The Kelly Society
Orthopaedic Journal Table of Contents
Letter from the Chair 1
Letter from the Residency Program Director 2
2020 Kelly Visiting Professor Bio 3
Kelly Day Agenda 4
The Graduating Class of 2020 6
PGY – 5 Manuscripts 7-101
Adam Boissonneault, MD 7-21
Zachary Grabel, MD 22-40
Gregory Kurkis, MD 41-77
Dale Segal, MD 78-85
David Shau, MD 86-101
PGY – 3 Manuscripts 102-187
William Godfrey, MD 102-124
Matthew Lunati, MD 125-147
Jeremy Pflederer, MD 148-167
Huai Ming Phen, MD 168-187
Meet the Class of 2025 188-191
Orthopaedic Surgery and Research Faculty 192
P a g e | 1
LETTER FROM THE CHAIR ______
We have a long tradition of outstanding leaders of our profession who have taken time from their personal and professional schedules to serve as the Kelly Day Visiting Professor. Christopher Bono, MD clearly ranks as a leader in Orthopaedics today, and I am particularly grateful to Dr. Bono for being with us and enriching our educational program. We are especially grateful for his participation in our virtual format due to the COVID-19 pandemic and the need for social distancing.
I next want to acknowledge all of our faculty, residents, students, researchers, and staff. Each of you help contribute to “giving patients back their lives,” a privilege that not every medical specialty can claim with the same vigor. We can all be proud of how the Department has continued to grow and excel in both patient care and quality of patient, and family experience. We are excited to expand to six fellowship-trained trauma surgeons at Grady, and a number of other key faculty additions at a variety of our locations. The MSK service line
Scott D. Boden, MD remains a critically important component of Emory Healthcare, with approximately 1 out of every 3.5 new patients to Emory entering through Professor and Chair Orthopaedics, and planning for new facilities for the clinical and research Orthopaedic Surgery enterprise continues.
The blend of clinical efficiency, academic opportunities, and diverse patient care environments have resulted in the Emory Orthopaedic Residency and Fellowship training programs becoming increasingly among the most sought after in the country. We’re excited to expand our fellowship offerings to include Upper Extremity, Trauma, and Pediatrics.
We continue to undergo a major expansion of our basic/translational and clinical research team. Our Vice-Chair for Research, Hicham Drissi, PhD and our Director of Clinical Research, Michael Gottschalk, MD are leading those efforts with outstanding early success and momentum. Our new Sports Performance and Research Center (SPARC) will open in January 2021 at the Falcons Flowery Branch, and focus on sports injury prevention and recovery. The 15,000 sq.ft. research laboratory and surgical skills training lab in the new MSK Institute will be a game changer for our program when it opens in September 2021.
Support and contributions from our alumni are increasingly needed to enhance resident education and research, which are an integral component of these Departmental achievements. Thank you for your continued support.
With sincere appreciation for each of your efforts that contribute to Team MSK,
Scott
Scott D. Boden, M.D. P a g e | 2
LETTER FROM THE RESIDENCY PROGRAM DIRECTOR ______
For over 40 years, “Kelly Days” have represented the culmination of the academic year for the Orthopaedic Surgery Residency Program at Emory. The event provides us with a time to recognize the research efforts of our residents and serves as a venue by which we can discuss innovative approaches to patient care, quality improvement and knowledge gaps within our field.
These are difficult times indeed! The 2019 Corona Virus Pandemic has created chaos in almost every sector of modern life. It is our responsibility to continue to distance ourselves to minimize viral spread. As such, this is the first time in history that Kelly Day will be converted to a virtual format.
It is truly an honor to host Dr. Chris Bono as the 2020 Kelly Visiting Professor.
Chris will also hold the distinction of being the first “Virtual Kelly Visiting Professor.” Dr. Bono’s numerous accomplishments, involving both the clinical care and research of diseases of the spine, are a testament to his energy,
Tom Bradbury, MD intellect, interpersonal skills, and character. We are fortunate to briefly pause from the “daily grind” to spend time with him discussing our research initiatives Associate Professor and ways to improve the care we provide. Dr. Bono, thank you! Program Director Orthopaedic Surgery As importantly, Kelly Day gives us the opportunity to recognize the accomplishments of the Residency Class of 2020. Over the past 5 years, David, Adam, Dale, Greg, Nick and Zach have endured the rigor of training to be an orthopaedic surgeon. They have dedicated almost every waking hour to the pursuit of knowledge and surgical skill. They have sacrificed their own personal interest to attain the ability provide the best care for their own patients as they transition to the next phase of their careers. Over the past 3 months, they’ve risen to the challenge laid before them by the pandemic. They have defined what it means to be a team. In my mind, they embody our Residency Program’s ultimate objective … to produce the kind of doctor you would want to care for your own family. Please join me this in celebrating the achievements of this outstanding group!
Tom Bradbury, M.D.
P a g e | 3
2020 KELLY VISITING PROFESSOR
Christopher M. Bono, M.D.
Dr. Christopher M. Bono is an experienced adult spine surgeon with particular interest in the management of cervical and lumbar degenerative disorders. He is a national and international leader in spinal research and evidence-based medicine, enabling him to incorporate the latest treatment data into patient discussion for informed surgical decision-making.
He is a board-certified, fellowship-trained adult spine surgeon in the Orthopaedic Spine Center at Massachusetts General Hospital. After obtaining his medical degree from the State University of New York Health Science Center at Brooklyn, he completed his orthopaedic residency at the University of Medicine and Dentistry of Christopher Bono, MD New Jersey Medical School (now called Rutgers New Jersey Executive Vice Chair Medical School). Dr. Bono completed a spine surgery fellowship at Orthopaedic Surgery Dept. the University of California, San Diego.
Director, MGH/BWH Additionally, Dr. Bono specializes in cervical spinal stenosis, Orthopaedic Spine Surgery cervical myelopathy, cervical disc herniations, lumbar spinal stenosis, lumbar disc herniations, and lumbar spondylolisthesis. Fellowship Program
Dr. Bono is Professor of Orthopaedic Surgery at Harvard Medical Professor School. He is past president of the North American Spine Society Orthopaedic Surgery (2015-2016) and continues to be quite active in a number of Harvard Medical School committees. He serves as the current Editor-in-Chief of The Spine Journal, which holds the highest impact factor among journals dedicated to spine. Dr. Bono is also a member of the American Academy of Orthopaedic Surgeons, Cervical Spine Research Society, and the International Society for the Study of the Lumbar Spine, holding or having held many leadership positions in these societies as well. P a g e | 4
2020 Kelly Day Virtual Agenda 8:00 AM Opening Remarks Scott Boden, MD Professor & Chair Department of Orthopaedic Surgery 8:10 AM Resident Research Presentations: Session I Adult Spinal Deformity Knowledge in Orthopaedic Spine Surgeons: Impact of Fellowship Training, Experience, and Practice Characteristics Zachary Grabel, MD – PGY 5
Red Flags for Low Back Pain Are Not Always Really Red: A Prospective Evaluation of the Clinical Utility of Commonly Used Screening Questions for Low Back Pain William Godfrey, MD – PGY 3
Anterior vs Posterior Revision Total Hip Arthroplasty: A Comparative Analysis of Complications Gregory Kurkis, MD – PGY 5
Pre-operative Depression is a Risk Factor for Complication and Increased Health Care Utilization Following Total Shoulder Arthroplasty Matthew Lunati, MD – PGY 3
Fusions Ending Above the Sagittal Stable Vertebrae in Adolescent Idiopathic Scoliosis: Does it Matter? Dale Segal, MD – PGY 5
Opioid Use Following Total Shoulder Arthroplasty: Who Requires Refills and for How Long? Jeremy Pflederer, MD – PGY 3
8:55 AM Discussion and Feedback: Dr. Bono and Emory faculty
9:10 AM Basic Science Presentation I: Application of a Repurposed FDA Approved Drug for Spinal Fusion Nick Willett, PhD
9:25 AM Resident Research Presentations: Session II Pre-operative Echocardiogram in Hip Fracture Patients: Are We Hindering Care? Adam Boissonneault, MBChB – PGY 5
Fully Threaded Sacroiliac Lag Screws Have Higher Load to Failure When P a g e | 5
Compared to Partially Threaded Screws: A Biomechanical Study Ming Phen, MBBS – PGY 3
Current Trends in Clinical Practice for the Direct Anterior Approach Total Hip Arthoplasty Nick Patel, MD – PGY 5
Simultaneous Carpal and Cubital Tunnel Release: Who Needs Both? William Runge, MD – PGY 3
Medicaid Payer Status is Associated with Increased 90-Day Morbidity and Resource Utilization Following Primary Shoulder Arthroplasty: A Propensity- Score-Matched Analysis David Shau, MD – PGY 5
The P-mJOA – A Patient-Derived, Self-Reported Outcome Instrument for Evaluating Cervical Myelopathy: Comparison with the mJOA Jeff Shi, MD – PGY 3
10:10 AM Discussion and Feedback: Dr. Bono and Emory faculty
10:25 AM Trending Topic Presentation: COVID-19 in 2020 Colleen Kraft, MD Director, Clinical Virology Research Laboratory Assoc. Professor, Department of Medicine, Division of Infectious Diseases Assistant Professor, Department of Pathology
11:05 AM Basic Science Presentation II: Diet & Disc Degeneration Svenja Illien-Junger, PhD
11:20 AM Research Update Hicham Drissi, PhD
11:35 AM Kelly Day Lecture 2020 Kelly Visiting Professor Christopher Bono, MD
12:30 PM Closing Remarks Scott Boden, MD & Tom Bradbury, MD
P a g e | 6
The Graduating Class of 2020
Adam Boissonneault, MBChB Zachary Grabel, MD
Trauma, Cowley Shock Trauma Center Spine Surgery, Cleveland Clinic Gregory Kurkis, MD Nick Patel, MD
Adult Recon, Colorado Joint Replacement Adult Recon, Brigham & Women’s Hospital Dale Segal, MD David Shau, MD
Spine Surgery, Brigham & Women’s Hospital Hip & Knee Arthroplasty, UCSF
P a g e | 7
Pre-Operative Echocardiogram in Hip Fracture Patients: Are We Hindering Care?
Adam Boissonneault MBChB, Andrew Schwartz MD, Christopher Johnson BS, Rachel Mather MD, Roberto Hernandez-Irizarry MD, William Reisman MD, McKenzie Hollon MD, Mara Schenker MD
P a g e | 8
Introduction patient and systematic factors that predict at-risk
Hip fracture care continues to be a near- populations. The standardized modified frailty universal part of on-call orthopaedic duties, driven index quantifies a patients physiologic reserve as a largely by the increasingly older population; function of their comorbidity profile, and has been international projections exceed 4.5 million found to be positively associated with short-term patients by the year 2050.[1] Despite its complication and mortality risk,[4, 5] even in the commonality, hip fracture care also remains young patient.[6] Conceptually similar, though complex, with a high incidence of both morbidity physiologically distinct, serologic markers of and mortality in the short-term postoperative malnutrition also portend morbidity and mortality course. While this is multimodal, these patients in hip fracture patients.[7-10] Further, in rare cases tend to be chronically ill, with over 43% of patients of coexistence of these composite markers of stress presenting with at least three comorbid medical response capacity, morbidity and mortality suffer conditions, [2] and nearly 20% of all patients noted as a synergistic function of both conditions.[11] to suffer at least one medical complication during Surgeons have also been implicated as modifiers of the postoperative inpatient course.[3] The stakes postoperative outcome, with delays to are high, as 30-day and 1-year mortality rates are intervention,[12-15] general anesthesia,[16-18] and noted to be at least 4 and 16%, respectively, though non-comanaged perioperative care [19-21] being the rate of mortality in patients experiencing any associated with worsened risk. complication are over 8 times higher at 30 days Given the commonality of hip fractures, the and nearly 3 times higher at 1 year.[3] ubiquity to all orthopaedic practices, and the thin
Given the baseline morbidity and mortality margin for error in an inherently vulnerable associated with geriatric hip fractures, multiple population, a great deal of investigation has offered investigations have used refined measures of evidence-based guidelines to replace or underscore P a g e | 9 dogmatic practice. One recurring theme is a focus technology (CPT) codes, including 27236 (open on early surgery on the otherwise optimized treatment of femoral fracture, proximal end, neck, patient, as geriatric physiology does not tolerate internal fixation or prosthetic replacement), 27244 prolonged, strict bedrest.[22-26] A common cause (treatment of intertrochanteric, peritrochanteric, or for delay in these patients is overuse of the subtrochanteric femoral fracture; with plate/screw preoperative echocardiogram,[27, 28] despite prior type implant, with or without cerclage), and 27245 retrospective studies that have focused on study (treatment of intertrochanteric, peritrochanteric, or stewardship.[29] Thus, the primary aim of this subtrochanteric femoral fracture; with study is to evaluate the perioperative surgical and intramedullary implant, with or without anesthetic implications of a preoperative interlocking screws and/or cerclage). Included echocardiogram, and thus quantify the necessity of patients had an acute, isolated proximal femur this oft-ordered study and risk stratify patients for fracture, were >60 years old, and underwent its use. A secondary aim was to evaluate the surgical treatment of their hip fracture on the same interrelationship between frailty index scores and inpatient stay after presentation to the emergency prevalence of peri-operative exams and cardiac department. Excluded patients had subacute or complications. chronic presentations, concomitant injuries,
Methods diagnoses other than acute fracture of the native
Patient Selection proximal femur (e.g. infection, tumor-associated
We reviewed the records of consecutive fracture, periprosthetic fracture), and/or patients presenting to a single, urban, government- preoperative diagnosis of acute coronary funded American College of Surgeons Level I syndrome. trauma center. Patients were identified in our Data Collected institutional database via current procedural Patient demographics were collected: age, P a g e | 10 sex, race, injury mechanism, and baseline basis according to physician preference by either ambulatory status. Comorbid conditions were cardiologists, hospitalists, and/or collected as part of the standardized 11-item anesthesiologists), binary presence or absence of
Modified Frailty Index (mFI), which is well- any procedural cardiac intervention after described in its resolution to predict outcomes after echocardiogram, and intraoperative complications hip fractures.[11, 30] This includes diabetes (venous thromboembolism [VTE], MI, death). mellitus, non-independent functional status, Postoperative outcomes included: cardiac chronic obstructive pulmonary disease, congestive complications (MI, cardiac arrest, CHF heart failure (CHF) requiring admission within exacerbation), length of hospital stay (LOS), length thirty days prior to assessment, history of of intensive care unit (ICU) stay if applicable myocardial infarction (MI), history of (otherwise noted as 0), discharge disposition, 30- percutaneous coronary intervention or angina, day readmission to the same institution, and non- peripheral vascular disease or rest pain, cardiac complications that include acute kidney hypertension requiring medication, impaired injury, altered mental status, delayed ICU sensorium, history of transient ischemic attack or admission, acute respiratory distress syndrome, cerebral vascular accident (CVA), and history of urinary tract infection, central line-associated
CVA with resulting neurological deficit. The bacterial infection, surgical site infection, VTE number of conditions is then divided by 11, with (deep vein thrombosis or pulmonary embolus), the resultant ratio defining the mFI. respiratory distress, CVA, and/or mortality.
Perioperative data included time from Statistical Analysis presentation to surgery (in hours), anesthetic type Data was analyzed using Stata. Two
(binary: general versus spinal), binary presence or cohorts were created: those who did and did not absence of an echocardiogram (obtained on case receive echocardiogram studies preoperatively. P a g e | 11
Patient demographic, preoperative, intraoperative, between groups. Among the entire cohort, there and postoperative data were analyzed using two- were 7 patients (4%) with major cardiac events and tailed t-test for normally distributed data and Mann an overall 35% in-hospital complication rate. Of
Whitney U test for nonparametric data. The the patients with major cardiac events, only one primary outcomes of interest were intraoperative had a pre-operative echo. In patients with moderate complications and postoperative cardiac to severe frailty (mFI > or = 0.27) and a personal complications. All other outcomes constituted history of cardiac event (prior MI or percutaneous secondary outcomes. An α<0.05 was selected to coronary intervention), the odds of a major peri- indicate statistical significance. operative cardiac event increased by over 11-fold
Results (p=0.004). Of the patients that met criteria of mFI
The mean age of the cohort was 77 years > or = 0.27, only two (14%) of these patients
(range 60-98 years). Of the 181 patients, 21 received a pre-operative echo. patients (12%) received a pre-operative echo. Of Discussion these 21 patients, no patient had a pre-operative Our findings indicate little discernible cardiac intervention and no patient had an obvious difference between patients who were indicated for change in anesthetic plan. There was no difference a preoperative echocardiogram by either a in mean age between patients that did and did not cardiologist, hospitalist, or anesthesiologist, but a receive an echo (Table 1). Patients that received an greater than two times longer wait for a surgical echo had a higher mean ASA score, 3.1 compared hip fracture treatment. In these twenty one patients, to 2.7 (p=0.003). The mean time to surgery was none of them received a preoperative cardiac over twice as long for patients who required a pre- procedure, nor did any receive a change in operative echo, 73 compared to 30 hours anticipated anesthetic type. In the seven patients
(p<0.001). There was no difference in mFI who experienced a postoperative cardiac event, P a g e | 12 only one had a preoperative echocardiogram. not new, as prior literature has investigated focused
Similarly, while an mFI ≥ 0.27 or a history of methods to decrease the use of unindicated either MI or PCI portended an 11-fold increase in preoperative echocardiography.[31] There is some risk of cardiac event, only 14% of these patients concern that the uncertainty of this inherently received a preoperative echo. This suggests that vulnerable patient population has spurred less decision-making in preoperative echocardiogram restrictive echocardiogram algorithms, though triage at our institution was likely guided by medical teams that focus on comanaged hip improper and/or inconsistent criteria, and thus did fracture care has proven a successful avenue for a not yield any reflexive preoperative optimizing refined echocardiogram triage system.[19, 32-34] interventions, and did not identify the at-risk While an echocardiogram is a useful tool to patient for postoperative cardiac complications. direct cardiac interventions and adjust
Our findings are of particular concern, intraoperative protocols, our findings support the because the overarching goals of hip fracture care notion that they are associated with delays to are early surgical intervention, immediate surgical treatment for hip fracture patients.[35, 36] postoperative mobility, and While many orthopaedic interventions are anticipation/minimization of complications in the generally unaffected by a delay of surgical medically fragile. Unintentionally paradoxical, intervention by an extra day, hip fracture patients however, non-protocolized use of echocardiograms are notoriously intolerant of surgical delays; prior was found to be associated with longer delays to literature has demonstrated both the morbidity and surgery without commensurate reflexive mortality of prolonged presurgical periods, even preoperative cardiac intervention, and does not after controlling for patient comorbidities and other capture the patients who ultimately have a potential drivers of bias towards delay in sicker postoperative cardiac complication. This concern is patients.[12, 37-40] While there is sensitivity in the P a g e | 13 data to protocol and outcomes of interest, the care of these patients leading up to surgery. general target for surgery is within the first 24-72 Of further concern, the majority of patients hours to optimize outcomes. In our study, patients who went on to experience a cardiac event who did not receive an echocardiogram postoperatively were not in the cohort who was preoperatively were treated within 30 hours of initially indicated for an echocardiogram admission, yet those who did undergo an preoperatively. This underscores the previously- echocardiogram took over twice that long, with an introduced concept that preoperative patient average of 73 hours. This is owed at least in part to echocardiogram triage continues to be important in the labor-intensive process of obtaining and the postoperative period. While it is possible that interpreting a high-quality echocardiogram, which the poor performance of subjective hospitalist, typically involves a bedside technologist and cardiologist, and/or anesthesiologist cardiology specialist at a minimum. This delay and echocardiogram indications were an institutional utilization of resources would potentially be phenomenon, this supports prior literature that justifiable by an optimizing alteration in care raises similar worries.[29, 41-44] A prior leading up to surgery, however no patient comparative benchmark for adherence to a underwent a preoperative cardiology intervention standardized indications protocol is the American or deviation in typical anesthetic plan as a result. College of Cardiology/American Heart Association
Rather, the preoperative echocardiogram simply Clinical Practice Guidelines, though this is not served to remind us that these patients are specific to hip fracture patients.[45] Rather, we simultaneously chronically ill, yet generally not have identified two underlying commonalities that indicated for urgent corrective procedures before reflect contemporary advancements in the study of expedient hip fracture surgery. Ultimately, physiologic decline and cardiac risk. We found that preoperative echocardiogram added little to the patients with an eleven-item mFI ≥0.27 of prior P a g e | 14 history of acute coronary syndrome equivalent (MI limitations to our study. First, the study is or PCI) offered a reliable and narrow predictive retrospective. Thus, the hindsight afforded by history for patients who would experience a knowledge of the ultimate patient outcome does postoperative cardiac event, which supports prior not necessarily justify the decision to proceed with findings in a randomized clinical trial in this a preoperative echocardiogram; it is possible that a population.[46] Of note, a history of heart failure negative study ruled out a condition that would’ve did not predict either a change in preoperative altered the preoperative course or intraoperative cardiac treatment or the odds of a postoperative protocol. Further, despite reviewing a consecutive cardiac event, despite this being a common cited series of cases, it remains subject to selection bias. reason to obtain a preoperative echocardiogram. It is possible that the selected patients did not
While preoperative cardiac risk is in defined by the possess the risk factors or risk factor severity that ejection fraction calculated by the preoperative an echocardiogram was attempting to identify echocardiogram,[27] it is not likely an acutely when initially ordered. As such, it’s feasible that modifiable risk factor, so this may be a relatively some findings would’ve altered the pre-surgical or noncritical step (and potentially wasteful and surgical plan, or, in rare cases, surgical indications surgery delaying) in the preoperative cardiac in the first place. Additionally, while a single- clearance process. institution study likely enhances the consistency of
Despite our findings of the preoperative, data and verifiability of results, it also potentially intraoperative, and postoperative cardiac course of confounds our findings with sensitivity to hospital patients who underwent and did not undergo protocol(s) that may not be universally applied. preoperative echocardiogram, and clear definitions However, our findings and triage recommendations of who best to reserve this costly and inherently are easily applicable after basic history-taking, delaying diagnostic study, there remains which is ubiquitous to any institution in the world. P a g e | 15
Conclusion
Geriatric hip fractures, though common at nearly every hospital in the world, present unique clinical challenges given their inherent physiologically compromised host and relatively urgent need for surgery to preserve mobility and minimize the risk of acute and rapid decline. While conceptually pragmatic, a preoperative echocardiogram adds little to the preoperative and intraoperative care of hip fracture patients.
