statistical process control is employed to detect the adequacy of performance and determine when proficiency has been achieved. Lee et al.,25 used LC-CUSUM to show that there is a substantial learning period necessary in the optimal positioning of an acetabular cup during THR. Journal of Surgical Education Volume ]/Number ] ] ]]]]
The CUSUM technique too has received criticism for occasionally generating false-negatives; classifying adequate procedures as inadequate.24 Assessment and support of individual learning curves have not traditionally been a part of UK specialty training. However, with the development of these tools, the changing hospital dynamics and the reduction in individual operating numbers there is a recognized need for the inclusion of more comprehensive strategies.
SIMULATION AND COMPUTER NAVIGATION Simulation is quickly becoming an integral part of modern training. It provides conceptual scaffolding, promotes collaborative learning and reinforces motor learning, all within a safe nonclinical environment. There is an abundance of evidence showing that simulation can enhance technical ability and decision making in trainees in other specialties such as general surgery, anesthetics and obstetrics, and gynecology.26-31 The benefits have been well described for laparoscopy and arthroscopy, however, there is a lack of evidence in the literature detailing the advances of simulation technology in the training of open surgical skills. Some authors showed that after using computer navigation in the operating department, even experienced surgeons demonstrated improved accuracy in freehand component positioning.32 Given its potential to improve understanding of the operative field and the individual steps, it has been assumed to be a valuable educational resource. Similarly, other authors have shown that computer navigation technology can reduce the length of the learning curve in hip resurfacing.33 Perhaps, unsurprisingly within the United States it is being increasingly utilized with surgical trainees.34 There has, however, been a concern of the development of operator dependence on feedback from navigation technology with prolonged use. Motor learning studies have repeatedly demonstrated that extrinsic feedback (e.g., from a trainer or navigation technology) needs to be appropriately diminished over time to avoid the development of learner dependence. As a result its role within orthopedic training internationally is yet to be clarified. In a randomized study, looking at the effect of computer navigation on trainee learning of surgical skills Gofton et al.34 concluded that computer assisted orthopedic surgery resulted in improved early performance and equivalent learning. Each Specialist Advisory Committee has appointed a lead for simulation curriculum development. There are, however, logistical challenges to national expansion, such as funding for purchasing and maintaining the technology, availability of facilities for delivering training, manpower for administration 3
and training, and providing equal availability throughout the country. Despite this, many feel that it is in the best interest of the public to make this investment. There is evidence to suggest that time spent on simulators can reduce learning curves. While simulation in arthroscopy is becoming commonplace, there remains a paucity of information regarding simulation technology in arthroplasty.
4. Burke M, Callum KG, Gray AJG, et al. The 2001
report of the national confidential enquiry into perioperative deaths. London; 2001. 5. Kang S-N, Sanghrajka A, Amin A, Lee J, Briggs T.
Modernising medical careers: orthopaedic traineesʼ perspectives. Ann R Coll Surg Engl. 2005;87(suppl): S310-S312. 6. Bell RH, Biester TW, Tabuenca A. Operative experi-
CONCLUSION In 1979 Luft et al.35 reported a positive relationship between clinical outcomes and volume of procedures. This has been supported by several other studies, but despite this there is a lack of high-level data available to make an evidenced-based decision on number of procedures required to achieve competency. Sound systems for the recording and analysis of learning curves have emerged. These include LC-CUSUM, TAA, and perhaps even WBAs. The great benefit is the collection of numerical data and statistical analysis would allow for comparison of individuals, units, and educational establishments. A small step toward greater understanding of learning within the United Kingdom, could be to have trainees collect and keep their own data pertaining to their learning curve. Published evidence indicates that trainees are able to achieve good clinical outcomes with no difference in complication rates when supervised appropriately. However, there is a consistently reported longer duration of surgery, which is to be expected, and has obvious economic repercussions. However, despite this, if we do not make the necessary adjustments to accommodate for training and the learning curve, how would we produce proficient and skilled orthopedic surgeons? What certain is, more research is required to identify accurate methods of assessing and supporting ones learning curve. Perhaps this could be used to determine competency rather than obligatory numbers for each procedure. For this it seems technology may hold the key, in the form of simulation, computer navigation, or TAA.
REFERENCES
7. Sher JL, Reed MR, Calvert P, Wallace WA, Lamb A.
Influencing the national training agenda: the UK and Ireland orthopaedic E-logbook. J Bone Joint Surg Br. 2005;87(9):1182-1186. 8. Clamp J, Baiju D, Copas D, Hutchinson J, Rowles J.
Do independent sector treatment centres (ISTC) impact on specialist registrar training in primary hip and knee arthroplasty? Ann R Coll Surg Engl. 2008;90 (6):492-496. 9. Hopper A, Jamison M, Lewis G. Learning curves in
surgical practice. Postgrad Med J. 2007;83(986): 777-779. 10. Jackson.B. The Bristol enquiry report. Available at: 〈www.
bristol-inquiry.org.uk/final_report/annex_a/chapter_14_ 15.htm〉 Accessed 15.03.16. 11. Marston RA, Cobb AG, Bentley G. Stanmore com-
pared with Charnley total hip replacement: a prospective study of 413 arthroplasties. J Bone Joint Surg Br. 1996;78(2):178-184. 12. Unwin AJ, Thomas M. Dislocation after hemiarthro-
plasty of the hip: a comparison of the dislocation rate after posterior and lateral approaches to the hip. R Coll Surg Eng. 1994;76(5):327-329. 13. Mahaluxmivala J, Bankes MJ, Nicolai P, et al. The
effect of surgeon experience on component positioning in 673 pres fit condylar posterior cruciate sacrificing total knee arthroplasties. J Arthroplasty. 2001;16 (5):635-640.
1. Wadey V, Maloney W, Dev P, Walker D. Orthopae-
14. Moran M, Yap SL, Walmsley P, Brenkel IJ. Clinical
dic surgery core curriculum hip and knee reconstruction. Can J Surg. 2008;51(2):135-141.
and radiographic outcome of of total hip arthroplasty performed by trainee compared with consultant orthopaedic surgeons. J Arthroplasty. 2004;19(7):853-857.
2. Frankel S, Eachus J, Pearson N. Population require-
ment for primary hip replacement surgery: a cross sectional study. Lancet. 1999;353(9161):1304-1309. 3. Lim E, Tsui S. Impact of the European working time
directive on exposure to operative cardiac surgical training. Eur J Cardiothorac Surg. 2006;30(4): 574-577. 4
ence of residents in US general surgery programs: a gap between expectation and experience. Ann Surg. 2009;249(5):719-724.
15. Palan J, Gulati A, Andrew J, Murray DW, Beard DJ.
The trainer, the trainee and the surgeons assistant— clinical outcomes following total hip replacement. J Bone Joint Surg Br. 2009;91:928-934. 16. Smabrekke A, Espehaug B, Havelin LI, Furnes O.
Operating time and survival of primary total hip Journal of Surgical Education Volume ]/Number ] ] ]]]]
replacements: an analysis of 31,745 primary cemented and uncemented total hip replacements from local hospitals reported to the Norwegian Arthroplasty Register. Acta Orthop Scand. 2004;75(5):524-532.
26. Temple J. A review of the impact of the European
17. Sosa JA, Bowman HM, Tielsch JM, et al. The
27. Howells N, Brinsden MD, Gill RS, Carr AJ, Rees JL.
importance of surgeon experience for clinical and economic outcomes from thyroidectomy. Ann Surg. 1998; 228(3):320-330.
Motion analysis: a validated method for showing skill levels in arthroscopy. Arthroscopy. 2008;24(3): 335-342.
18. Pitts D, Rowley DI, Sher JL. Assessment of perform-
28. Howells NR, Gill RS, Carr AJ, Price AJ, Rees JL.
ance in orthopaedic training. J Bone Joint Surg [Br]. 2005;87(9):1187-1191. 19. Marriott J, Purdie H, Crossley J, Beard JD. Evaluation of
procedure-based assessment for assessing traineesʼ skills in the operating theatre. Br J Surg. 2011;98(3):450-457. 20. Minekus JP, Rozing PM, Valstar ER, Dankelman J.
Evaluation of humeral head replacements using timeaction analysis. J Shoulder Elbow Surg. 2003;12:152-157. 21. Van Oldenrijk J, Schafroth M, Bhandari M, Runne W,
Poolman R. Time-action analysis (TAA) of the surgical technique implanting the collum femoris preserving (CFP) hip arthroplasty. TAASTIC trial Identifying pitfalls during the learning curve of surgeons participating in a subsequent randomized controlled trial. BMC Musculoskelet Disord. 2008;9(93):1-9. 22. Den Boer KT, Straatsburg IH, Schellinger AV, de Wit
LT, Dankelman J, Gouma DJ. Quantitative analysis of the functionality and efficiency of three surgical dissection techniques: a time-motion analysis. J Laparoendosc Adv Surg Tech A. 1999;9(5):389-395.
Working Time Directive on the quality of training. Retrieved from Time for Training. Available at: 〈www.mee.nhs.uk〉.
Transferring simulated arthroscoic skills to the operating theatre. A randomised blinded study. J Bone Joint Surgery Br. 2008;90(4):494-499. 29. Aggarwal R, Ward J, Balasundaram I, et al. Proving
the effectiveness of virtual reality simulation for training in laparoscopic surgery. Ann Surg. 2007;246 (5):771-779. 30. Aggarwal R, Grantcarov TP, Erikson JR, et al. An
evidence-based virtual reality training programme for novice laparoscopic surgeons. Ann Surg. 2006;244(2): 310-314. 31. Aggarwal R, Darzi A. Technical skills training in the
21st century. N Engl J Med. 2006;355:2695-2697. 32. Leenders T, Vandevele D, MAhieu G, Nuyts R.
Reduction in variability of acetabular cup abduction using computer assisted surgery: a prospective and randomized study. Comput Aided Surg. 2002;7(2): 99-106. 33. Seyler T, Lai L, Sprinkle D, Ward W, innah R. Does
Grimbergen CA. Comparison of efficiencies of three techniques for colon surgery. J Laparoendosc Adv Surg Tech A. 2000;10(1):47-53.
computer-assisted surgery improve accuracy and decrease the learning curve in hip resurfacing? A radiographic analysis J Bone Joint Surg Am. 2008;90(1):71-80.
24. Biau DJ, Porcher R, Salomon LJ. CUSUM: a tool for
34. Gofton W, Dubrowski A, Tabloie F, Backstein D. The
ongoing assessment of performance. Ultrasound Obstet Gynecol. 2008;31(3):252-255.
effect of computer navigation on trainee learning of surgical skills. J Bone Joint Surg Am. 2007;89(12): 2819-2827.
23. Sjoerdsma W, Meijer DW, Jansen A, den Boer KT,
25. Lee YK, Biau DJ, Yoon BH, Kim TY, Ha YC,
Koo KH. Learning curve of acetabular cup positioning in total hip arthroplasty using a cumulative summation test for learning curve (LC-CUSUM). J Arthroplasty. 2014;29(3):586-589.
Journal of Surgical Education Volume ]/Number ] ] ]]]]
35. Luft HS, Bunker JP, Enthoven AC. Should operations
be regionalized? The empirical relation between surgical volume and mortality N Engl J Med. 1979;301(25): 1364-1369.
5