The risks of surgical delay likely outweigh the potential benefits before a surgery of necessity. We identify a mFI score ≥0.27 and history of MI and/or PCI as risk factors for postoperative cardiac complications, and thus recommend a postoperative echocardiogram in these patients, as it’s possible, then, that they may be a candidate for a same-stay cardiac procedure to enhance these patients’ cardiac health. This likely has more potential meaningful clinical benefit than a preoperative echocardiogram. We propose rare indications for a preoperative echocardiogram, and perhaps consideration for expeditious and focused echocardiogram administered by a subspecialized cardiac anesthesiologist, when available, prior to surgery in those deemed exceptionally high risk of imminent cardiac events during surgical treatment of a hip fracture. P a g e | 16 Table 1. Outcome measures by echo status
P a g e | 17
References
1. Veronese N, Maggi S. Epidemiology and social costs of hip fracture. Injury 49(8): 1458, 2018 2. Bekeris J, Wilson LA, Bekere D, Liu J, Poeran J, Zubizarreta N, Fiasconaro M, Memtsoudis SG. Trends in Comorbidities and Complications Among Patients Undergoing Hip Fracture Repair. Anesth Analg, 2019 3. Lawrence VA, Hilsenbeck SG, Noveck H, Poses RM, Carson JL. Medical complications and outcomes after hip fracture repair. Arch Intern Med 162(18): 2053, 2002 4. Boissonneault A, Mener A, Schwartz A, Wilson J, Staley C, Schenker M. Impact of Frailty on 30-Day Morbidity and Mortality of Patients With Intertrochanteric Femur Fractures. Orthopedics 42(6): 344, 2019 5. Traven SA, Reeves RA, Althoff AD, Slone HS, Walton ZJ. New Five-Factor Modified Frailty Index Predicts Morbidity and Mortality in Geriatric Hip Fractures. J Orthop Trauma 33(7): 319, 2019 6. Rege RM, Runner RP, Staley CA, Vu CCL, Arora SS, Schenker ML. Frailty predicts mortality and complications in chronologically young patients with traumatic orthopaedic injuries. Injury 49(12): 2234, 2018 7. Bohl DD, Shen MR, Hannon CP, Fillingham YA, Darrith B, Della Valle CJ. Serum Albumin Predicts Survival and Postoperative Course Following Surgery for Geriatric Hip Fracture. J Bone Joint Surg Am 99(24): 2110, 2017 8. Li S, Zhang J, Zheng H, Wang X, Liu Z, Sun T. Prognostic Role of Serum Albumin, Total Lymphocyte Count, and Mini Nutritional Assessment on Outcomes After Geriatric Hip Fracture Surgery: A Meta-Analysis and Systematic Review. J Arthroplasty 34(6): 1287, 2019 9. Helminen H, Luukkaala T, Saarnio J, Nuotio M. Comparison of the Mini-Nutritional Assessment short and long form and serum albumin as prognostic indicators of hip fracture outcomes. Injury 48(4): 903, 2017 10. Chung AS, Hustedt JW, Walker R, Jones C, Lowe J, Russell GV. Increasing Severity of Malnutrition Is Associated With Poorer 30-Day Outcomes in Patients Undergoing Hip Fracture Surgery. J Orthop Trauma 32(4): 155, 2018 11. Wilson JM, Boissonneault AR, Schwartz AM, Staley CA, Schenker ML. Frailty and Malnutrition Are Associated With Inpatient Postoperative Complications and Mortality in Hip Fracture Patients. J Orthop Trauma 33(3): 143, 2019 P a g e | 18 12. Pincus D, Ravi B, Wasserstein D, Huang A, Paterson JM, Nathens AB, Kreder HJ, Jenkinson RJ, Wodchis WP. Association Between Wait Time and 30-Day Mortality in Adults Undergoing Hip Fracture Surgery. JAMA 318(20): 1994, 2017 13. Leer-Salvesen S, Engesaeter LB, Dybvik E, Furnes O, Kristensen TB, Gjertsen JE. Does time from fracture to surgery affect mortality and intraoperative medical complications for hip fracture patients? An observational study of 73 557 patients reported to the Norwegian Hip Fracture Register. The bone & joint journal 101-b(9): 1129, 2019 14. Beaupre LA, Khong H, Smith C, Kang S, Evens L, Jaiswal PK, Powell JN. The impact of time to surgery after hip fracture on mortality at 30- and 90-days: Does a single benchmark apply to all? Injury 50(4): 950, 2019 15. Ryan DJ, Yoshihara H, Yoneoka D, Egol KA, Zuckerman JD. Delay in Hip Fracture Surgery: An Analysis of Patient-Specific and Hospital-Specific Risk Factors. J Orthop Trauma 29(8): 343, 2015 16. Radcliff TA, Henderson WG, Stoner TJ, Khuri SF, Dohm M, Hutt E. Patient risk factors, operative care, and outcomes among older community-dwelling male veterans with hip fracture. J Bone Joint Surg Am 90(1): 34, 2008 17. Desai V, Chan PH, Prentice HA, Zohman GL, Diekmann GR, Maletis GB, Fasig BH, Diaz D, Chung E, Qiu C. Is Anesthesia Technique Associated With a Higher Risk of Mortality or Complications Within 90 Days of Surgery for Geriatric Patients With Hip Fractures? Clin Orthop Relat Res 476(6): 1178, 2018 18. Qiu C, Chan PH, Zohman GL, Prentice HA, Hunt JJ, LaPlace DC, Nguyen VT, Diekmann GR, Maletis GB, Desai V. Impact of Anesthesia on Hospital Mortality and Morbidities in Geriatric Patients Following Emergency Hip Fracture Surgery. J Orthop Trauma 32(3): 116, 2018 19. Swart E, Vasudeva E, Makhni EC, Macaulay W, Bozic KJ. Dedicated Perioperative Hip Fracture Comanagement Programs are Cost-effective in High-volume Centers: An Economic Analysis. Clin Orthop Relat Res 474(1): 222, 2016 20. Neuerburg C, Forch S, Gleich J, Bocker W, Gosch M, Kammerlander C, Mayr E. Improved outcome in hip fracture patients in the aging population following co-managed care compared to conventional surgical treatment: a retrospective, dual-center cohort study. BMC Geriatr 19(1): 330, 2019 21. Gosch M, Hoffmann-Weltin Y, Roth T, Blauth M, Nicholas JA, Kammerlander C. Orthogeriatric co-management improves the outcome of long-term care residents with fragility fractures. Arch Orthop Trauma Surg 136(10): 1403, 2016 P a g e | 19 22. Endo A, Baer HJ, Nagao M, Weaver MJ. Prediction Model of In-Hospital Mortality After Hip Fracture Surgery. J Orthop Trauma 32(1): 34, 2018 23. Sasabuchi Y, Matsui H, Lefor AK, Fushimi K, Yasunaga H. Timing of surgery for hip fractures in the elderly: A retrospective cohort study. Injury 49(10): 1848, 2018 24. Shiga T, Wajima Z, Ohe Y. Is operative delay associated with increased mortality of hip fracture patients? Systematic review, meta-analysis, and meta-regression. Can J Anaesth 55(3): 146, 2008 25. Nyholm AM, Gromov K, Palm H, Brix M, Kallemose T, Troelsen A. Time to Surgery Is Associated with Thirty-Day and Ninety-Day Mortality After Proximal Femoral Fracture: A Retrospective Observational Study on Prospectively Collected Data from the Danish Fracture Database Collaborators. J Bone Joint Surg Am 97(16): 1333, 2015 26. Rosso F, Dettoni F, Bonasia DE, Olivero F, Mattei L, Bruzzone M, Marmotti A, Rossi R. Prognostic factors for mortality after hip fracture: Operation within 48 hours is mandatory. Injury 47 Suppl 4: S91, 2016 27. Stitgen A, Poludnianyk K, Dulaney-Cripe E, Markert R, Prayson M. Adherence to Preoperative Cardiac Clearance Guidelines in Hip Fracture Patients. J Orthop Trauma 29(11): 500, 2015 28. Smeets SJM, van Wunnik BPW, Poeze M, Slooter GD, Verbruggen J. Cardiac overscreening hip fracture patients. Arch Orthop Trauma Surg 140(1): 33, 2020 29. Adair C, Swart E, Seymour R, Patt J, Karunakar MA. Clinical Practice Guidelines Decrease Unnecessary Echocardiograms Before Hip Fracture Surgery. J Bone Joint Surg Am 99(8): 676, 2017 30. Patel KV, Brennan KL, Brennan ML, Jupiter DC, Shar A, Davis ML. Association of a modified frailty index with mortality after femoral neck fracture in patients aged 60 years and older. Clin Orthop Relat Res 472(3): 1010, 2014 31. Soong C, Cram P, Chezar K, Tajammal F, Exconde K, Matelski J, Sinha SK, Abrams HB, Fan-Lun C, Fabbruzzo-Cota C, Backstein D, Bell CM. Impact of an Integrated Hip Fracture Inpatient Program on Length of Stay and Costs. J Orthop Trauma 30(12): 647, 2016 32. Shaw M, Pelecanos AM, Mudge AM. Evaluation of Internal Medicine Physician or Multidisciplinary Team Comanagement of Surgical Patients and Clinical Outcomes: A Systematic Review and Meta-analysis. JAMA network open 3(5): e204088, 2020 33. Kates SL, Mendelson DA, Friedman SM. Co-managed care for fragility hip fractures (Rochester model). Osteoporos Int 21(Suppl 4): S621, 2010 P a g e | 20 34. Lau TW, Fang C, Leung F. The effectiveness of a multidisciplinary hip fracture care model in improving the clinical outcome and the average cost of manpower. Osteoporos Int 28(3): 791, 2017 35. Garcia AE, Bonnaig JV, Yoneda ZT, Richards JE, Ehrenfeld JM, Obremskey WT, Jahangir AA, Sethi MK. Patient variables which may predict length of stay and hospital costs in elderly patients with hip fracture. J Orthop Trauma 26(11): 620, 2012 36. Ricci WM, Brandt A, McAndrew C, Gardner MJ. Factors affecting delay to surgery and length of stay for patients with hip fracture. J Orthop Trauma 29(3): e109, 2015 37. Vrahas MS, Sax HC. Timing of Operations and Outcomes for Patients With Hip Fracture-It's Probably Not Worth the Wait. JAMA 318(20): 1981, 2017 38. Panesar SS, Simunovic N, Bhandari M. When should we operate on elderly patients with a hip fracture? It's about time! Surgeon 10(4): 185, 2012 39. Simunovic N, Devereaux PJ, Sprague S, Guyatt GH, Schemitsch E, Debeer J, Bhandari M. Effect of early surgery after hip fracture on mortality and complications: systematic review and meta-analysis. CMAJ 182(15): 1609, 2010 40. Moja L, Piatti A, Pecoraro V, Ricci C, Virgili G, Salanti G, Germagnoli L, Liberati A, Banfi G. Timing matters in hip fracture surgery: patients operated within 48 hours have better outcomes. A meta-analysis and meta-regression of over 190,000 patients. PLoS One 7(10): e46175, 2012 41. Jettoo P, Kakwani R, Junejo S, Talkhani I, Dixon P. Pre-operative echocardiogram in hip fracture patients with cardiac murmur--an audit. J Orthop Surg Res 6: 49, 2011 42. Marcantonio A, Steen B, Kain M, Bramlett KJ, Tilzey JF, Iorio R. The Clinical and Economic Impact of Preoperative Transthoracic Echocardiography in Elderly Patients with Hip Fractures. Bulletin of the Hospital for Joint Disease (2013) 73(4): 239, 2015 43. Yonekura H, Ide K, Onishi Y, Nahara I, Takeda C, Kawakami K. Preoperative Echocardiography for Patients With Hip Fractures Undergoing Surgery: A Retrospective Cohort Study Using a Nationwide Database. Anesth Analg 128(2): 213, 2019 44. Mutlu H, Bilgili F, Mutlu S, Karaman O, Cakal B, Ozkaya U. The effects of preoperative non-invasive cardiac tests on delay to surgery and subsequent mortality in elderly patients with hip fracture. J Back Musculoskelet Rehabil 29(1): 49, 2016 45. Fleisher LA, Fleischmann KE, Auerbach AD, Barnason SA, Beckman JA, Bozkurt B, Davila-Roman VG, Gerhard-Herman MD, Holly TA, Kane GC, Marine JE, Nelson MT, Spencer CC, Thompson A, Ting HH, Uretsky BF, Wijeysundera DN. 2014 ACC/AHA guideline on perioperative cardiovascular evaluation and management of patients P a g e | 21 undergoing noncardiac surgery: executive summary: a report of the American College of Cardiology/American Heart Association Task Force on practice guidelines. Developed in collaboration with the American College of Surgeons, American Society of Anesthesiologists, American Society of Echocardiography, American Society of Nuclear Cardiology, Heart Rhythm Society, Society for Cardiovascular Angiography and Interventions, Society of Cardiovascular Anesthesiologists, and Society of Vascular Medicine Endorsed by the Society of Hospital Medicine. J Nucl Cardiol 22(1): 162, 2015 46. Canty DJ, Heiberg J, Yang Y, Royse AG, Margale S, Nanjappa N, Scott D, Maier A, Sessler DI, Chuan A, Palmer A, Bucknill A, French C, Royse CF. Pilot multi-centre randomised trial of the impact of pre-operative focused cardiac ultrasound on mortality and morbidity in patients having surgery for femoral neck fractures (ECHONOF-2 pilot). Anaesthesia 73(4): 428, 2018
P a g e | 22
Adult Spinal Deformity Knowledge in Orthopaedic Spine Surgeons: Impact of Fellowship Training, Experience, and Practice Characteristics
Zachary J. Grabel, MD; Robert A. Hart, MD; Aaron P. Clark, MD; Sara Heejung Park, BS; Christopher I. Shaffrey, MD; Justin K. Scheer BS; Justin S. Smith, MD; Michael P. Kelly, MD; J. Mason DePasse, MD; Munish C. Gupta, MD; Christopher P. Ames, MD; Alan H. Daniels, MD P a g e | 23
Abstract spinal deformity surgeons.
Study Design: Survey study. Results: Complete responses were received
Objective: The purpose of this paper was to from 413 orthopaedic spine surgeons. The assess the level of adult spine deformity (ASD) overall correct answer rate was 69.0%. knowledge among orthopaedic spine surgeons Surgeons in practice < 10 years had a and identify areas for improvement in spine higher correct answer rate compared to surgery training those who have practiced for ≥10 years.
Summary of Background Data: ASD is (74% vs. 67%; p=0.000003). Surgeons increasingly encountered in spine surgery with ≥75% of their practice dedicated to practice. While ASD knowledge among spine had a higher overall correct rate neurosurgeons has been evaluated, ASD compared to surgeons whose practice is knowledge among orthopaedic spine <75% surgeons has not previously been spine (71% vs. 63%; p=0.000029). reported. Completion of spine fellowship was associated with a higher overall correct answer rate
Methods: A survey of orthopaedic spine compared to respondents who did not surgeon members of North American Spine complete a spine fellowship (71% vs. 59%; Society (NASS) was conducted to assess p<0.00001). level of spine surgery training, practice Conclusions: Completion of spine fellowship experience, and spinal deformity and having a dedicated spine surgery practice knowledge base. The survey utilized was were significantly associated with improved previously completed by a group of performance on this ASD knowledge survey. neurological surgeons with published Unlike neurosurgeons, orthopaedic spine results. The survey utilized 11 questions surgeons who have practiced for less than 10 developed and agreed upon by experienced P a g e | 24 years performed better than those who have benefit from spine surgical intervention in practiced for more than 10 years. Ongoing carefully selected patients. As life emphasis on spine deformity education should expectancy increases and the aging be emphasized to improve adult spinal population remains active, the number of deformity knowledge base. patients undergoing surgical treatment for
Introduction spine deformity will also likely continue to
5 Most orthopaedic spine surgeons increase .
complete postgraduate fellowship training in In order to optimally treat ASD, spine
spine surgery prior to entering independent surgeons must learn and understand basic
practice1. The majority of spine surgery spinal deformity principles including coronal
fellowships include education in the evaluation and sagittal plane alignment, spinopelvic
and management of spinal disease due to parameters, clinical spine deformity
degenerative, traumatic, and malignant evaluation, indications for spinopelvic fixation
processes in addition to spinal deformity. The and methods of spinal and spinopelvic
6-9 exposure to spinal deformity varies widely instrumentation . According to guidelines
among postgraduate spine surgery training established by Herkowitz et al., these
programs, encompassing a range of 5%-50% principles should be introduced during
of the spine surgery fellowship experience2. orthopaedic residency and further emphasized
10 Adult spinal deformity (ASD) is during postgraduate fellowship training .
increasingly encountered in modern spine Because of the complex nature of treating
11-13 surgery practice3. ASD patients experience spine deformity , over 90% of orthopaedic
inferior health-related quality of life surgery and neurosurgery residency program
(HRQOL) compared to the general directors believe surgeons from either
population4 and may experience substantial specialty who want to perform spine deformity surgery must complete a spine surgery P a g e | 25 fellowship1. surgeons (18.7% response rate) with
However, the effect of fellowship training on published results the Journal of
9 ASD knowledge in orthopaedic surgeons is Neurosurgery: Spine . The survey consisted incompletely understood. of 14 multiple-cho ice questions (Appendix
While ASD knowledge among 1), which were developed by Clark et al. The practicing neurosurgeons has been studied9, survey was administered to an additional 10
ASD knowledge among practicing expert spine deformity surgeons for orthopaedic spine surgeons has not previously validation. The inter-rater reliability amongst been reported. The purpose of this study was the 10 expert spine deformity surgeons was
9 to assess the level of ASD knowledge among 85.4%. The first three questions focus on orthopaedic spine surgeons and to identify training and practice experience. The areas for improvement in orthopaedic spine remaining 11 questions were knowledge surgery training. based questions (Figure 1).
Methods These questions were subgrouped into 5 categories: 1) radiology/spinopelvic alignment,
An online survey was distributed to 2) health-related quality of life (HRQOL), 3) 2,222 orthopaedic spine surgeon members of surgical indications, 4) operative technique, North American Spine Society (NASS) to and 5) clinical evaluation. Selected questions assess level of spine surgery training, practice were sub-grouped into more than one category. experience and spinal deformity knowledge Demographic variables were base. The survey was distributed via dichotomized for analysis. Responses to “time surveymonkey.com (SurveyMonkey, Palo in practice” were grouped into “<10 years” Alto, CA) from 03/2015 to 05/2015. and “≥10 years.” Responses to “percent of The survey utilized was previously practice dedicated to spine surgery” were completed by a group of 7,781 neurological grouped into “<75%” and “≥75%.” Responses P a g e | 26 to knowledge base questions were recorded as Of those currently in practice, 96.7% correct or incorrect. completed a spine fellowship (354/366).
Differences in the rate of correct Those in practice for less than 10 answers (%) between groups were compared years (excluding trainees) were more likely to utilizing Chi-square analysis based on have completed a spine fellowship than those dichotomized participant demographic practicing for 10 years or longer (99% vs. characteristics including years of practice, 91%, respectively; p=0.00126). Excluding completion of a spinal surgery fellowship, and trainees, those who completed a spine percentage of practice comprising spinal fellowship were more likely to focus at least surgery with SPSS (IBM Corporation, 75% of their practice on spine than those who
Armonk, NY). P values <0.05 were deemed did not complete a fellowship (91% vs. 60%, statistically significant. respectively; p=0.00001).
Results Overall Knowledge Base Demographics The overall correct answer rate was Survey responses were received from 69.0% (3102/4494) (Table 2). Surgeons in 413 orthopaedic spine surgeons (18.6% practice less than 10 years had a significantly response rate): 11.4% (47/413) of respondents higher correct answer rate compared to those were still in training, 34.6% (143/413) who have practiced for 10 years or longer practiced less than 10 years, and 53.5% (74% vs. 67%; p=0.000003) (Figure 2). (221/413) practiced for 10 years or longer Surgeons with at least 75% of their practice (Table 1). dedicated to spine had a significantly higher Of the total respondents, 85.7% overall correct rate compared to surgeons (354/413) completed a spine fellowship whose practice is <75% spine (71% vs. 63%; and 83.0% (343/413) reported at least p=0.00003). Completion of spine fellowship 75% of their practice is focused on spine. P a g e | 27 was associated with a significantly higher performance on this ASD knowledge survey. overall correct answer rate compared to Both of these factors have previously been respondents who did not complete a spine shown to predict improved ASD knowledge fellowship (71% vs. 59%; p<0.00001). in neurosurgeons9. Unlike neurosurgeons,
Operative technique questions had however, orthopaedic spine surgeons who the lowest percentage correct (62%), while have practiced for less than 10 years
HRQOL questions had the highest (81%) performed better than those who have
(Table 3). Practice for less than 10 years, at practiced for more than 10 years. This may least 75% of practice focused on spine, and be due an ongoing increase in emphasis and completion of a fellowship was associated educational focus on adult spinal deformity with a higher correct response rate in all 5 in current orthopaedic spine fellowships. subcategories (Figures 3A-E). Completion of a spine fellowship was
Question 8 and question 11 the factor that most strongly influenced correct displayed the lowest correct response answer response rate amongst orthopaedic rates. Both questions had a correct spine surgeons, with an increased correct response rate below 50% (Table 2). answer rate of 12 percentage points compared
to those who had not completed a spine Discussion surgery fellowship. In contrast, neurosurgeons This investigation examined benefited most from increased specialization knowledge regarding the diagnostic of 75% of their practice dedicated to spine evaluation and treatment of adult spinal surgery. A possible explanation for this deformity in orthopaedic spine surgeons. finding is that orthopaedic surgeons perform Completion of spine surgery fellowship and significantly fewer spine surgery procedures having a dedicated spine surgery practice compared to neurosurgeons during were significantly associated with improved residency14, thus a larger portion of P a g e | 28 orthopaedic spine surgeons’ ASD knowledge spine deformity cases during residency on may be derived from fellowship. average, compared to 9.5 for orthopaedic
Additionally, neurosurgeons may also be residents, highlighting an important difference less apt to treat ASD patients, unless in spine training between the specialties14. their practice is highly specialized in In this study, orthopaedic spine complex spine reconstruction. surgeons performed best on HRQOL questions
Orthopaedic spine surgeons performed and worst on operative technique and surgical better than neurosurgeons on the survey indication questions. The demographic factor presented (69% vs. 42% overall correct with the greatest effect on the correct response response rate) 9. However, it is important to rate in HRQOL questions was length of note that only subspecialty focused practice experience; surgeons who have orthopaedic spine surgeons completed this practiced for less than 10 years performed 11 test, as compared to all neurosurgeons. When percentage points better than surgeons who controlling for completion of a spine have practiced for at least 10 years. It is likely fellowship and more than 75% of practice that ASD is more commonly encountered in the focused on spine, orthopaedic surgeons still spine surgeon’s practice than it was a decade had a greater overall correct response rate ago and spine fellowship training continues to compared to neurosurgeons (71% vs. 47%, emphasize ASD education3. Practice and 71% vs. experience of less than 10 years also had a
50%, respectively). Despite guidelines significant influence on correct response rate of surrounding spine surgeon training which are operative technique questions. This is similar for neurosurgery and orthopaedics10, consistent with the fact that surgical treatment the training experience clearly differs in terms techniques for ASD are constantly evolving of exposure to spinal deformity. Daniels et al. and new instrumentation technology is rapidly illustrated that neurosurgeons perform only 2 advancing15-16. P a g e | 29 The demographics of our study cohort controversial. Specifically, questions 8 and 11 highlight a strong trend toward completion of are concerning, as both had correct response spine surgery fellowship among orthopaedic rates below 50%. Question 8 also had the spine surgeons, as 99% of orthopaedic spine lowest correct response rate in the correlate surgeons with under 10 years of spine practice neurosurgery study9. This poor knowledge of responding to this survey completed a spine coronal plane deformity may be due to surgery fellowship. Not surprisingly, those insufficient training in coronal plane who complete a spine fellowship are more correction. Interestingly, question 11 appears likely to focus the majority of their practice on to be a straightforward multiple choice spine. question asking the average lordosis of the L5-
There are several potential S1 disc space. However, this had the lowest limitations to this study. The response rate correct response rate amongst 10 spine may be seen as a limitation, although deformity experts used to validate the several recent survey studies in the spine interrater reliability of the survey questions in literature have lower response rates9,17-18. the Clark et al study9. This suggests that
Moreover, the study of neurosurgeons question 11 may not be a reliable question to employing the same survey had a nearly assess spine deformity knowledge. The identical 19% response rate. questions, however, were identical to those
9 An additional possible concern is that previously utilized in the work by Clark et al although the survey was developed by a group to allow for comparison between the two study of highly experienced spinal deformity cohorts. Furthermore, the questions were surgeons, the answers may be viewed as only designed with some ambiguity for relative subjectively correct as evaluation and prioritization similar to the ABOS and ABNS treatment of ASD lacks consensus and can be written board questions. This study was not intended to directly P a g e | 30 compare the two specialties or to infer one with several positive and potentially negative specialty is more knowledgeable than the impacts on the orthopaedic and neurosurgery other. Rather, it was designed to assess establishments which will require further knowledge gaps in each specialty with the goal discussion amongst the spine surgery of improving ASD training and education. The leadership. results of this study may reflect exposure to Conclusion similar questions in training and for board This study assessed orthopaedic spine studying, or the practice patterns of the survey surgeons’ knowledge regarding ASD. respondents, rather than the specialty as a Completing a spine surgery fellowship whole. enhances spine deformity knowledge for
Adult spine deformity management is both orthopaedic surgeons and becoming an increasingly prominent part of neurosurgeons. With respect to orthopaedic current spine surgery practice. As such, spine spine surgery, particular attention should be surgeons must be adequately trained to paid towards operative technique and evaluate and treat (or refer) patients with this surgical indications education as these condition. With two separate training pathways categories represented the greatest ASD towards spine surgery specialization knowledge gaps among practicing
(orthopaedic surgery and neurosurgery), it orthopaedic surgeons. presents a challenge to identify knowledge gaps and improve and standardize training.
Early subspecialization during residency for spine surgeons or dedicated categorical spine surgery residency training may offer solutions to improve and standardize training of spine surgeons19. However, this is a complex topic P a g e | 31
Figure Legend:
Table 1: Respondent demographic data
Table 2: Survey overall correct answer response data
Table 3: Survey correct answer response data by question subgroup
Figure 1: Example question (with responses)
Figure 2: Overall correct response rates of orthopaedic spine surgeons comparing different demographic groups. Y= Yes; N=No.
Figure 3A-E: Correct response rates for each of the question subcategories comparing different demographic groups. Y=Yes; N=No.
Appendix I: Adult spine deformity survey questions. P a g e | 32
Table 1: Respondent demographic data
Demographic Data Responses Percent Total Respondents 413 100% <10 years experience 143 34.6% ≥10 years experience 221 53.5% Still in training 47 11.4% Unanswered 2 0.5%
<75% spine practice 69 16.7% ≥75% spine practice 343 83.1% Unanswered 1 0.2%
Surgeon in practice 366 100% Fellowship trained 354 96.7% No spine fellowship 8 2.2% Unanswered 4 1.1%
Table 2: Survey overall correct answer response data Correct Incorrect Total Correct Question # Responses Responses Responses (%) 4 329 80 409 80% 5 330 81 411 80% 6 230 176 406 57% 7 383 28 411 93% 8 156 251 407 38% 9 346 61 407 85% 10 282 129 411 69% 11 194 214 408 48% 12 276 133 409 67% 13 282 121 403 70% 14 294 118 412 71% Overall Correct Responses 69% P a g e | 33
Table 3: Survey correct answer response data by question subgroup
Correct Incorrect Total % Question Subgroups Responses Responses Responses Correct Radiology/Spinopelvic alignment Overall 1967 896 2863 69 Fellowship 1726 725 2451 70 No Fellowship 224 160 384 58 <10 years experience 732 259 991 74 ≥10 years experience 1032 498 1530 67 <75% spine practice 295 176 471 63 ≥75% spine practice 1668 707 2375 70 HRQOL Overall 994 237 1231 81 Fellowship 854 200 1054 81 No Fellowship 130 35 165 79 <10 yearse experience 299 40 339 88 ≥10 years experience 605 185 790 77 <75% spine practice 156 40 196 80 ≥75% spine practice 837 185 1022 82 Operative Technique Overall 1008 621 1629 62 Fellowship 901 493 1394 65 No Fellowship 101 118 219 46 <10 yearse experience 388 178 566 69 ≥10 years experience 529 339 868 61 <75% spine practice 147 128 275 53 ≥75% spine practice 859 483 1342 64 Surgical Indications Overall 1290 742 2032 63 Fellowship 1146 592 1738 66 No Fellowship 136 138 274 50 <10 yearse experience 499 206 705 71 ≥10 years experience 664 419 1083 61 <75% spine practice 191 153 344 56 ≥75% spine practice 1096 579 1675 65 Clinical Evaluation Overall 1618 436 2054 79 Fellowship 1406 353 1759 80 No Fellowship 196 79 275 71 <10 yearse experience 520 103 623 83 ≥10 years experience 947 284 1231 77 <75% spine practice 246 88 334 74 P a g e | 34
Figure 1: Example question (with responses)
P a g e | 35
Figure 2: Overall correct response rates of orthopaedic spine surgeons comparing different demographic groups. Y= Yes; N=No.
Correct Response Rates
80
p=0.000003
75 p=0.000029 p<0.00001
70
% correct 65 Y responses N
60
55
50 Fellowship <10 yr ≥75% P a g e | 36
Figure 3A-E: Correct response rates for each of the question subcategories comparing different demographic groups. Y=Yes; N=No.
Radiology/Spinopelvic Alignment
80 p=0.000002 p=0.006 p=0.001127
75
70
% Correct 65 Y Responses N 60
55
50 Fellowship <10 yr ≥75%
HRQOL
90 p=0.498238 p=0.000007 p=0.019656
85
% Correct 80 Y Responses N
75
70 Fellowship <10 yr ≥75% P a g e | 37
Operative Technique
70 p<0.000001 p=0.003365 p=0.001 65
60
55
% Correct 50 Y Responses N 45
40
35
30 Fellowship <10 yr ≥75%
Surgical Indications
75 p=0.000041 70 p<0.000001 p=0.000497 65
60
55 % Correct Y Responses 50 N
45
40
35
30 Fellowship <10 yr ≥75% P a g e | 38
Clinical Evaluation
85 p=0.001069 p=0.00109 5 p=0.00624 7 80
75
% Correct Y Responses N 70
65
60 Fellowship <10 yr ≥75%
P a g e | 39
References
1. Daniels AH, DePasse JM, Magill ST, Fischer SA, Palumbo MA, Ames CP, et al. The Current State of United States Spine Surgery Training: A Survey of Residency and Spine Fellowship Program Directors. Spine Deformity.2(3):176-85.
2. NASS. Spine Fellowship Directory. https://www.spine.org/Documents/PolicyPractice/FellowshipDirectory.pdf. Accessed 3/23/2017.
3. Schwab F, Dubey A, Gamez L, El Fegoun AB, Hwang K, Pagala M, et al. Adult scoliosis: prevalence, SF-36, and nutritional parameters in an elderly volunteer population. Spine (Phila Pa 1976). 2005;30(9):1082-5.
4. Bess S, Boachie-Adjei O, Burton D, Cunningham M, Shaffrey C, Shelokov A, et al. Pain and disability determine treatment modality for older patients with adult scoliosis, while deformity guides treatment for younger patients. Spine (Phila Pa 1976). 2009;34(20):2186-90.
5. Schwab F, Lafage V, Farcy JP, Bridwell K, Glassman S, Ondra S, et al. Surgical rates and operative outcome analysis in thoracolumbar and lumbar major adult scoliosis: application of the new adult deformity classification. Spine (Phila Pa 1976). 2007;32(24):2723-30.
6. Schwab F, Patel A, Ungar B, Farcy JP, Lafage V. Adult spinal deformity-postoperative standing imbalance: how much can you tolerate? An overview of key parameters in assessing alignment and planning corrective surgery. Spine (Phila Pa 1976). 2010;35(25):2224-31.
7. Ames CP, Smith JS, Scheer JK, Bess S, Bederman SS, Deviren V, et al. Impact of spinopelvic alignment on decision making in deformity surgery in adults: A review. J Neurosurg Spine. 2012;16(6):547-64.
8. Moshirfar A, Rand FF, Sponseller PD, Parazin SJ, Khanna AJ, Kebaish KM, et al. Pelvic fixation in spine surgery. Historical overview, indications, biomechanical relevance, and current techniques. J Bone Joint Surg Am. 2005;87 Suppl 2:89-106.
9. Clark AJ, Garcia RM, Keefe MK, Koski TR, Rosner MK, Smith JS, et al. Results of the AANS membership survey of adult spinal deformity knowledge: impact of training, practice experience, and assessment of potential areas for improved education: Clinical article. J Neurosurg Spine. 2014;21(4):640-7.
10. Herkowitz HN, Connolly PJ, Gundry CR, Varlotta GP, Zdeblick TA, Truumees E. Resident and fellowship guidelines: educational guidelines for resident training in spinal surgery. Spine (Phila Pa 1976). 2000;25(20):2703-7. 11. Charosky S, Guigui P, Blamoutier A, Roussouly P, Chopin D. Complications and risk factors of primary adult scoliosis surgery: a multicenter study of 306 patients. Spine (Phila Pa 1976). 2012;37(8):693-700.
12. Cho SK, Bridwell KH, Lenke LG, Yi JS, Pahys JM, Zebala LP, et al. Major complications in revision adult deformity surgery: risk factors and clinical outcomes with 2- to 7-year follow-up. Spine (Phila Pa 1976). 2012;37(6):489-500.
13. Drazin D, Shirzadi A, Rosner J, Eboli P, Safee M, Baron EM, et al. Complications and outcomes after spinal deformity surgery in the elderly: review of the existing literature and future directions. Neurosurg Focus. 2011;31(4):E3.
14. Daniels AH, Ames CP, Smith JS, Hart RA. Variability in spine surgery procedures performed during orthopaedic and neurological surgery residency training: an analysis of ACGME case log data. J Bone Joint Surg Am. 2014;96(23):e196.
15. Bridwell KH, Berven S, Edwards C, 2nd, Glassman S, Hamill C, Schwab F. The problems and limitations of applying evidence-based medicine to primary surgical treatment of adult spinal deformity. Spine (Phila Pa 1976). 2007;32(19 Suppl):S135-
16. Bridwell KH, Berven S, Glassman S, Hamill C, Horton WC, 3rd, Lenke LG, et al. Is the SRS-22 instrument responsive to change in adult scoliosis patients having primary spinal deformity surgery? Spine (Phila Pa 1976). 2007;32(20):2220-5.
17. Hahn T, Kelly C, Murphy E, Whissel P, Brown M, Schenk R. Clinical decision- making in the management of cervical spine derangement: a case study survey using a patient vignette. J Man Manip Ther. 2014;22(2):115-21.
18. Mayer JE, Dang RP, Duarte Prieto GF, Cho SK, Qureshi SA, Hecht AC. Analysis of the techniques for thoracic- and lumbar-level localization during posterior spine surgery and the occurrence of wrong-level surgery: results from a national survey. Spine J. 2014;14(5):741- 8.
19. Daniels AH, Ames CP, Garfin SR, Shaffrey CI, Riew KD, Smith JS, Anderson PA, Hart RA. Spine surgery training: is it time to consider categorical spine surgery residency? Spine J. 2015 Jul 1;15(7):1513-8.
P a g e | 41
Anterior Versus Posterior Revision Total Hip Arthroplasty: A Comparative Analysis of Complications
Gregory Kurkis, MD; Samir Chihab, MD; Kevin X Farley, BS; Albert Anastasio, MD; Thomas L Bradbury, MD; George N Guild, MD
P a g e | 42
ABSTRACT RESULTS The DAA cohort demonstrated an increased risk of BACKGROUND superficial wound complications (7.1% versus Wound complication after primary direct anterior 0.5%, p=0.003) and a decreased dislocation rate (DAA) hip replacement has been reported in the (2.0% versus 13.1%, p=0.002). There was a trend literature. Additionally, the DAA has demonstrated towards increased overall complications in the PA a lower dislocation rate compared to posterolateral group (OR 1.71, p=0.078). approach (PA) in primary THA. There has been no comparison in the literature regarding wound CONCLUSIONS complications and/or dislocation after revision Revision DAA THA is associated with an anterior versus revision posterior hip replacement. increased risk of superficial wound complications The authors hypothesize that anterior approach but may also impart a decreased dislocation rate. revision surgery may have increased wound complications and decreased dislocations compared to posterior hip revisions and also report Level of Evidence: III - Retrospective Cohort on secondary outcome metrics. Study
METHODS Keywords: Revision, Total Hip Arthroplasty, 99 DAA and 191 PA revisions were included for Complications, Dislocation, Wound- analysis. Preoperative demographic characteristics, indication for revision, operative details, type of Complications, Approach, Direct Anterior revision performed, components utilized, and Approach postoperative complications were compared between DAA and PA groups including multivariate analysis. INTRODUCTION Postoperative wound-healing complications 2.1% in their series of 48 revision DAA THA cases have been documented after primary direct anterior [5]. Kennon et al., found 7 wound hematomas in a approach (DAA) total hip arthroplasty (THA) and series of 468 DAA revision THA, but did not occur in 1-11.5% of primary anterior surgeries [1- report on all superficial wound-healing
4]. The incidence of wound-healing complications complications [6]. Although longer-term patient- after revision THA performed via a DAA has been reported outcomes have shown no difference minimally reported in the literature, with no between DAA and PA for primary THA, some comparative studies between DAA and posterior research has suggested an early postoperative approach (PA). Nogler et al. documented an benefit in anterior approach hip replacement, incidence of superficial wound healing issues of which remains a popular approach [7-10]. There P a g e | 43 may be an early benefit in DAA hips, but a notable series of 51 and 48 patients respectively [24,5]. concern is the increased risk of postoperative Cogan et al. performed 73 isolated acetabular wound complications, particularly for obese or component revision surgeries via a DAA and diabetic patients [11]. found a 6.6% dislocation rate [25].
Additionally, postoperative dislocation is While less commonly used than the PA for the most common complication encountered after revision THA, revision DAA surgery has been revision total hip arthroplasty (THA) [12,13]. described and may be an option for some patients
Reported dislocation rates after revision THA are [26]. The purpose of this study is to determine if high, ranging from 5.68% to 28% the anterior approach portends to differences in
[12,13,14,15,16,17,18,19]. The majority of these complication rates, specifically increased wound dislocations occur in the early postoperative complications or decreased dislocations in revision period, within 90 days of the surgical procedure total hip surgery. To the author’s knowledge, no
[20,21]. In primary THA, studies have shown previous studies have compared complications, decreased dislocation rates with a direct anterior including wound issues and dislocation, after approach (DAA) compared to a posterolateral revision THA performed via a DAA versus PA and approach (PA) [22,23]. There have been few case hypothesize that revision surgery performed via a series published on revision DAA THA DAA will result in more superficial wound
[6,24,5,25]. Kennon et al. showed a 3.0% complications and a lower dislocation rate. We dislocation rate in a single surgeon experience of also report on 90-day complications amongst
468 revision THA procedures performed via a groups.
DAA, with 6-month minimum follow-up [6]. Mast METHODS et al. and Nogler et al. both reported no dislocation After obtaining approval from our events after revision THA via the DAA in case Institutional Review Board, a retrospective review P a g e | 44 of all revision THA procedures performed by two was collected, including age, gender, medical surgeons at our institution over an eight-year comorbidities, BMI, tobacco usage, and period (2011-2019) was conducted. A minimum preoperative opioid use. Prior approach and
90-day postoperative follow-up was required for surgical indication for the index THA were noted, inclusion in the study. All revision surgeries were as well as, preoperative diagnosis and indication performed by two fellowship-trained adult for revision surgery. reconstructive surgeons. Of note, both surgeons Operative details were recorded for each perform high annual volumes of revision anterior patient, including type of revision procedure hip replacement. Decision-making for type of performed, operative time, estimated blood loss, approach used was based on prior primary design-type of components utilized, and use of an approach and acetabular bone loss pattern. Medical extended trochanteric osteotomy. The type of comorbidities, preoperative physical examination, revision procedure performed was classified as 1) patient demographic variables, indication for isolated acetabular component revision, 2) isolated revision, and type of revision surgery were femoral component revision, 3) combined generally not used in surgical approach decision acetabular and femoral component revisions, or 4) making. All surgeries were performed at the same head-ball and polyethylene liner exchange. orthopaedic specialty hospital, which utilizes Additionally, the design-type of components protocol-driven pathways in perioperative care. utilized were also recorded for the acetabular
Anterior revision surgeries were done via a Smith- component (use of augment), femoral component
Peterson interval in the supine position. Posterior (diaphyseal-fit revision versus regular stem), and approach revisions were conducted through a liner (standard, walled, versus dual mobility).
Kocher-Langenbeck interval in a lateral decubitus Postoperative 90-day complications, in- position. Baseline preoperative demographic data hospital complications, medical complications, P a g e | 45 postoperative transfusion requirement, need for mass index (BMI) categories (≤29, ≥30), and readmission within 90-days after surgery, and revision surgery type (femoral component reason for readmission were recorded. replacement, acetabular component replacement,
Complications specifically analyzed within 90- both femoral and acetabular component days postoperatively included superficial wound replacement, or a head-ball and liner exchange). dehiscence, total wound complications (including Furthermore, type of acetabular component insert superficial wound cellulitis as well as (neutral, walled/raised, or modular dual mobility) noninfectious wound complications), dislocation, and presence of preoperative hip instability were surgical site infection, hematoma formation, and included in the regression model that identified component loosening or complication. independent risk factors for dislocations. All
Patients were separated into DAA or PA analyses were performed with SPSS software revision cohorts for analysis. Nonparametric (IBM Corp, Armonk, NY). A priori power analysis analysis was performed using Mann–Whitney U was calculated using an effect size of .5, power .8 tests and Kruskal–Wallis tests for continuous and p<.05 with minimum sample size per group at variables. Chi-square or Fischer-Exact was used 70. for categorical variables, as appropriate. To RESULTS identify anterior or posterior revision surgery as 290 revision surgeries were included in this independent risk factors for complications and study. 99 revisions were done via a DAA and 191 readmissions, a binomial logistic regression by a PA. There was no difference between the analysis was used to control for potential DAA and PA groups in gender or medical confounders. Confounders entered into the comorbidities. The PA group was slightly older regression analysis included age categories (≤44 with an average age of 66 versus 63 years for the yrs, 45-64 yrs, 65-84 yrs, and >85 yrs), sex, body DAA patients (p = 0.037). The DAA cohort had a P a g e | 46 higher average BMI (31.5) compared to the the DAA group and only 2.1% of the PA group posterior group (29.0, p<.001). The majority, 84%, (p=.007). The indications for revision surgery in of patients who underwent DAA revision had a the DAA and PA groups are summarized in Table prior DAA for their initial primary arthroplasty 2. procedure. Conversely, for patients that underwent There was no difference in the proportions
PA revision THA, 77% had prior posterior or of isolated acetabular or isolated femoral lateral approach for their primary procedure. There component revision between the DAA and PA was no significant difference in the presence of groups. However, combined acetabular and preoperative hip instability between the two groups femoral component revisions were performed more
(9.1% in DAA versus 14.1% in PA, p=0.22). frequently in the PA compared to DAA group
Patient demographic and preoperative (33.5% versus 18.2%, p=.006). In contrast, head- characteristics are summarized in Table 1. ball and liner exchange procedures were more
Preoperative indications for revision common in the DAA group (24.2% vs 8.9%, surgery were overall similar between the two p<.001). The majority of cases utilized tapered groups. There was no significant difference in the modular titanium stems (TMFT) for femoral proportions of patients that underwent revision for fixation. Acetabular augmentation use was greater component loosening, limb length inequality, in the PA group (12.2% versus 6.1%). Type of prosthetic joint infection, periprosthetic fracture, acetabular liner used was not significantly different recurrent instability/dislocation, acetabular between the two groups. Use of an extended component malposition/migration, or conversion trochanteric osteotomy was infrequent and not from prior hemiarthroplasty. The DAA group had significantly different between the two groups. an increased incidence of revisions done for Notably, hospital length of stay trended shorter in polyethylene liner failure, accounting for 10.1% of the DAA group compared to the PA group by 0.2 P a g e | 47 days (3.2 versus 3.0 days, p=.09). Operative details 120.97, p=0.019) and PA had increased odds ratio and postoperative hospital length of stay are of periprosthetic dislocation (OR 7.49, 95% CI summarized in Table 3. 1.62-34.69, p=0.010) (Table 5).
Univariate analysis revealed the overall 90- DISCUSSION day complication rate was lower in the DAA This study demonstrates a higher rate of group, with 22.2% versus 35.3% of PA patients superficial wound complications with revision experiencing a complication (p=0.023). When THA performed via a DAA compared to PA, comparing individual complications, there was no which persisted after multivariate regression difference in deep surgical site infection, 5.1% in analysis. Additionally, this study also demonstrates
DAA and 5.2% in PA (p=0.95). A significantly a decreased risk of dislocation with the DAA increased incidence of superficial wound compared to PA revision surgery. There was a complications, including both noninfectious trend towards decreased overall 90-day dehiscence and superficial infection, was noted in complications in DAA group, with no difference in the DAA group, 7.1% versus 0.5% (p=0.003). 90-day surgical site infection. This is the first
There was a significantly higher dislocation rate in comparative study to report on wound the PA group (13.1%) compared to the DAA group complications in the revision population and is
(2.0%) (p=.002). The 90-day readmission rate was consistent with prior research that has identified an not significantly different between the groups increased risk of wound healing issues in primary
(9.1% in DAA versus 8.9% in PA, p=0.96) (Table DAA THA [1-3]. Prior studies have demonstrated
4). that both obesity and diabetes are associated with
Multivariate analysis revealed that DAA postoperative wound healing complications patients had an increased odds ratio of superficial requiring reoperation in DAA THA [2]. The wound dehiscence (OR 13.65, 95% CI 1.54- incidence of postoperative primary DAA wound P a g e | 48 complications reported in the literature ranges from reducing post-operative instability after posterior
1.0% to 11.5% [2,3], and the rate of postoperative approach primary total hip arthroplasty has been wound complications seen in this revision study well documented. It stands to reason that capsular falls within this observed range. It should also be repair after revision posterior total hip arthroplasty noted that the DAA group in this study had a would be equally or more important than with slightly higher BMI (31.5 vs 29.0) that could have primary surgery. However, revision surgery is contributed to the increased incidence of wound- typically associated with more expansile surgical healing issues encountered; however, this was exposures which may comprise the ability to controlled for by multivariate analysis. Although a successfully repair the posterior capsule. It is statistically significant difference in BMI between possible that the higher instability rate after groups on univariate analysis was found, it likely revision posterior THA in our study is related to a does not meet the minimal clinically important reduction in the ability to re-establish the integrity difference (MCID) for obesity. The deep surgical of the posterior capsule. In addition, cadaveric site infection rates seen in our study of 5.1% and evidence suggests a difference in muscle damage
5.2% for the DAA and posterior groups patterns between anterior and posterior approaches respectively, are in keeping with previously during primary THA. The anterior approach may published infection rates after revision THA, be protective of the abductors at the cost of which range from 1.35% to 7.84% [27,28]. damage to the tensor fascia lata muscle. This is
As a secondary outcome metric, this study consistent with the authors’ intraoperative demonstrates a decreased risk of postoperative observations during femoral component revision dislocation with DAA compared to PA revision using both approaches. The combination of surgery that persisted after multivariate regression external rotation and hip extension during femoral analysis. The importance of capsular repair in instrumentation with the anterior approach may P a g e | 49 provide some protection of the abductors at their idiosyncrasies are present in any revision hip confluence at the proximal aspect of the greater comparison manuscript and must be accounted for trochanter. Such damage may influence stability if by multivariate analysis. The multivariate analysis it results in compromise of abductor muscle controlled for potential confounders and showed function. However, abductor muscle damage was persistent statistical significance in superficial not evaluated with post-operative imaging in our wound complications and dislocation rates study and the clinical significance of muscle between the two groups, where other significance damage patterns between the two approaches in on univariate analysis was lost. unknown. Although there was a statistically significant decrease in the overall 90-day complication rate in the DAA group on univariate analysis (p=.023), the significance was lost on multivariate analysis.
Limitations of this study include the inherent variability in revision hip surgery that can exist between seemingly similar patients. The PA group is on average 3 years older than the DAA group, had a greater proportion of revisions done for metal-on-metal/ALTR, yet there were more head/liner exchanges in the DAA group. One may argue that failed MOM hip revision portends to increased dislocation, but the same argument can be made for head/liner exchanges. These types of P a g e | 50
CONCLUSIONS
This manuscript confirms the authors’ hypothesis that revision THA performed via the
DAA has an increased odds ratio of superficial wound-healing complications, but may be associated with lower postoperative dislocation rate compared to PA revision surgery. This is the first comparative study to examine revision DAA and PA outcomes to the authors’ knowledge.
This study should not be used to justify the superiority of one approach over another in revision surgery; instead, it provides the arthroplasty surgeon with additional information to make an informed decision when considering the surgical approach during preoperative planning for revision THA. Future work should utilize prospective randomized models to evaluate differences between revision THA approaches. P a g e | 51
Table 1. Descriptive Statistics of Patient Characteristics by Surgical Approach
Patient Overall (n=290) Direct Anterior Posterolateral P-value Demographics Approach (n=99) Approach (n=191) Age (years) 64.9 12.6 62.9 12.1 65.9 12.7 0.037$ BMI (kg/m2) 29.8 8.1 31.5 6.4 29.0 8.8 < 0.001$ Gender (% Female) 159/290 (54.8%) 47/99 (47.5%) 112/191 (58.6%) 0.07* Medical 221/290 (76.5%) 76/99 (76.8%) 145/191 (76.3%) 0.93* Comorbidities Preoperative Hip 36/290 (12.4%) 9/99 (9.1%) 27/191 (14.1%) 0.22* Instability Continuous statistics: mean SD; Categorical statistics: n/N (%); $ Wilcoxon Rank Sum Test; * Chi-square test
Table 2. Indications for Revision Surgery by Surgical Approach
Indication for Revision Overall (n=290) Direct Anterior Posterolateral P-value Approach (n=99) Approach (n=191) Loose component 110/290 (37.9%) 33/99 (33.3%) 77/191 (40.3%) 0.25* Metal-on- 36/290 (12.4%) 6/99 (6.1%) 30/191 (15.7%) 0.018* Metal/ALVAL Psoas impingement 10/290 (3.4%) 9/99 (9.1%) 1/191 (0.5%) <.001# Limb-Length 8/290 (2.8%) 5/99 (5.1%) 3/191 (1.6%) 0.13# Inequality Prosthetic Joint 46/290 (15.9%) 14/99 (14.1%) 32/191 (16.8%) 0.56* Infection Periprosthetic Fracture 46/290 (15.9%) 13/99 (13.1%) 33/191 (17.3%) 0.36* Recurrent Instability 36/290 (12.4%) 9/99 (9.1%) 27/191 (14.1%) 0.22* Hemiarthroplasty 10/290 (3.4%) 2/99 (2.0%) 8/191 (4.2%) 0.50# Conversion Ceramic Fracture 2/290 (0.7%) 1/99 (1.0%) 1/191 (0.5%) 1.00# Acetabular Component 19/290 (6.6%) 3/99 (3.0%) 16/191 (8.4%) 0.08* Malposition Polyethylene Wear / 24/290 (8.3%) 8/99 (8.1%) 16/191 (8.4%) 0.93* Osteolysis Failed Liner 14/290 (4.8%) 10/99 (10.1%) 4/191 (2.1%) 0.007# Heterotopic 1/290 (0.3%) 0/290 (0%) 1/191 (0.5%) 1.00# Ossification Categorical statistics: n/N (%); # Fisher’s Exact Test; * Chi-square test
P a g e | 52
Table 3. Descriptive Statistics of Operative Characteristics by Surgical Approach
Operative Overall (n=290) Direct Anterior Posterolateral P-value Characteristics Approach (n=99) Approach (n=191) Estimated Blood 588 541 469 445 649 576 <.001$ Loss (mL) Operative Time 153 55 141 57 159 53 0.002$ (min) Length of Stay 3.1 2.6 3.0 2.7 3.2 2.6 0.09$ (days) Type of Revision Acetabulum Only 92/290 (31.7%) 27/99 (27.3%) 65/191 (34.0%) 0.24* Femur Only 73/290 (25.2%) 29/99 (29.3%) 44/191 (23.0%) 0.24* Acetabulum & 82/290 (28.3%) 18/99 (18.2%) 64/191 (33.5%) 0.006* Femur Head-Ball/Liner 41/290 (14.1%) 24/99 (24.2%) 17/191 (8.9%) <.001* Exchange ORIF Fracture 10/290 (3.4%) 5/99 (5.1%) 5/191 (2.6%) 0.32# Femoral Component Non-modular 17/99 (17.2%) 13/190 (6.8%) 0.011* Proximal 30/289 (10.4%) Metaphyseal-Fit Modular Tapered 111/289 (38.4%) 29/99 (29.3%) 82/190 (43.2%) 0.030* Diaphyseal-Fit Monoblock 12/289 (4.2%) 0.70* Tapered 3/99 (3.0%) 9/190 (4.7%) Diaphyseal-Fit Acetabular Component Augment Used 9/286 (10.1%) 6/98 (6.1%) 23/188 (12.2%) <.001* Liner Type Standard 174/256 (68.0%) 58/87 (66.7%) 116/169 (68.6%) 0.73* Walled/raised 47/256 (18.4%) 15/87 (17.2%) 32/169 (18.9%) Constrained 25/256 (9.8%) 11/87 (12.6%) 14/169 (8.3%) ETO performed 8/290 (2.8%) 1/99 (1.0%) 7/191 (3.7%) 0.27# Continuous statistics: mean SD; Categorical statistics: n/N (%); $ Wilcoxon Rank Sum Test; * Chi-square test; # Fisher’s Exact Test
P a g e | 53
Table 4. Univariate Analysis of 90-Day Postoperative Complications by Surgical Approach
Complication Overall (n=290) Direct Anterior Posterolateral P-value Approach (n=99) Approach (n=191) Any Complication 89/290 (30.8%) 22/99 (22.2%) 67/191 (35.3%) 0.023* 90-Day Readmission 26/290 (9.0%) 9/99 (9.1%) 17/191 (8.9%) 0.96* Dislocation 27/290 (9.3%) 2/99 (2.0%) 25/191 (13.1%) 0.002* Deep Surgical Site 15/290 (5.2%) 5/99 (5.1%) 10/191 (5.2%) 0.95* Infection Superficial Wound 8/290 (2.8%) 7/99 (7.1%) 1/191 (0.5%) 0.003# Complication Acetabular Component 5/290 (1.7%) 1/99 (1.0%) 4/191 (2.1%) 0.66# Loosening Femoral Component 2/290 (0.7%) 0/99 (0%) 2/191 (1.0%) 0.55# Subsidence Limb Length Inequality 6/290 (2.1%) 0/99 (0%) 6/191 (3.1%) 0.10# Periprosthetic fracture 2/290 (0.7%) 0/99 (0%) 2/191 (1.0%) 0.55# Categorical statistics: n/N (%);* Chi-square test; # Fisher’s Exact Test
Table 5. Multivariate Analysis of 90-Day Postoperative Complication, Readmission or Dislocation by Surgical Approach
Complication Odds Ratio (PA 95% Confidence P-value vs DAA) Interval Any Complication 1.71 0.94 – 3.10 0.078 90-Day Readmission 1.02 0.40 – 2.56 0.970 Superficial wound 13.65* 1.54-120.97 0.019 Dislocation 7.49 1.62 – 34.69 0.010 *Odds ratio for superficial wound complication presented as DAA vs PA P a g e | 54
REFERENCES
[1] Christensen CP, Karthikeyan T, Jacobs CA. Greater prevalence of wound complications requiring reoperation with direct anterior approach total hip arthroplasty. J Arthroplasty. 2014 Sep;29(9):1839-41. [2] Jahng KH, Bas MA, Rodriguez JA, Cooper HJ. Risk Factors for Wound Complications After Direct Anterior Approach Hip Arthroplasty. J Arthroplasty. 2016 Nov;31(11):2583- 2587. [3] Purcell RL, Parks NL, Cody JP, Hamilton WG. Comparison of Wound Complications and Deep Infections With Direct Anterior and Posterior Approaches in Obese Hip Arthroplasty Patients. J Arthroplasty. 2018 Jan;33(1):220-223. [4] Statz JM, Duethman NC, Trousdale RT, Taunton MJ. Outcome of Direct Anterior Total Hip Arthroplasty Complicated by Superficial Wound Dehiscence Requiring Irrigation and Debridement. J Arthroplasty. 2019 Jul;34(7):1492-1497. [5] Nogler M, Mayr E, Krismer M. [The direct anterior approach to the hip revision]. Oper Orthop Traumatol. 2012 Apr;24(2):153-64. [6] Kennon R, Keggi J, Zatorski LE, Keggi KJ. Anterior approach for total hip arthroplasty: beyond the minimally invasive technique. J Bone Joint Surg Am. 2004;86-A Suppl 2:91-7. [7] 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. [8] Peters RM, van Beers LWAH, van Steenbergen LN, Wolkenfelt J, Ettema HB, Ten Have BLEF, Rijk PC, Stevens M, Bulstra SK, Poolman RW, Zijlstra WP. Similar Superior Patient-Reported Outcome Measures for Anterior and Posterolateral Approaches After Total Hip Arthroplasty: Postoperative Patient-Reported Outcome Measure Improvement After 3 months in 12,774 Primary Total Hip Arthroplasties Using the Anterior, Anterolateral, Straight Lateral, or Posterolateral Approach. J Arthroplasty. 2018 Jun;33(6):1786-1793. [9] Miller LE, Gondusky JS, Bhattacharyya S, Kamath AF, Boettner F, Wright J. Does Surgical Approach Affect Outcomes in Total Hip Arthroplasty Through 90 Days of
P a g e | 55
Follow-Up? A Systematic Review With Meta-Analysis. J Arthroplasty. 2018 Apr;33(4):1296-1302 [10] Post Z, Orozco F, Diaz-Ledezma C, Hozack W, Ong A: Direct Anterior Approach for Total Hip Arthroplasty: Indications, Techniques, and Results. JAAOS 2014; 22 (9): 595- 603. [11] Jahng KH, Bas MA, Rodriguez JA, Cooper HJ. Risk Factors for Wound Complications After Direct Anterior Approach Hip Arthroplasty. J Arthroplasty. 2016 Nov;31(11):2583- 2587. [12] Springer B, Fehring T, Griffin W, Odum S, Masonis J: Why Revision Total Hip Arthroplasty Fails. Clinical Orthopaedics and Related Research 2009; 467 (1): 166-173. [13] Carter A, Sheehan E, Mortazavi S, Purtill J, Sharkey P, Parvizi J. Revision for Recurrent Instablity: What are the Predictors of Failure?. Journal of Arthroplasty. 2011; 26(6): 46-52. [14] Wetters N, Murray T, Moric M, Sporer S, Paprosky W, Della Valle C: Risk Factors for Dislocation After Revision Total Hip Arthroplasty. Clinical Orthopaedics and Related Research 2013; 471 (2): 410-416. [15] Alberton G, High W, Morrey B: Dislocation After Revision Total Hip Arthroplasty. Journal of Bone and Joint Surgery 2002; 84A (10): 1788-1792. [16] Carter AH, Sheehan EC, Mortazavi SM, Purtill JJ, Sharkey PF, Parvizi J. Revision for recurrent instability: what are the predictors of failure? J Arthroplasty. 2011 Sep;26(6 Suppl):46-52. [17] Berend KR, Sporer SM, Sierra RJ, Glassman AH, Morris MJ. Achieving stability and lower limb length in total hip arthroplasty. Instr Course Lect. 2011;60:229-46. [18] Faldini C, Stefanini N, Fenga D, Neonakis E, Perna F, Mazzotti A, Pilla F, Triantafyllopoulos K, Traina F. How to prevent dislocation after revision total hip arthroplasty: a systematic review of the risk factors and a focus on treatment options. Journal of Orthopaedics and Traumatology. 2018; 19 (17). [19] Kosashvili Y, Backstein D, Safir O, Lakstein D, Gross AE. Dislocation and infection after revision total hip arthroplasty: comparison between the first and multiply revised total hip arthroplasty. J Arthroplasty. 2011 Dec;26(8):1170-5. [20] Meek RM, Allan DB, McPhillips G, Kerr L, Howie CR. Epidemiology of dislocation after total hip arthroplasty. Clin Orthop Relat Res. 2006 Jun;447:9-18.
P a g e | 56
[21] Jolles BM, Zangger P, Leyvraz PF. Factors predisposing to dislocation after primary total hip arthroplasty: a multivariate analysis. J Arthroplasty. 2002 Apr;17(3):282-8. [22] Sheth D, Cafri G, Inacio MC, Paxton EW, Namba RS. Anterior and Anterolateral Approaches for THA Are Associated With Lower Dislocation Risk Without Higher Revision Risk. Clin Orthop Relat Res. 2015 Nov;473(11):3401-8. [23] Zijlstra WP, De Hartog B, Van Steenbergen LN, Scheurs BW, Nelissen RGHH. Effect of femoral head size and surgical approach on risk of revision for dislocation after total hip arthroplasty. Acta Orthop. 2017 Aug;88(4):395-401. [24] Mast NH, Laude F. Revision total hip arthroplasty performed through the Hueter interval. J Bone Joint Surg Am. 2011 May;93 Suppl 2:143-8. [25] Cogan A, Klouche S, Mamoudy P, Sariali E. Total hip arthroplasty dislocation rate following isolated cup revision using Hueter's direct anterior approach on a fracture table. Orthop Traumatol Surg Res. 2011 Sep;97(5):501-5. [26] Nogler MM, Thaler MR. The Direct Anterior Approach for Hip Revision: Accessing the Entire Femoral Diaphysis Without Endangering the Nerve Supply. J Arthroplasty. 2017 Feb;32(2):510-514. [27] Kosashvili Y, Backstein D, Safir O, Lakstein D, Gross AE. Dislocation and infection after revision total hip arthroplasty: comparison between the first and multiply revised total hip arthroplasty. J Arthroplasty. 2011 Dec;26(8):1170-5. [28] Jafari SM, Coyle C, Mortazavi SMJ, Sharkey PF, Parvizi J. Revision Hip Arthroplasty: Infection is the Most Common Cause of Failure. Clinical Orthopaedics and Related Research. 2010;468(8):2046-2051.
P a g e | 57
Current Trends in Clinical Practice for the Direct Anterior Approach Total Hip Arthroplasty
Nick N. Patel, MD; Jason A. Shah, BS; Greg A. Erens, MD
P a g e | 58
ABSTRACT patient market share. Only 3.0% of current non- performers plan to adopt the DAA in the future.
Background Despite increased popularity of the direct Conclusion anterior approach (DAA) for total hip This is the first study of its kind to arthroplasty (THA), current practice trends and highlight current trends and clinical practices specific driving factors leading to adoption are from a surgeon perspective regarding the DAA. not well established. Specifically, it provides comprehensive data regarding perceptions of clinical outcomes, Methods practice economics, and driving factors for We conducted an electronic e-mail survey choice of surgical approach for surgeons who do of members of the American Association of Hip and do not perform the DAA. and Knee Surgeons inquiring into the choice of THA surgical approach, perceptions of clinical Level of Evidence: IV outcomes, and economic implications associated Keywords with the DAA. Direct anterior approach; total hip arthroplasty; total hip approach; surgical preference; surgical Results approach
Of 996 total respondents (44.3% American Association of Hip and Knee Surgeons INTRODUCTION member response rate), 56.2% currently perform The evolution of surgical techniques and the DAA. DAA performers have been in practice implant design has led total hip arthroplasty for statistically less time than non-performers (17.0 years vs 20.9 years, P < .001). Similarly, (THA) to be one of the most successful high-volume DAA surgeons have been in orthopedic procedures [1-3]. Recent increased practice for less time than low-volume surgeons. emphasis on improving patient outcomes and DAA performers felt that revision case status (79.3%), complex anatomy (65.0%), and body reducing healthcare costs has propelled THA habitus (53.0%) were factors leading to surgical approaches into the forefront of current preferential use of the posterior approach. We discussion. Despite significant literature on the also provide comprehensive data for perceived outcomes comparing the DAA and posterior matter, much controversy continues to exist approach. For current non-performers, the top regarding the most effective THA surgical reasons for not utilizing the DAA were feelings approach [4-9]. The direct anterior approach of worse outcomes, no clinical benefit, and (DAA) for THA has gained popularity in recent concern for the learning curve. Economically, 76.1% of DAA performers reported increased years primarily due to claims of reduced surgical patient market share by performing the DAA trauma, shorter recovery time, and improved while 65.8% of non-performers endorsed lost patient outcomes [10-15]. Furthermore,
P a g e | 59 marketing and promotion by industry, hospitals, 2249 active AAHKS members. The full survey and surgeons alike have played a substantial role can be found in the Appendix. All survey takers in the DAA’s expanding adoption [16-18]. were asked about the length of time being in However, the DAA has been associated with a practice and whether he or she had completed an steep learning curve including increased revision arthroplasty fellowship. Additionally, everyone rate and overall rate of complication [19-23]. was asked whether they performed DAA THA; Additionally, limited literature exists establishing however, no further inquiry was made into other the long-term benefits when compared to other THA approaches utilized. DAA THA performers surgical techniques. As a result, some authors and non-performers were each subsequently suggest choosing a surgical approach for THA directed to a separate series of questions. DAA based on surgeon preference, experience, and performers were asked for quantification of their patient-specific considerations [24]. Although annual DAA THA case count, the percentage of there is emerging literature on the clinical their THA patients requesting the DAA, and if outcomes and even patient perceptions associated they felt performing the DAA led to increased with DAA THA, current practice trends and patient market share. Using a 3-level Likert specific driving factors leading to adoption by scale, DAA performing surgeons were asked to arthroplasty surgeons are not well established compare the DAA to the posterior approach (PA) [4,25]. Through this investigation, we aim to for THA in regards to the following categories: identify surgeon preferences for the DAA, in blood loss, operative time, post-operative pain, particular those related to clinical outcomes, hospital length of stay, rate of discharge home vs patient perceptions, and practice economics. rehab facility, dislocation rate, femoral fracture Furthermore, we hope to gain insight into rate, infection rate, overall complication rate, practice trends and factors that drive the choice postoperative functional status, leg length of surgical approach for those who do and do not equality, and accuracy of acetabular positioning. utilize the DAA. Finally, DAA performers were asked to select all factors that would lead them to perform a PA METHODS THA when they otherwise would have utilized In May 2018, an electronic survey that the DAA. An “other” option was available for was approved by the American Association of additional considerations as well. DAA non- Hip and Knee Surgeons (AAHKS) research performers were first asked to select the single committee was distributed via e-mail link to all most important reason for not performing the
P a g e | 60
DAA THA. An “other” option was available for cases annually (P = .04 and 0.05, respectively). manual entry of a factor not listed. This cohort of In addition, 43% of responding DAA surgeons surgeons was finally asked if they plan to claimed that over 50% of their THA patients perform DAA THA in the future, and if they felt request DAA (Fig. 1). When asked about not performing the DAA has led to patients being considerations that would lead DAA performers lost to other surgeons. In terms of statistical to use an alternative approach, respondents analysis, a 2-tailed Student’s t-test was utilized to identified revision case (79.3%), complex compare numerical means. An N-1 two- anatomy (65.0%), and body habitus (53.0%) proportion test was implemented to compare respectively as the most common considerations proportions of binomial categorical variables. for use of the PA (Table 3). About 7.9% of current DAA surgeons felt that there were no RESULTS indications for the PA. Of the 2249 active AAHKS members In regards to DAA peri-operative who were e-mailed, a total of 996 individuals variables, 73.2% and 70.0% of DAA performing completed the survey resulting in a 44.3% surgeons believe that there is decreased post- response rate. An additional 10 surveys were not operative pain and length of hospital stay, entirely completed and were excluded from the respectively, when compared to PA THA. analysis. Of the 996 total respondents, 560 Moreover, 53.8% of DAA performers believe (56.2%) currently perform DAA THA while 436 that there is no difference in blood loss between (43.8%) do not. DAA performing surgeons had a the 2 approaches. In total, 40.7% of DAA significantly shorter mean time in practice (17.0 surgeons report no difference in operative time vs 20.9 years, P < .001); however, no significant compared to the PA while 35.8% believe it is difference was observed in percentage of actually increased. When compared to the PA, respondents who are arthroplasty fellowship 41.9% of DAA performers feel that the anterior trained (70.7% vs 75.7%, P = .08; Table 1). Of approach was associated with higher rate of the DAA THA performers, a majority (53.6%) discharge home vs rehab while 47.0% report no reported a DAA case volume of greater than 100 difference (Fig. 2). cases annually (Table 2). Those performing >50 Analysis of perceived complications and >100 DAA cases annually had been in demonstrated that 72.4% of DAA surgeons practice for a statistically shorter mean time believe that there is a lower dislocation rate with when compared to those performing <10 DAA the DAA when compared to the PA. In addition,
P a g e | 61
57.9% of this cohort feel that the DAA was 65.8% of DAA non-performers believe that they associated with no difference in the rate of lose patients to DAA performing surgeons. Only femoral fracture compared to the PA, while 3.0% of those currently not performing DAA, 38.2% perceive an increased fracture rate. An however, indicated a plan to adopt the anterior equal percentage of DAA surgeons believe that approach in the future (Fig. 6). Subgroup there was an increase (45.0%) or no difference analysis of DAA non-performers revealed that (45.0%) in the rate of infection or wound those who endorsed patient loss to other surgeons complication in obese patients. When had a significantly longer mean time in practice considering the overall complication rate of the compared to those who denied patient loss (21.8 DAA vs the PA, 54.9% of DAA surgeons feel vs 19.1 years, P = .005). However, no significant that no difference exists (Fig. 3). difference was appreciated between the 2 groups Regarding patient post-operative in regards to percentage of arthroplasty functional status of the DAA compared to PA, fellowship trained surgeons (74.6% vs 77.9%, P 80.3% of DAA surgeons perceive increased level = .46) (Table 4). Additional subgroup analysis of in the short term, but 85.2% feel no long-term DAA non-performers demonstrated that those difference exists. About 73.3% of DAA who plan to adopt DAA in the future have performers consider this approach to be significantly fewer years in practice compared to associated with increased accuracy of acetabular those who deny plans to adopt DAA (14.0 vs positioning compared to the PA (Fig. 4). 21.1 years, P = .01). Similarly, no significant Surgeons who do not perform DAA THA difference was appreciated between the 2 groups were asked to report the single most important in regards to percentage of arthroplasty reason for not currently performing DAA. fellowship trained surgeons (69.2% vs 76.1%, P Analysis identified belief of worse outcomes = .57) (Table 5). Of the surgeons who plan to when using DAA (30.7%), no benefit over other adopt DAA in the future, a majority indicated surgical approaches (25.7%), and concern about that the single most important barrier with DAA the learning curve (23.6%) as the 3 most is concern of the learning curve (61.5%) common reasons for not performing DAA (Fig. followed by not being trained with the procedure 5). Inquiry into economic perceptions (23.1%) (Fig. 7). demonstrated that 76.1% of DAA performing surgeons believe that they have increased patient market share due to performing DAA. Similarly,
P a g e | 62
DISCUSSION completing arthroplasty fellowships has also Although significant literature exists risen rapidly in recent years [29]. Although regarding surgical technique and clinical significant, the difference of approximately 4 outcomes of various approaches for THA, data years in practice time was less than anticipated. regarding current practices by arthroplasty This may be explained by the fact that non- surgeons are limited. This investigation is the fellowship trained surgeons routinely adopt the first, to our knowledge, to describe actual DAA throughout their careers. Similarly, our surgeon preferences and current clinical trends in investigation demonstrated no difference in the THA approaches, particularly regarding the rate of arthroplasty fellowship training between DAA. DAA performers and non-performers. It would The DAA for THA has become be interesting to see how these findings would increasingly popular for a multitude of reasons differ if a more comprehensive population of including extensive marketing and perceptions of arthroplasty surgeons could be surveyed. improved patient outcomes [10-18]. Through When considering DAA case volume, the large database analysis, Maratt et al and Sheth et majority (53.6%) of surgeons performed >100 al demonstrated an increase in percentage of DAA cases annually while only 8.9% performed THA using the DAA from 4% during 2001-2011 <10 cases annually. Those performing >100 to 17% during 2012-2014 [26, 27]. With 56.2% cases and even >50 cases annually had of currently surveyed arthroplasty surgeons significantly lower mean practice times performing the DAA and more planning to adopt compared to those performing <10 DAA THAs this technique, it is expected that the annual annually. The finding that higher volume DAA DAA THA volume will continue to rise. surgeons have been in practice for less time than The present study demonstrated that low-volume DAA surgeons may similarly be surgeons who perform DAA THA have spent explained by the timeline of popularization of the significantly fewer years in practice than those DAA. who do not. This discrepancy in practice time is The topic of case volume in relation to understandable as the DAA has become the DAA warrants discussion of the potential markedly popularized over the past 15-20 years learning curve associated with the procedure. [28]. Furthermore, not only has exposure to DAA The learning curve accompanying the DAA has become more prevalent in residency and been well documented including increased fellowship training, but the rate of surgeons complication rate, operative time, and need for
P a g e | 63 revision surgery up to the first 100 cases including length of hospital stay, post-operative performed [23, 30]. With 8.9% and 24.1% of pain, dislocation rate, accuracy of acetabular DAA surgeons performing less than 10 and 50 positioning, and short-term functional status. anterior cases per year respectively, it merits More than half of DAA surgeons felt that there discussion whether a minimum annual case count was no difference between approaches in terms is needed to maintain one’s surgical skills with of blood loss, rate of femoral fracture, overall the DAA. complication rate, and long-term functional Bender et al [31] discuss potential status. The purpose of this article is to present contraindications for the DAA and mention that readers with comprehensive information consideration should be given to other regarding current trends rather than comparing approaches in the setting of femoral deformity, surgeon perceptions to published data. large anterior panniculus, and revision surgery Knowledge of the current practices and setting. Similarly, in this investigation, when preferences in conjunction with the existing asked about factors that would lead current DAA literature will perhaps provide THA surgeons surgeons to perform a PA, the top 3 responses with a better understanding of surgical were revision case, complex anatomy, and approaches and driving factors for their adoption. significant obesity/body habitus, respectively. This investigation provides insight not Our data suggest that the vast majority of DAA only into trends for DAA performers, but also THA surgeons have at least one or more perceptions by DAA non-performers. Critics of indications to use an alternate approach. In the DAA claim increased complication rate, lack contrast, only 7.9% of DAA surgeons felt that of clinical superiority compared to certain other there was no indication to use another approach. approaches, and steep learning curve which have Significant literature exists regarding the all been reported to some extent in the literature clinical outcomes and complication profile of the [30, 32-36]. This is mirrored by the results of this DAA compared to the PA; however, this study as the top 3 reasons for currently not continues to be the topic of much debate [4-8]. performing the DAA were belief of worse To our knowledge, this investigation is the first outcomes with the DAA (30.7%), no benefit over of its kind to provide insight into actual surgeon other surgical approaches (25.7%), and concern perceptions for these variables. A significant about the learning curve (23.6%). These majority (>70%) of DAA surgeons felt that perceptions continue to be the topic of much outcomes differed between the DAA and PA debate and further high-powered, prospective
P a g e | 64 investigations are needed to elicit a more statistically shorter practice times than those definitive consensus. without intention to adopt provokes Marketing, at various levels in the consideration of specific driving factors. A healthcare infrastructure, is considered perhaps combined 84.6% of surgeons planning to adopt one of the largest driving forces leading to recent the DAA in the future indicated concern for the popularization of the DAA for THA [16-18]. learning curve or lack of training with the Mohan et al [17] discuss the substantial online approach as the single most reason for currently marketing for the DAA from hospitals, surgeons, not performing the technique. What this appears and industry while calling into question the to suggest is that surgeons earlier in their practice quality of such easily accessible information. are perhaps more likely to adopt the DAA and Current DAA surgeons are seemingly aware of embrace the known associated learning curve. this phenomenon as 43.0% report that more than There are a few limitations associated half of their THA patients request the DAA. with this investigation. The obtained data may be Furthermore, in today’s era with increased subject to biases inherently associated with emphasis on cost containment and optimization, electronic questionnaire-based studies including economic implications of surgical practice response bias and sampling error. Only AAHKS cannot be ignored. An impressive 76.1% of members were queried in this study. This current DAA surgeons reported having increased potentially introduces bias due to the propensity patient market share due to performing the of AAHKS members to more likely be approach. Similarly, 65.8% of current DAA non- fellowship trained. Additionally, with a 44.3% performing surgeons reported losing patients to AAHKS member response rate, it warrants other surgeons as a direct result of not consideration of potentially differing results if performing the approach. Interestingly, only more society members were included. Multiple 3.0% of non-DAA performers indicate plans to other approaches for THA exist that were not adopt the DAA in the future. This implies that discussed here which may be construed as not despite perceptions of lost patient market share, being comprehensive. The objective of this the vast majority of DAA non-performers adhere investigation, however, was to compare the 2 to their choice in surgical approach even with most commonly utilized and debated THA known economic implications. The finding that approaches. This does leave opportunity for bias current DAA non-performing surgeons planning as other specific utilized THA approaches were to adopt the DAA have been in practice for not inquired about for either the DAA
P a g e | 65 performing or non-performing cohorts. Therefore, it is not possible with the data collected to distinguish between surgeons who utilize the DAA as the workhorse approach or those who perform it selectively. Similarly, not all DAA surgeons also perform the PA despite being asked to compare outcomes of the 2 approaches. This is the first study of its kind to highlight current trends and clinical practices from a surgeon perspective regarding the DAA for THA. Specifically, it sheds light into perceptions of outcomes, economic considerations, and driving factors for choice of surgical approach for those who do and do not perform the DAA.
P a g e | 66
TABLES
P a g e | 67
P a g e | 68
FIGURES
Fig. 1. Proportion of THA patients requesting the DAA as reported by current DAA surgeons.
Fig. 2. Peri-operative variables: DAA vs PA perceptions by all current DAA performers
P a g e | 69
Fig. 3. Complications: DAA vs PA perceptions by all current DAA performers
Fig. 4. Post-operative outcomes: DAA vs PA perceptions by all current DAA performers
P a g e | 70
Fig. 5. Reported single most important reason for not performing DAA by all current DAA non- performers
Fig. 6. Economic perceptions toward DAA by current performers and non-performers
P a g e | 71
Fig. 7. Single most important barrier for current DAA non-performers planning to adopt DAA in future
APPENDIX
P a g e | 72
Survey Page #1 Completed by All Responders.
Survey Page #2 Completed by Those Responding “No” to Performing DAA
Survey Page #2 Completed by Those Responding “Yes” to Performing DAA
P a g e | 73
P a g e | 74
REFERENCES [1] Learmonth ID, Young C, Rorabeck C. The operation of the century: total hip replacement. Lancet 2007;370:1508e19.
[2] Khanduja V. Total hip arthroplasty in 2017dcurrent concepts and recent advances. Indian J Orthop 2017;51:357e8.
[3] Laupacis A, Bourne R, Rorabeck C, Feeny D, Wong C, Tugwell P, et al. The effect of elective total hip replacement on health-related quality of life. J Bone Joint Surg Am 1993;75:1619e26.
[4] Jia F, Guo B, Xu F, Hou Y, Tang X, Huang L. A comparison of clinical, radiographic and surgical outcomes of total hip arthroplasty between direct anterior and posterior approaches: a systematic review and meta-analysis. Hip Int 2018 [Epub ahead of print].
[5] Higgins BT, Barlow DR, Heagerty NE, Lin TJ. Anterior vs. posterior approach for total hip arthroplasty, a systematic review and meta-analysis. J Arthroplasty 2015;30:419e34.
[6] Tripuraneni KR, Munson NR, Archibeck MJ, Carothers JT. Acetabular abduction and dislocations in direct anterior vs posterior total hip arthroplasty: a retrospective, matched cohort study. J Arthroplasty 2016;31:2299e302.
[7] Poehling-Monaghan KL, Kamath AF, Taunton MJ, Pagnano MW. Direct anterior versus miniposterior THA with the same advanced perioperative protocols: surprising early clinical results. Clin Orthop Relat Res 2015;473:623e31.
[8] Petis S, Howard JL, Lanting BL, Vasarhelyi EM. Surgical approach in primary total hip arthroplasty: anatomy, technique and clinical outcomes. Can J Surg 2015;58:128e39.
[9] Yue C, Kang P, Pei F. Comparison of direct anterior and lateral approaches in total hip arthroplasty: a systematic review and meta-analysis (PRISMA). Medicine (Baltimore) 2015;94:e2126.
[10] Wang Z, Hou JZ, Wu CH, Zhou YJ, Gu XM, Wang HH, et al. A systematic review and meta-analysis of direct anterior approach versus posterior approach in total hip arthroplasty. J Orthop Surg Res 2018;13:229.
P a g e | 75
[11] Christensen CP, Jacobs CA. Comparison of patient function during the first six weeks after direct anterior or posterior total hip arthroplasty (THA): a randomized study. J Arthroplasty 2015;30(9 Suppl):94e7.
[12] Zawadsky MW, Paulus MC, Murray PJ, Johansen MA. Early outcome comparison between the direct anterior approach and the mini-incision posterior approach for primary total hip arthroplasty: 150 consecutive cases. J Arthroplasty 2014;29:1256e60.
[13] Zhao HY, Kang PD, Xia YY, Shi XJ, Nie Y, Pei FX. Comparison of early functional recovery after total hip arthroplasty using a direct anterior or posterolateral approach: a randomized controlled trial. J Arthroplasty 2017;32:3421e8.
[14] Restrepo C, Parvizi J, Pour AE, Hozack WJ. Prospective randomized study of two surgical approaches for total hip arthroplasty. J Arthroplasty 2010;25:671e679.e1.
[15] Post ZD, Orozco F, Diaz-Ledezma C, Hozack WJ, Ong A. Direct anterior approach for total hip arthroplasty: indications, technique, and results. J Am Acad Orthop Surg 2014;22:595e603.
[16] Massin P, SFHG. Marketing the direct anterior approach to the hip: is the industry overstepping its role? Orthop Traumatol Surg Res 2016;102:277e8.
[17] Mohan R, Yi PH, Hansen EN. Evaluating online information regarding the direct anterior approach for total hip arthroplasty. J Arthroplasty 2015;30:803e7.
[18] Shofoluwe AI, Naveen NB, Inabathula A, Ziemba-Davis M, Meneghini RM, Callaghan JJ, et al. Internet promotion of direct anterior approach total hip arthroplasty by members of the American Association of Hip and Knee Surgeons. J Arthroplasty 2018;33:167e170.e1.
[19] Lee GC, Marconi D. Complications following direct anterior hip procedures: costs to both patients and surgeons. J Arthroplasty 2015;30(9 Suppl):98e101.
[20] Hartog YM, Vehmeijer SB. High complication rate in the early experience of minimally invasive total hip arthroplasty by the direct anterior approach. Acta Orthop 2013;84:116e7.
[21] Spaans AJ, van den Hout JA, Bolder SB. High complication rate in the early experience of minimally invasive total hip arthroplasty by the direct anterior approach. Acta Orthop 2012;83:342e6.
P a g e | 76
[22] Stone AH, Sibia US, Atkinson R, Turner TR, King PJ. Evaluation of the learning curve when transitioning from posterolateral to direct anterior hip arthroplasty: a consecutive series of 1000 cases. J Arthroplasty 2018;33:2530e4.
[23] Masonis J, Thompson C, Odum S. Safe and accurate: learning the direct anterior total hip arthroplasty. Orthopedics 2008;31(12 Suppl. 2).
[24] Cheng TE, Wallis JA, Taylor NF, Holden CT, Marks P, Smith CL, et al. A prospective randomized clinical trial in total hip arthroplastydcomparing early results between the direct anterior approach and the posterior approach. J Arthroplasty 2017;32:883e90.
[25] Trousdale WH, Taunton MJ, Mabry TM, Abdel MP, Trousdale RT. Patient perceptions of the direct anterior hip arthroplasty. J Arthroplasty 2017;32:1164e70.
[26] Maratt JD, Gagnier JJ, Butler PD, Hallstrom BR, Urquhart AG, Roberts KC. No difference in dislocation seen in anterior vs posterior approach total hip arthroplasty. J Arthroplasty 2016;31(9 Suppl):127e30.
[27] Sheth D, Cafri G, Inacio MC, Paxton EW, Namba RS. Anterior and anterolateral approaches for THA are associated with lower dislocation risk without higher revision risk. Clin Orthop Relat Res 2015;473:3401e8.
[28] Kennon RE, Keggi JM, Wetmore RS, Zatorski LE, Huo MH, Keggi KJ. Total hip arthroplasty through a minimally invasive anterior surgical approach. J Bone Joint Surg Am 2003;85-A(Suppl. 4):39e48.
[29] Ruddell JH, Eltorai AEM, DePasse JM, Kuris EO, Gil JA, Cho DK, et al. Trends in the orthopaedic surgery subspecialty fellowship match: assessment of 2010 to 2017 applicant and program data. J Bone Joint Surg Am 2018;100:e139.
[30] Hartford JM, Bellino MJ. The learning curve for the direct anterior approach for total hip arthroplasty: a single surgeon's first 500 cases. Hip Int 2017;27: 483e8.
[31] Bender B, Nogler M, Hozack WJ. Direct anterior approach for total hip arthroplasty. Orthop Clin North Am 2009;40:321e8.
[32] Graves SC, Dropkin BM, Keeney BJ, Lurie JD, Tomek IM. Does surgical approach affect patient-reported function after primary THA? Clin Orthop Relat Res 2016;474:971e81.
P a g e | 77
[33] Christensen CP, Karthikeyan T, Jacobs CA. Greater prevalence of wound complications requiring reoperation with direct anterior approach total hip arthroplasty. J Arthroplasty 2014;29:1839e41.
[34] Russo MW, Macdonell JR, Paulus MC, Keller JM, Zawadsky MW. Increased complications in obese patients undergoing direct anterior total hip arthroplasty. J Arthroplasty 2015;30:1384e7.
[35] Jewett BA, Collis DK. High complication rate with anterior total hip arthroplasties on a fracture table. Clin Orthop Relat Res 2011;469:503e7.
[36] Sculco TP. Anterior approach in THA improves outcomes: opposes. Orthopedics 2011;34:e459e61.
P a g e | 78
Fusions Ending Above the Sagittal Stable Vertebrae in Adolescent Idiopathic Scoliosis: Does it Matter?
Dale N. Segal, MD; Keith J. Orland, MD; Eric Yoon; Tracey Bastrom; Nicholas D. Fletcher, MD
P a g e | 79
P a g e | 80
P a g e | 81
P a g e | 82
P a g e | 83
P a g e | 84
P a g e | 85
P a g e | 86
Medicaid Payer Status is Associated with Increased 90-Day Morbidity and Resource Utilization Following Primary Shoulder Arthroplasty: a Propensity-Score-Matched Analysis
David Shau, MD; Kevin Farley, BS; Will Godfrey, MD; Eric Wagner, MD
P a g e | 87
Introduction: Medicaid payer status has been score matching was utilized to control for shown to affect risk-adjusted patient baseline confounding variables. One-to-one outcomes and healthcare utilization across matching between the payer groups and type multiple medical specialties and orthopaedic of arthroplasty (reverse, anatomic, and hemi) procedures. However, the impact of Medicaid was performed, and the odds ratio (OR) and payer status on 90-day morbidity and resource 95% confidence interval (CI) for 90-day utilization following primary shoulder readmission and reoperation rates were arthroplasty (reverse, anatomic, and hemi) has calculated along with a comparison of LOS not yet been assessed. The purpose of this and cost between the propensity score study was to examine 90-day readmission and matched cohorts. reoperation rates, hospital length of stay Results: A total of 4,667 Medicaid and
(LOS), and direct cost following primary 161,147 non-Medicaid patients were shoulder arthroplasty. identified from the 2011-2016 NRDs.
Methods: The National Readmission Propensity score analysis was performed, and
Database (NRD) was queried for all patients 4,637 Medicaid patients were matched to undergoing primary anatomic, reverse, and 4,637 non-Medicaid patients, each with 1504 hemi-shoulder arthroplasty (aTSA, rTSA, (32.4%) rTSAs, 1934 (41.7%) aTSAs, and
HA) from 2011 to 2016 and corresponding 1199 (25.9%) HAs. Patients with Medicaid
“Medicaid” or “non-Medicaid” payer status payer status yielded significant increase in 90- were determined. Patient demographics and day all-cause readmission rates of 11.6% vs comorbidities along with 90-day readmission, 9.3% (p < 0.001; OR = 1.28 [95% CI, 1.12 to
90-day reoperation, LOS, and inflation- 1.46]), 90-day shoulder-related readmission adjusted cost was assessed and propensity rates of 3.3% versus 2.3% (p = 0.004; OR =
P a g e | 88
1.44 [95% CI, 1.12 to 1.85]), and 90-day Introduction reoperation rates of 2.0% vs. 1.3% (p = 0.008, Medicaid is the single largest health
OR=1.54 [95% CI, 1.12 to 1.94]). insurance provider in the United states, with
Furthermore, there was an increased risk of coverage for over 72 million patients(1). extended LOS > 2 days (28.4% versus 25.7%; With the expansion to the Medicaid program p=0.004; OR = 1.14 [95% CI, 1.04 to 1.25]) under the Affordable Care Act (ACA), the along with increased direct cost (median of number of individuals insured under the
$17,612 versus $16,775; p < 0.001). program is expected to grow by
Discussion: This study demonstrates that approximately 12 million(2). The impact that
Medicaid payer status is independently Medicaid payer status has on outcomes and associated with increased 90-day readmission resource utilization has been shown across and reoperation rates, LOS, and direct cost multiple medical specialties, including following primary shoulder arthroplasty. orthopaedics.
Providers may have a disincentive to treat Recently, resource utilization and patient populations who require increased Medicaid payer status has been studied resource utilization following surgery, and extensively with regards to total joint risk adjustment models accounting for arthroplasty. Multiple studies have
Medicaid payer status are necessary to avoid demonstrated a significant increase in decreased access to care for this patient resource utilization by Medicaid patients population and to avoid financial penalty for undergoing total hip arthroplasty and total physicians and hospitals alike. knee arthroplasty(3, 4). However, similar
Level of Evidence: III sized studies have not been performed on
patients undergoing primary shoulder
P a g e | 89 arthroplasty, which includes anatomic and The purpose of this study was to reverse total shoulder arthroplasty, as well as examine 90-day readmission and reoperation hemi-arthroplasty. rates, hospital length of stay (LOS), and direct
Shoulder arthroplasty is a widely cost associated with Medicaid payer status accepted and well tolerated procedure for following primary shoulder arthroplasty. managing degenerative conditions of the While current evidence suggests Medicaid shoulder. Patient outcomes are reliably payer status leads to increased resource improved following these procedures(5-8), utilization following shoulder arthroplasty(10, and the frequency with which shoulder 11), these studies are limited by small sample arthroplasty is being performed continues to size. This publication contains a large sample increase, with projections that the number of size and utilizes a national database to primary shoulder arthroplasties performed in facilitate a matched comparison of Medicaid the US will increase by 755% by 2030 patients with control patients who differ only compared to 2011 totals (9). As such, it is in payer status. We hypothesize that important to understand factors that contribute Medicaid payer status leads to increased 90- to cost and resource utilization following day readmission and reoperation rates, LOS, primary shoulder arthroplasty. While and direct cost following primary shoulder multiple studies have shown that low arthroplasty. socioeconomic status has a negative impact Methods on patient outcomes, there is a paucity of data The National Readmission Database regarding the impact of Medicaid payer status (NRD) was queried for all patients on resource utilization and cost following undergoing primary reverse, anatomic, and primary shoulder arthroplasty. hemi-shoulder arthroplasty (aTSA, rTSA,
P a g e | 90
HA) from 2011 to 2016. Corresponding logistic regression, with insurance type as the
“Medicaid” or “non-Medicaid” payer status dependent variable, was used to estimate was also determined. Patient demographics propensity scores. The independent variable
(age, sex, and discharge to a skilled nursing in this model were covariates that may be facility), arthroplasty for proximal humerus associated with the insurance type and fracture, comorbidities (Congestive heart outcome measured. The propensity score was failure, peripheral vascular disease, chronic then calculated with the use of the resulting obstructive pulmonary disease [COPD], logistic regression equation as the probability acquired immune deficiency syndrome of a patient being in the Medicaid insurance
[AIDS], coagulopathy, rheumatoid arthritis, group. chronic blood loss anemia, alcohol use One-to-one matching between the disorder, drug use, and depression), 30-day payer groups and type of arthroplasty readmission, 90-day readmission, 90-day (reverse, anatomic, and hemi) was performed reoperation, LOS, and inflation-adjusted cost using a greedy matching strategy(14). With were assessed. Cost was defined as actual this strategy, a Medicaid patient is selected, expenses caused during the production of and then a control (a patient with another hospital services, and is not the hospital insurance) who has the closest propensity charges. Cost is also only reported for the score that lies within a defined distance from primary procedure and hospital stay, and does the selected Medicaid patient’s propensity not include the cost of readmission. To score to be their match. This defined distance control for confounding variables between is called the propensity score caliper, and in payer groups, a propensity score based this analysis a caliper of 0.2 standard matching model was used(12, 13). Binary deviations of the logit of the propensity score
P a g e | 91 was used(15). If multiple non-Medicaid were evaluated using the Wilcoxon signed- patients had propensity scores that were rank test. equally close to that of a Medicaid patient, Results then 1 of the non-Medicaid patients was selected at random. Standardized differences Patient Demographics and Baseline were used to assess the balance of Information confounders between the 2 exposure groups. A total of 165,810 patients undergoing
The final step in the analysis required shoulder arthroplasty were identified: 4,663 estimating the effect of exposure on (2.8%) with Medicaid payer status and readmission after primary shoulder 161,147 (97.2%) with insurance of other arthroplasty. Multivariable regression is not types. This included 68,044 (42.2%) necessary since matching on the propensity anatomic TSA, 72,353 (44.9%) reverse TSA, scores addressed confounding. The relative and 20,750 (12.9%) hemiarthroplasty. Patient risk (RR) and its 95% confidence interval (CI) demographics and comorbidities are for readmission were calculated on the basis displayed in Table 1. Prior to propensity of the 2-by-2 table for matched pairs, iwhere score matching, Medicaid patients were the outcome is readmission and the predictor younger (55.29 ± 10.26 yrs vs. 69.81 ± 9.92 is type of insurance. A RR estimate of >1.0 yrs, p<0.001), more likely to undergo suggests that Medicaid patients were at higher hemiarthroplasty (25.9% vs. 12.9%, p<0.001), risk of readmission after primary shoulder undergo surgery for a proximal humerus arthroplasty than patients with other types of fracture (13.0% vs. 11.9%, p<0.001), be male insurance. LOS and total cost comparisons (46.0% vs. 41.6%, p<0.001), and generally
have a higher pre-operative comorbidity
P a g e | 92 burden, including chronic lung disease, p=0.004), however this difference was small. acquired immunodeficiency syndrome All-Cause Readmission Rates (AIDS), coagulopathies, alcohol use disorders, drug use, and depression. In unmatched cohorts, 30-day readmission rate for Medicaid patients was After propensity score matching, 5.3% vs 4.0% for non-Medicaid patients differences in cohorts were less pronounced (p<0.001), and 90-day readmission rate was (Table 1). The mean difference in age 11.6% for Medicaid vs. 8.2% for non- decreased to <1 year (Medicaid: 55.44 yrs, Medicaid patients (p<0.001) (Table 2). In the control: 56.13 yrs, p<0.001). As exact propensity score matched cohorts, controlling matching was performed on both surgery type for patient demographic and comorbid data in and proximal humerus fracture in the Table 1, Medicaid patients had significantly propensity score matching process, both of higher odds of 90-day readmission (Odds these variables were exactly equally Ratio [OR]:1.28, 95% Confidence Interval distributed in both the Medicaid and control [CI]: 1.12- 1.46, p<0.001), but no increased cohort (both p=0.999). Both cohorts included risk of 30-day readmission (OR: 1.13, 95% 599 patients (12.9%) undergoing surgery for a CI: 0.93-1.36) (Table 2). proximal humerus fracture, and 1,934
(41.7%) receiving an aTSA, 1,504 (32.4%) Surgical Complications rTSA, and 1,199 (25.9%) HA. All other Surgical complications, including 90- differences in cohorts were reduced and not day readmission for a shoulder specific significant with p>0.05. The only significent complication, 90-day infection (surgical site differences still present were higher rates of infection or prosthetic joint infection), and 90- COPD in the control cohort (30.6% vs 27.8%, day revision surgery, were all higher in
P a g e | 93
Medicaid payer patients in unmatched cohorts was significantly increased, with a mean
(Table 2). Specifically, shoulder specific difference of $1,112 (95% CI: 504-1,721) and complications leading to readmissions were a median difference of $837 (p<0.001). seen in 3.3% of the Medicaid patients, but Additionally, in matched cohorts, there was a
1.7% of the non-Medicaid controls, infection 0.40 (95% CI: 0.17-0.63, p<0.001) day was present in 1.0% of Medicaid patients and increase in LOS in Medicaid patients.
0.5% of non-Medicaid patients, and revision Additionally, Medicaid patients had a 1.14 surgery in 2.1% of Medicaid patients and times increased odds of an extended LOS > 2
1.1% of non-Medicaid patients (all p<0.001). days (p=0.004, 95% confidence interval 1.04-
In the propensity score adjusted analysis, 1.25)
Medicaid payer patients had a 1.44 times Discussion increased odds of a shoulder specific In today’s healthcare climate, complication readmission (95% CI: 1.12-1.85, understanding identifiable risk factors that p=0.004) and a 1.54 times increased odds of contribute to increased cost of care are
90-day revision surgery (95% CI: 1.12-2.13, increasingly important. This study p=0.008). However, in the matched cohorts, demonstrates that Medicaid payer status is there was loss of significance differences in independently associated with increased odd the rates of infection (OR: 1.26, 95% CI: of 90-day readmission (OR:1.28, 95% CI:
0.83-1.94, p=0.279). 1.12- 1.46, p<0.001) and reoperation rates
(OR: 1.54, 95% CI: 1.12-2.13, p=0.008), LOS Hospital Cost and Length of Stay (OR: 1.14, 95% CI:1.04-1.25, p=0.004), and
Table 3 displays a comparison of cost increased hospital cost with a mean difference and LOS of Medicaid and non-Medicaid of $1,112 (95% CI: 504-1,721) following patients. In matched cohorts, hospital cost
P a g e | 94 primary shoulder arthroplasty. These findings reoperation have the potential to drastically are consistent with previous studies in increase the cost of a patient’s episode of orthopaedic literature, which demonstrated care. This study shows that Medicaid payer increased LOS and total cost following status is associated with both increased direct primary total knee and hip arthroplasty(3, 4, cost as well as increased 90-day readmission
16). Matsen et. al evaluated data for 17,133 rates following primary shoulder arthroplasty, patients and found patients with Medicaid just as it has been shown following total hip payer had a 14% longer LOS follow primary arthroplasty and total knee arthroplasty(3, 4, shoulder arthroplasty (10), which is similar to 16). This could potentially lead to disparities our findings. Our study assesses a large in patients’ access to care, as providers could cohort through a national database and be hesitant to accept patients with Medicaid accounts for a wide variety of confounding due to the increased financial risk associated patient factors found in the Medicaid patient with their care. As such, adjusted re- population, such as younger age and increased imbursement models would be beneficial in a medical comorbidities. The use of propensity setting where TSA is being covered by score matching allowed us to control for these alternative payment models such as bundled confounders. payments.
With the increasing popularity of Additionally, shoulder arthroplasty alternate payment models, physicians and surgeons should consider the challenging health care systems have increasing incentive socioeconomic pressures that their Medicaid to provide quality care at the lowest possible patients may face. Patients with Medicaid cost. Certain aspects of a patients care, such often have complex social and economic as post discharge care, re-admission, or challenges that could influence anything from
P a g e | 95 their ability to receive inpatient rehab, to variables through propensity-score matching. travelling further for postoperative follow up There are inherent limitations present in any
(17). As such, certain interventions taken by large database study, such as inaccurate the surgeon and their practice during the coding and incomplete data collection as well perioperative period, such as closer as lack of patient outcomes. And while postoperative follow up, or the use of propensity score matching is a useful tool to healthcare navigators may help mitigate these control for measured covariates, it does not outcomes. Promising results from a ensure unmeasured covariates are controlled randomized controlled trial regarding the use for. In this case, the NRD does not track of care navigators for Medicaid patient in income, employment status, or education emergency departments (ED) showed 1.4 level and as such these could not be included fewer ED visits (p=0.01) and 1.0 fewer in the analysis. hospitalizations (p=0.001) in the group that While we show that patients with received assistance from a patient navigator Medicaid payer status are at risk for increased
(18). The use of patient navigators in the LOS, 90-day readmission, and total cost perioperative period could potentially following primary shoulder arthroplasty, these overcome some barriers to care, and warrants procedures remain remarkably beneficial. further study with regards to orthopaedic care. This study does not delve into patient reported
One strength of this study lies in its outcomes, however patients with Medicaid size, with 4,637 Medicaid patients and their payer status have been shown to have matched controls, which was accomplished excellent functional improvements following with the use of the NRD. Another strength is primary shoulder arthroplasty (11, 19). the ability to control for confounding Patients with Medicaid often have complex
P a g e | 96 social and economic barriers to care, and while determining how to overcome these social challenges is complicated, it is imperative to better optimize care for these at- risk patients.
P a g e | 97
Conclusion
With the continued use of alternative payment models such as bundled payments, access to care for patients with increased risk of complications and readmission could be threatened by pressures to limit cost in health care systems. We have shown that Medicaid payer status is independently associated with increased 90-day readmission and reoperation rates, LOS, and direct cost following primary shoulder arthroplasty. Risk adjustment models accounting for
Medicaid payer status will be necessary to ensure good access to care for this patient population by avoiding penalties for physicians and hospital systems. It is important to develop a better understanding of why Medicaid payer status is associated with an increased rate of readmission,
LOS, and cost. In doing so, it will allow shoulder arthroplasty surgeons to provide improved care to Medicaid patients while optimizing resource utilization following primary shoulder arthroplasty.
P a g e | 98
Table 1. Shoulder Arthroplasty Patient Demographics and Comorbidities in Unmatched and Matched Cohorts
Unmatched Cohorts Matched Cohorts b Patient Characteristic Control Medicaid P-Value Control Medicaid P-Value
Total 161,147 (97.2%) 4,663 (2.8%) 4,637 (50.0%) 4,637 (50.0%) Age, yrs (Mean ± SD) 69.81 ± 9.92 55.29 ± 10.26 <0.001 56.13 ± 11.02 55.44 ± 10.03 <0.001 Surgery Type a Total Shoulder Arthroplasty 68,044 (42.2%) 1,944 (41.7%) <0.001 1,934 (41.7%) 1,934 (41.7%) 0.999 Reverse Shoulder Arthroplasty 72,353 (44.9%) 1,511 (32.4%) 1,504 (32.4%) 1,504 (32.4%) Hemiarthroplasty 20,750 (12.9%) 1,208 (25.9%) 1,199 (25.9%) 1,199 (25.9%) Sex Male 67,057 (41.6%) 2,145 (46.0%) <0.001 2,094 (45.2%) 2,134 (46.0%) 0.404 Female 94,090 (58.4%) 2,518 (54.0%) 2,543 (54.8%) 2,503 (54.0%) Non-Home Discharge 23,239 (14.4%) 367 (7.9%) <0.001 398 (8.6%) 366 (7.9%) 0.227 Proximal Humerus Fracture 19,112 (11.9%) 607 (13.0%) <0.001 599 (12.9%) 599 (12.9%) 0.999
Comorbidities Congestive Heart Failure 6,043 (3.8%) 155 (3.3%) 0.131 186 (4.0%) 155 (3.3%) 0.087 Perivascular Disease 5,040 (3.1%) 71 (1.5%) <0.001 76 (1.6%) 71 (1.5%) 0.678 COPD 29,233 (18.1%) 1,296 (27.8%) <0.001 1,417 (30.6%) 1,290 (27.8%) 0.004 AIDS 73 (0.1%) 18 (0.4%) <0.001 14 (0.3%) 18 (0.4%) 0.479 Coagulopathy 2,695 (1.7%) 98 (2.1%) 0.025 105 (2.3%) 97 (2.1%) 0.569 Rheumatoid Disease 8,803 (5.5%) 258 (5.5%) 0.835 249 (5.4%) 255 (5.5%) 0.783 Chronic blood loss anemia 1,077 (0.7%) 25 (0.5%) 0.273 38 (0.8%) 25 (0.5%) 0.100 Deficiency Anemia 14,053 (8.7%) 385 (8.3%) 0.268 411 (8.9%) 383 (8.3%) 0.299 Alcohol Use Disorders 2,520 (1.6%) 262 (5.6%) <0.001 270 (5.8%) 257 (5.5%) 0.560 Drug Use 1,332 (0.8%) 214 (4.6%) <0.001 177 (3.8%) 209 (4.5%) 0.096 Depression 23,787 (14.8%) 931 (20.0%) <0.001 991 (21.4%) 924 (19.9%) 0.086 a Exact Matching Performed on Surgery Type (RSA, TSA, or HA) and Humerus Fracture Status b 1-to-1 propensity score matched patients with greedy matching using calipers of a specified width (0.20 standard deviation of the logit of the propensity score)
Table 2. Comparison of Readmission, Complications, and Revision Surgery After Shoulder Arthroplasty (N = 4,637)
Unmatched Cohorts Matched Cohorts b Complication Control Medicaid P-Value a Control Medicaid Odds (95% CI) P-Value a All Cause 30-Day Readmission 6,525 (4.0%) 248 (5.3%) <0.001 215 (4.6%) 241 (5.2%) 1.13 (0.93-1.36) 0.212 All Cause 90-Day Readmission 13,266 (8.2%) 543 (11.6%) <0.001 431 (9.3%) 536 (11.6%) 1.28 (1.12-1.46) <0.001 Readmission for Shoulder 2,811 (1.7%) 152 (3.3%) <0.001 106 (2.3%) 151 (3.3%) 1.44 (1.12-1.85) 0.004 Specific Complications Infection (SSI or PJI) 830 (0.5%) 48 (1.0%) <0.001 38 (0.8%) 48 (1.0%) 1.26 (0.83-1.94) 0.279 90-Day Revision 1,794 (1.1%) 96 (2.1%) <0.001 62 (1.3%) 95 (2.0%) 1.54 (1.12-2.13) 0.008 a Chi-square statistic p-value
Table 3. Comparison of Cost and LOS After Shoulder Arthroplasty in Matched and Unmatched Cohorts (N = 4,637)
Unmatched Cohorts Matched Cohorts b Characteristic Control Medicaid P-Value Control Medicaid P-Value Hospital Cost (USD, $) Mean ± SD a 18,799 ± 9,530 20,374 ± 15,385 <0.001 19,188 ± 14,385 20,300 ± 15,233 <0.001 Median [Interquartile Range] b 17,003 (13,363-21,844) 17,616 (13,609-23,282) <0.001 16,775 (13,128-21,788) 17,612 (13,614-23,249) <0.001 Length of Stay Mean ± SD a 2.38 ± 2.96 2.89 ± 7.20 <0.001 2.46 ± 3.53 2.86 ± 7.14 <0.001 LOS >2 Days (N, %) 44,152 (27.4%) 1,333 (28.6%) 0.074 1,192 (25.7%) 1,315 (28.4%) c 0.004 a Independent Samples T-Test, b Mann-Whitney-U Test c Odds Ratio 1.14, 95% confidence interval 1.04-1.25
P a g e | 99
Table 4. Comparison of Readmission, Complications, and Revision Surgery After Shoulder Arthroplasty
Matched Cohorts Complication Control Medicaid P-Value a All Cause 30-Day Readmission 48 (2.5%) 72 (3.7%) 0.026 All Cause 90-Day Readmission 117 (6.0%) 162 (8.4%) 0.005 Readmission for Shoulder 22 (1.1%) 30 (1.6%) 0.264 Anatomic Shoulder Arthroplasty Specific Complications Infection (SSI or PJI) 7 (0.4%) 12 (0.6%) 0.250
90-Day Revision 12 (0.6%) 18 (0.9%) 0.271
All Cause 30-Day Readmission 91 (6.1%) 79 (5.3%) 0.343
All Cause 90-Day Readmission 167 (11.1%) 197 (13.1%) 0.094 Readmission for Shoulder Reverse Shoulder Arthroplasty 58 (3.9%) 79 (5.3%) 0.066 Specific Complications Infection (SSI or PJI) 19 (1.3%) 20 (1.3%) 0.872 90-Day Revision 38 (2.5%) 58 (3.9%) 0.038 All Cause 30-Day Readmission 76 (6.3%) 90 (7.5%) 0.260 All Cause 90-Day Readmission 147 (12.3%) 177 (14.8%) 0.073 Readmission for Shoulder Hemiarthroplasty 26 (2.2%) 42 (3.5%) 0.049 Specific Complications Infection (SSI or PJI) 12 (1.0%) 16 (1.3%) 0.447 90-Day Revision 12 (1.0%) 19 (1.6%) 0.206 a Chi-square statistic p-value, significent p-values (<0.05) displayed in bold
P a g e | 100
1. Medicaid Eligibility Criteria 2019 [Available from: https://www.medicaid.gov/medicaid/eligibility/index.html. 2. Estimates for the Insurance Coverage Provisions of the Affordable Care Act Updated for the Recent Supreme Court Decision 2012 [Available from: http://cbo.gov/sites/default/files/cbofiles/attachments/43472-07-24-2012-CoverageEstimates.pdf. 3. Shau D, Shenvi N, Easley K, Smith M, Bradbury T, Guild G, 3rd. Medicaid Payer Status Is Associated with Increased 90-Day Morbidity and Resource Utilization Following Primary Total Hip Arthroplasty: A Propensity-Score-Matched Analysis. J Bone Joint Surg Am. 2018;100(23):2041-9. 4. Shau D, Shenvi N, Easley K, Smith M, Guild G, 3rd. Medicaid is associated with increased readmission and resource utilization after primary total knee arthroplasty: a propensity score-matched analysis. Arthroplast Today. 2018;4(3):354-8. 5. Flurin PH, Marczuk Y, Janout M, Wright TW, Zuckerman J, Roche CP. Comparison of outcomes using anatomic and reverse total shoulder arthroplasty. Bull Hosp Jt Dis (2013). 2013;71 Suppl 2:101-7. 6. Kiet TK, Feeley BT, Naimark M, Gajiu T, Hall SL, Chung TT, et al. Outcomes after shoulder replacement: comparison between reverse and anatomic total shoulder arthroplasty. J Shoulder Elbow Surg. 2015;24(2):179-85. 7. Simovitch RW, Friedman RJ, Cheung EV, Flurin PH, Wright T, Zuckerman JD, et al. Rate of Improvement in Clinical Outcomes with Anatomic and Reverse Total Shoulder Arthroplasty. J Bone Joint Surg Am. 2017;99(21):1801-11. 8. Wright MA, Keener JD, Chamberlain AM. Comparison of Clinical Outcomes After Anatomic Total Shoulder Arthroplasty and Reverse Shoulder Arthroplasty in Patients 70 Years and Older With Glenohumeral Osteoarthritis and an Intact Rotator Cuff. J Am Acad Orthop Surg. 2019. 9. Padegimas EM, Maltenfort M, Lazarus MD, Ramsey ML, Williams GR, Namdari S. Future patient demand for shoulder arthroplasty by younger patients: national projections. Clin Orthop Relat Res. 2015;473(6):1860-7. 10. Matsen FA, 3rd, Li N, Gao H, Yuan S, Russ SM, Sampson PD. Factors Affecting Length of Stay, Readmission, and Revision After Shoulder Arthroplasty: A Population-Based Study. J Bone Joint Surg Am. 2015;97(15):1255-63.
P a g e | 101
11. Sabesan VJ, Lima DJL, Goss L, Gauger M, Whaley JD, Ghisa C, et al. Does Medicaid payer status affect patient's shoulder outcomes after shoulder arthroplasty? Musculoskelet Surg. 2019. 12. Rosenbaum PR, Rubin DB. The central role of the propensity score in observational studies for causal effects. Biometrika. 1983;70(1):41-55. 13. Rubin DB. The design versus the analysis of observational studies for causal effects: parallels with the design of randomized trials. Statistics in medicine. 2007;26(1):20-36. 14. Bergstralh E, Konsanke J. Computerized Matching of Cases to Controls: Technical Report No. 56. Department of Health Science Research, Mayo Clinic, Rochester, MN. 1995. 15. Austin PC. An introduction to propensity score methods for reducing the effects of confounding in observational studies. Multivariate behavioral research. 2011;46(3):399-424. 16. Browne JA, Novicoff WM, D'Apuzzo MR. Medicaid payer status is associated with in- hospital morbidity and resource utilization following primary total joint arthroplasty. J Bone Joint Surg Am. 2014;96(21):e180. 17. Martin CT, Callaghan JJ, Liu SS, Gao Y, Johnston RC. Disparity in preoperative patient factors between insurance types in total joint arthroplasty. Orthopedics. 2012;35(12):e1798-803. 18. Kelley L, Capp R, Carmona JF, D'Onofrio G, Mei H, Cobbs-Lomax D, et al. Patient Navigation to Reduce Emergency Department (ED) Utilization Among Medicaid Insured, Frequent ED Users: A Randomized Controlled Trial. J Emerg Med. 2020. 19. Waldrop LD, 2nd, King JJ, 3rd, Mayfield J, Farmer KW, Struk AM, Wright TW, et al. The effect of lower socioeconomic status insurance on outcomes after primary shoulder arthroplasty. J Shoulder Elbow Surg. 2018;27(6S):S35-S42.
P a g e | 102
Red Flags for Low Back Pain Are Not Always Really Red: A Prospective Evaluation of the Clinical Utility of Commonly Used Screening Questions for Low Back Pain
William Godfrey, MD, Ajay Premkumar, MD, MPH, Michael B. Gottschalk, MD, and Scott D. Boden, MD
P a g e | 103
ABSTRACT having a vertebral fracture, malignancy,
Introduction infection, or cauda equina syndrome were
Low back pain (LBP) has a high prevalence classified as having a red flag diagnosis. and morbidity, and is a source of significant Results health care spending. Numerous published Specific individual red flags and guidelines support the use of ‘red flag’ combinations of red flags were associated questions to screen for serious pathology for with an increased probability of underlying
LBP. However, there is limited data on the serious spinal pathology, e.g., recent trauma effectiveness of red flags. This paper and an age of >50 years were associated examines the effectiveness of red flag with vertebral fracture. The presence or questions as a screening tool for patients absence of other red flags, such as night presenting with LBP to a multidisciplinary pain, was unrelated to any particular academic spine center. diagnosis. For instance, for patients with no
Methods recent history of infection and no fever,
We conducted a retrospective review of the chills, or sweating, the presence of night cases of 9,940 patients with a chief pain was a false-positive finding for complaint of low back pain. The patients infection >96% of the time. In general, the completed a questionnaire that included absence of red flag responses did not several red flag questions during their first meaningfully decrease the likelihood of a physician visit. Diagnostic data for the same red flag diagnosis; 64% of patients with clinical episode were collected from medical spinal malignancy had no associated red records and were corroborated with imaging flags. reports. Patients who were diagnosed as
P a g e | 104
Conclusions driver of physician visits and hospital
While a positive response to a red flag admissions3-7. question may indicate the presence of While a substantial burden on our serious disease, a negative response to one health-care system, low back pain is or two red flag questions does not typically self-limited, with most cases meaningfully decrease the likelihood of a resolving after conservative management in red flag diagnosis. Clinicians should use 6 to 8 weeks8. While most low back pain is caution when utilizing red flag questions as nonspecific, patients with low back pain screening tools may have neurologic impairments (e.g.,
Level of Evidence radiculopathy or spinal stenosis) or a serious
Level III, Retrospective Cohort Study. underlying pathology (e.g., malignancy or
Keywords infection) that requires timely and accurate
Red flags, Low back pain, Fracture, diagnosis9-11. Low back pain that is due to
Infection, Malignancy, Cauda equina serious pathology occurs between 1% and syndrome 4% of the time, and has been generally
INTRODUCTION classified as related to four etiologies:
Low back pain is a major cause of fracture, malignancy, infection, and cauda disability worldwide1, 2. In the United States, equina syndrome10-12. While they are rare, low back pain has a lifetime prevalence appropriate identification of these diagnoses approaching 85%, is the most common and specialist consultation for their cause of activity limitation in people management are essential. Imaging for low younger than 45 years, and is a leading back pain can be costly, and there is a high
prevalence of abnormal findings in
P a g e | 105 asymptomatic patients12-14. Due to the numbers14, 15. Recent Cochrane reviews have prevalence, burden, and associated health- indicated a need for more robust data and care cost of low back pain, it is paramount to have noted that it is currently challenging to utilize noninvasive and inexpensive provide strong recommendations for or screening tools to detect serious disease. against the clinical application of these
Within this context, red flag questions14, 20. screening questions were developed and In the present study, we analyzed encouraged for use in the detection of records for individuals seen with a chief serious spine pathology in the primary care complaint of low back pain at a large or multidisciplinary setting15. These red flag academic multidisciplinary spine center. We symptoms are widely believed to indicate sought to evaluate the utility of red flag specific underlying pathology, and multiple questions in uncovering their classically practice guidelines for the management of associated diagnoses. We hypothesized that low back pain recommend against the use of red flag questions would improve diagnostic imaging or procedures for low detection and aid in the workup and back pain unless such red flag symptoms are management of infection, malignancy, present9, 16-18. Specific red flag questions are fracture, and cauda equina syndrome. not consistent between guidelines, and there MATERIALS AND METHODS is a paucity of evidence to support their Participants inclusion aside from consensus decision14, 15, Institutional Review Board approval
19-21. was granted for this study. Patients
Prior studies examining the utility of completed a questionnaire that included red red flag questions are hampered by small flag questions during their first visit, before
P a g e | 106 seeing their surgeon. Diagnosis information advocated for use in this context in was drawn directly from physician entry and published clinical practice guidelines9, 15-18. corroborated by imaging reports by an Table 1 shows the specific red flags in our independent radiologist at our institution, if questionnaire and their associated diagnoses. the patient received imaging. Diagnosis Each questionnaire answer was assessed information was also confirmed by categorically as one of two responses: “Yes” reviewing records for International or “No.”
Classification of Diseases (ICD) categories Outcomes for each patient. The primary dependent variable of
All unique patients who presented as interest was the presence of spinal fracture, a new patient with a chief complaint of LBP malignancy, infection, or cauda equina with or without leg pain between July 2005 syndrome. We identified these four and February 2016, to one of six fellowship diagnoses as “red flag diagnoses,” as they trained spine surgeons at our institution, have been previously categorized in this were identified. Patients presenting with a manner in the literature.12, 15, 21 chief complaint of “deformity” rather than If a patient had more than one of
“pain” were excluded from analysis, as were these diagnoses, we treated each diagnosis patients with a chief complaint of neck pain. as a separate entity for analysis. For
All 9,940 patients identified by the above example, a patient with both a diagnosed criteria were included in our analysis. malignancy and fracture was evaluated with
Variables of Interest other patients diagnosed with a malignancy,
The hypothesized red flags for spinal as well as with other patients who had a pathology in our questionnaire have been fracture.
P a g e | 107
Statistical Analysis and twenty-eight patients (8.3%) were
Statistical analysis was performed identified as having a red flag diagnosis. using the SAS software package version 9.4 Fracture was by far the most common red
(SAS institute Inc. Cary, NC). Student t tests flag diagnosis, followed by tumor and were performed for continuous data and chi- infection. Cauda equina syndrome was the square or Fisher exact tests were performed least common red flag diagnosis (Table II). for categorical data, as appropriate. Table III outlines which red flag questions
Sensitivity, specificity, positive likelihood have been classically associated with which ratios (+LR), and negative likelihood ratio (- diagnosis in the setting of low back pain, as
LR) were reported with their corresponding stated in prior practice guidelines8, 14, 15, 23, 24.
95% confidence intervals (CIs) using Analysis of Red Flag Questions &
Wilson scoring methods.22 Two-tailed p Individual Red Flag diagnoses values <0.05 were considered statistically Univariable analysis examining each significant. red flag question and its associated diagnosis
RESULTS identified several significant findings. The
As shown in Table 1, the average age sensitivities, specificities, and positive and of patients in our cohort was 56.8 (standard negative likelihood ratios for each red flag deviation [SD] 15.0) years, and 48.7% were are detailed in Table IV. women. Most patients (92.6%) possessed at Fracture least one listed red flag symptom at the time Presence of age greater than 50 of their new-patient visit. The most (+LR, 1.10 [1.05, 1.16], -LR 0.79 [0.69- commonly endorsed red flag symptom was 0.91]), age greater than 70 (+LR, 1.55 that of night pain (58.1%). Eight hundred [1.36, 1.76], -LR 0.86 [0.82-0.91]), history
P a g e | 108 of recent trauma (+LR, 2.18 [1.86, 2.54], - patients with spinal malignancy had no
LR 0.85 [0.81, 0.89]) were all associated associated red flags. with a significantly increased probability of Infection vertebral fracture when present and a Fever, chills or sweating (+LR, 1.71 significantly decreased probability of [1.04, 2.81]) and recent infection (+LR, 9.31 vertebral fracture when absent. [6.63, 13.07], -LR 0.78 [0.70-0.86]) both
Malignancy significantly increased the probability of a
Unexplained weight loss (+LR, 1.87 spinal infection when present. Absence of
[1.10, 3.17]) and personal history of cancer fever, chills, or sweating was not associated
(+LR, 7.25 [5.65, 9.30], -LR 0.71 with a lower probability of spinal infection
[0.64,0.79]) both significantly increased the (-LR 0.95 [0.89,1.01]). probability of malignancy when present; In addition, night pain ((+LR 0.99 however, absence of unexplained weight [0.85-1.15], -LR 1.02 [0.89-1.01]) did not loss was not significantly associated with significantly change the probability of spinal decreased probability of spinal malignancy infection when present or absent.
(-LR 0.96 [0.92,1.01]). Cauda Equina Syndrome
In addition age greater than 50 (+LR Recent loss of bladder (urination)
1.06 [0.96,1.17], -LR 0.87 [0.68,1.11]), age control (+LR, 2.31 [1.25, 4.27]) and recent greater than 70 (+LR 1.1 [0.82-1.47], -LR loss of bowel control other than diarrhea
0.97 [0.93-1.01]), and night pain ((+LR 0.85 (+LR, 2.78 [1.23, 6.30]) both significantly
[0.83-1.10], -LR 1.07 [0.90-1.27]) did not increased the probability of cauda equina significantly change the probability of spinal syndrome when present; however, absence malignancy when present or absent. 64% of of either of these symptoms did not
P a g e | 109 significantly decrease the probability of likelihood for the diagnosis of spinal cauda equina syndrome. infection (+LR, 13.15 [6.66, 25.97], -LR
Analysis of Combinations of Commonly 0.93 [0.88-0.98]). Lastly, the combination of
Associated Red Flag Questions & recent loss of bladder control and recent
Individual Diagnoses loss of bowel control was significant for an
Analysis of combinations of increased likelihood of a diagnosis of cauda individually significant red flags for each equina syndrome (+LR, 3.00 [1.01, 8.92]). diagnosis of interest was performed and Of note, while each of these combinations revealed several significant associations were relatively specific, all of the above
(Table V). Namely, the combination of age combinations of red flags were poorly greater than 50 and recent trauma was sensitive in identifying their diagnoses of significant for an increased probability of interest (Table V). vertebral fracture (+LR 2.54 [2.05, 3.16]). DISCUSSION
The combination of age greater than 70 and Low back pain is common and is recent trauma had an even stronger increase mostly self-limited, but red flag diagnoses in probability of vertebral fracture (+LR are rare, serious, and require urgent
4.35 [2.92, 6.48]). The combination of treatment. Due to their low prevalence, a unintentional weight loss and a personal large sample size is required to adequately history of cancer was significant for an study them. To our knowledge, this study increased likelihood of spinal tumor (+LR, contains the largest cohort of patients to date
10.25 [3.60, 29.21]). The combination of to assess the utility of red flag questions as fevers, chills or sweating and recent well as combinations of red flags in the infection had a significantly increased diagnosis of serious spinal pathology.
P a g e | 110
The prevalence of a red flag such as advanced imaging.15 Evidence for diagnosis for patients presenting with LBP this recommendation, however, is generally in our study was 8.3%, well above the lacking with most guidelines citing prior published rates in primary care settings of 1- guidelines and unpublished data as
4%, but consistent with rates in tertiary care sources.18, 25 A 2016 review of LBP settings.19 As the prevalence of red flag guidelines advocating for the use of red diagnoses in this population was very low, flags in LBP found that of 16 recently we found traditionally calculated positive or published guidelines, only three presented negative predictive values to be of little any reference to support the use of red flags. relevance (e.g. all negative predictive values Of these three, information on diagnostic were very high because case prevalence was accuracy was limited and when present, so low). Likelihood ratios, a ratio of the referenced studies with fewer than 20 cases post-question probability of diagnosis to the of a red flag diagnosis.15 Despite the paucity pre-question probability of diagnosis, are of information supporting their use, most independent of case prevalence and thus clinical guidelines explicitly recommend the more useful markers of diagnostic utility use of red flag questions as a screening tool. than predictive values in this setting. A key trait of screening tests is that
Clinical guidelines for the negative tests are very unlikely to be falsely management of LBP generally advocate the negative, and that negative test results use of red flag questions to raise a significantly alter the pre-test probability of clinician’s index of suspicion concerning a particular diagnosis. Therefore, serious underlying spinal pathology and as characteristics of good screening tools an indication for further diagnostic work-up, include high sensitivity and high negative
P a g e | 111 likelihood ratios. In our study, not a single underlying serious pathology. Conversely, red flag had a sensitivity over 75%, and the presence nor the absence of night pain most had a sensitivity well under 60%. affected the probability of a patient having a
Similarly, not a single red flag in our study red flag diagnosis. While this question may had a meaningfully low negative likelihood be of use to a clinician to gauge a patient’s ratio. pain and quality of life, its use as a screening
While not ideal screening tests for question is not supported by our data. the aforementioned reasons, our data did Over 90% of patients in this study reveal that particular combinations of red had at least one positive red flag during their flags have a high specificity and can new patient visit. This rate is higher, but significantly increase the probability of fairly consistent with the rate of 80% having certain diagnoses (Table 5). These reported in the primary care population.12 data suggest that when used in combination, Given the high prevalence of red flags, certain red flags can aid a clinician in ruling indiscriminate pursuit of further imaging in in a certain serious diagnosis. For example, the setting of a positive red flag may lead to our data suggests that a patient with a potentially unnecessary imaging, bringing personal history of cancer, presenting with with it increased costs and unnecessary
LBP, and experiencing unintentional weight radiation exposure.26 loss, should be evaluated for a spinal Analysis of the diagnostic accuracy malignancy. of individual and combinations of red flags,
Our data also suggests that a as seen in Tables IV and V, led to the personal history of cancer or a recent development of six recommendations infection should raise alarm for potential
P a g e | 112 regarding the use of traditional red flags in weight loss, alone, carries a 3% probability patients with LBP. of having a spinal infection; however, when
Recommendations both red flags were present, the probability
(1) Recent trauma in the setting of LBP of having a spinal malignancy in the setting should raise a clinician’s suspicion of of LBP was 14.3%, up from a baseline of vertebral fracture, especially in patients 1.6%. greater than 50 years of age. Suspicion (3) Recent history of infection in the setting should be even higher in patients greater of LBP should always raise concern for than 70 years of age. Presence of both recent possibility of infection and prompt questions trauma and age greater than 50 carries a about constitutional symptoms such
13.1% probability of vertebral fracture in the as fevers, chills or sweating, and vice versa. setting of LBP; similarly, presence of both The presence of recent history of infection recent trauma and age greater than 70 in the setting LBP carries a 10% probability carries a 20.5% probability of vertebral of having a spinal infection. History of fracture in the setting of LBP. fevers chills or sweating, alone, carries a 2%
(2) Personal history of cancer in the setting probability of having a spinal infection; of LBP should raise suspicion of malignancy however, when both red flags were present, and prompt questions about constitutional the probability of having an infection in the symptoms such as unintentional weight loss, setting of LBP was 13.8%, up from a and vice versa. The presence of personal baseline of 1.2%. history of cancer in the setting of LBP (4) Presence of loss of bladder control carries a 10.6% probability of having a (urination) and loss of bowel control (other spinal malignancy. History of unintentional than diarrhea) in the setting of LBP should
P a g e | 113 raise suspicion for cauda equina syndrome red flags does not help rule out a particular but is not highly predictive of the diagnosis. red flag diagnosis, as noted by the low
The presence of both red flags increased the change in probabilities, sensitivities and probability of having cauda equina negative likelihood ratios of red flag syndrome to 1.2%, up from a baseline of questions (Tables IV and V).
0.4%. Strengths
(5) Pain that awakens me from sleep, also This study has several strengths, with known as night pain, is not a useful question one of its greatest advantages being its large to determine any particular diagnosis (Table cohort size. With 9940 patients included in
4) and its presence should not be used to the study, it provides us with the ability to justify advanced imaging. More than 55% examine associations between red flag of patients reported this symptom, 60% of questions and rare disease entities, such as whom did not have a red flag diagnosis cauda equina syndrome. Another strength
Classically, night pain has been associated of this study is that it was performed at a with malignancy and infection, however the single large multidisciplinary spine center presence of this symptom was a false with an attached imaging suite, allowing positive over 85% of the time for robust diagnosis data to complement red flag malignancy, and over 96% of the time for question data. The study also used infection when no other associated red flags standardized questions, language, and were present. protocols for providing patients with pre-
(6) Other than age greater than 50 and a visit questionnaires across providers. diagnosis of vertebral fracture, the absence of any individual red flag or combination of
P a g e | 114
Limitations importance of answering the questionnaire
Our study has several limitations. accurately. Another limitation of our study
Not all red flag questions of interest were is that all data was collected at a large included in our questionnaire at time of multidisciplinary spine center. Our center inception in 2005, such as prolonged employs surgeons as well as physiatrists and corticosteroid use, which is traditionally sees primary, secondary, and tertiary care associated with vertebral fracture. As such, spine patients that can self-referred or be caution should be made to avoid drawing referred by another physician. As expected, conclusions on red flags that were not we saw a higher rate of “red flag diagnoses” included in this analysis. Further analyses in our population than published rates in would benefit from including larger primary care populations. While this is a selections and combinations of not only red strength, as a high volume of patients with flag symptoms, but also clinical examination serious spine pathology is a requisite to findings. Another limitation in our study is study the utility of red flag questions, it also that data regarding the presence or absence introduces inherent referral bias. Given our of red flag symptoms was obtained via a finding that red flag questions are not ideal questionnaire, and it is possible that patients screening questions in our multidisciplinary may respond differently to these questions spine center, it is likely they would be of when asked verbally by their physician. In even less benefit in a purely primary care light of this possible issue, prior to population with a lower prevalence of cases. completing the questionnaire, each patient viewed a computer-generated personalized message from their physician regarding the
P a g e | 115
CONCLUSION
The presence of red flag symptoms should be used cautiously as a screening tool for serious spine pathology in the setting of LBP. While a positive response to a red flag question may indicate the presence of disease, a negative response to 1 or 2 red flag questions does not meaningfully decrease the likelihood of a red flag diagnosis. Recommendations for more efficient use of red flags are proposed in this study. Further refinements in clinical practice guidelines to improve screening for serious spinal pathology is indicated.
P a g e | 116
TABLES
Table I
P a g e | 117
Table II
P a g e | 118
Table III
P a g e | 119
Table IV
P a g e | 120
Table V
P a g e | 121
REFERENCES
1. Global Burden of Disease Study C. Global, regional, and national incidence, prevalence, and years lived with disability for 301 acute and chronic diseases and injuries in 188 countries,
1990-2013: a systematic analysis for the Global Burden of Disease Study 2013. Lancet. 2015
Aug 22;386(9995):743-800. Epub 2015/06/13.
2. Disease GBD, Injury I, Prevalence C. Global, regional, and national incidence, prevalence, and years lived with disability for 310 diseases and injuries, 1990-2015: a systematic analysis for the Global Burden of Disease Study 2015. Lancet. 2016 Oct 08;388(10053):1545-
602.
3. Balague F, Mannion AF, Pellise F, Cedraschi C. Non-specific low back pain. Lancet.
2012 Feb 4;379(9814):482-91.
4. Frank A. Low back pain. BMJ. 1993 Apr 3;306(6882):901-9.
5. Hart LG, Deyo RA, Cherkin DC. Physician office visits for low back pain. Frequency, clinical evaluation, and treatment patterns from a U.S. national survey. Spine (Phila Pa 1976).
1995 Jan 1;20(1):11-9. Epub 1995/01/01.
6. Praemer A, Furner S, Rice DP, Kelsey JL. Musculoskeletal conditions in the United
States. 1992.
7. Taylor VM, Deyo RA, Cherkin DC, Kreuter W. Low back pain hospitalization: recent
United States trends and regional variations. Spine. 1994;19(11):1207-12.
8. Patrick N, Emanski E, Knaub MA. Acute and Chronic Low Back Pain. Med Clin North
Am. 2016 Jan;100(1):169-81.
P a g e | 122
9. Chou R, Qaseem A, Snow V, Casey D, Cross JT, Shekelle P, et al. Diagnosis and treatment of low back pain: a joint clinical practice guideline from the American College of
Physicians and the American Pain Society. Annals of internal medicine. 2007;147(7):478-91.
10. Jarvik JG, Deyo RA. Diagnostic evaluation of low back pain with emphasis on imaging.
Ann Intern Med. 2002 Oct 01;137(7):586-97.
11. Raison NT, Alwan W, Abbot A, Farook M, Khaleel A. The reliability of red flags in spinal cord compression. Arch Trauma Res. 2014 Mar;3(1):e17850.
12. Henschke N, Maher CG, Refshauge KM, Herbert RD, Cumming RG, Bleasel J, et al.
Prevalence of and screening for serious spinal pathology in patients presenting to primary care settings with acute low back pain. Arthritis Rheum. 2009 Oct;60(10):3072-80.
13. Boden SD, Davis D, Dina T, Patronas N, Wiesel S. Abnormal magnetic-resonance scans of the lumbar spine in asymptomatic subjects. A prospective investigation. J Bone Joint Surg
Am. 1990;72(3):403-8.
14. Williams CM, Henschke N, Maher CG, van Tulder MW, Koes BW, Macaskill P, et al.
Red flags to screen for vertebral fracture in patients presenting with low-back pain. Cochrane
Database Syst Rev. 2013;1:CD008643.
15. Verhagen AP, Downie A, Popal N, Maher C, Koes BW. Red flags presented in current low back pain guidelines: a review. Eur Spine J. 2016 Sep;25(9):2788-802.
16. Airaksinen O, Brox JI, Cedraschi C, Hildebrandt J, Klaber-Moffett J, Kovacs F, et al.
Chapter 4. European guidelines for the management of chronic nonspecific low back pain. Eur
Spine J. 2006 Mar;15 Suppl 2:S192-300.
P a g e | 123
17. de Jager JP, Ahern MJ. Improved evidence-based management of acute musculoskeletal pain: guidelines from the National Health and Medical Research Council are now available. Med
J Aust. 2004 Nov 15;181(10):527-8.
18. van Tulder M, Becker A, Bekkering T, Breen A, del Real MT, Hutchinson A, et al.
Chapter 3. European guidelines for the management of acute nonspecific low back pain in primary care. Eur Spine J. 2006 Mar;15 Suppl 2:S169-91.
19. Downie A, Williams CM, Henschke N, Hancock MJ, Ostelo RW, de Vet HC, et al. Red flags to screen for malignancy and fracture in patients with low back pain: systematic review.
BMJ. 2013;347:f7095.
20. Henschke N, Maher CG, Ostelo RW, de Vet HC, Macaskill P, Irwig L. Red flags to screen for malignancy in patients with low-back pain. Cochrane Database Syst Rev.
2013;2:CD008686.
21. Koes BW, van Tulder M, Lin CW, Macedo LG, McAuley J, Maher C. An updated overview of clinical guidelines for the management of non-specific low back pain in primary care. Eur Spine J. 2010 Dec;19(12):2075-94.
22. Newcombe RG. Two‐sided confidence intervals for the single proportion: comparison of seven methods. Statistics in medicine. 1998;17(8):857-72.
23. Henschke N, Maher CG, Refshauge KM. Screening for malignancy in low back pain patients: a systematic review. Eur Spine J. 2007 Oct;16(10):1673-9.
24. Henschke N, Maher CG, Refshauge KM. A systematic review identifies five "red flags" to screen for vertebral fracture in patients with low back pain. J Clin Epidemiol. 2008
Feb;61(2):110-8.
P a g e | 124
25. Bigos S, Bowyer O, Braen G. Acute low back pain in adults. Clinical practice guideline
No. 14. AHCPR Publication No 95. 1994;642.
26. Underwood M, Buchbinder R. Red flags for back pain. BMJ. 2013 Dec 12;347:f7432.
P a g e | 125
Pre-operative Depression is a Risk Factor for Complication and Increased Healthcare Utilization Following Total Shoulder Arthroplasty
Matthew P Lunati, MD; Jacob M Wilson, MD; Kevin X Farley, BA; Michael B Gottschalk, MD; Eric R Wagner, MD
P a g e | 126
ABSTRACT analysis comparing complication and
Introduction: The incidence of total healthcare utilization between cohorts. This shoulder arthroplasty (TSA) continues to analysis controlled for baseline patient increase. While researchers expect demographic, comorbid, and surgical preoperative depression to influence factors. outcomes following primary hip and knee Results: We included 22,623 patients arthroplasty, there is a paucity of data on undergoing TSA in this study. Of these, this relationship after primary TSA. The 3,209 (14%) had a preoperative diagnosis of purpose of this study is to define the depression. Multivariate analysis relationship between a preoperative demonstrated that the following were more diagnosis of depression and postoperative common in depressed patients: sepsis (Odds outcomes following TSA. Ratio (OR) 2.04, 95% Confidence Interval Methods: This is a Level III retrospective (CI) 1.14-3.65, p=0.022), one-year revision cohort study. We identified patients (OR 1.92, 95%CI 1.45-2.55, p<0.001), one- undergoing TSA between 2009 and 2017 year prosthetic joint infection (OR 1.41, OR from the Truven Marketscan® database and 1.04-1.90, p=0.025), return to operating created two cohorts, patients with and room for irrigation and debridement (OR without depression. We included patients 2.72, 95%CI 1.67-4.42, p<0.001), prosthetic who were continuously enrolled in the complication (OR 1.54, 95%CI 1.26-1.88, database for one year pre- and p<0.001), and wound complication (OR postoperatively. We collected demographic 1.84, 95%CI 1.2-2.79, p=0.004). Similarly, data, complication data, and healthcare depressed patients had greater healthcare utilization factors, then performed statistical
P a g e | 127 utilization including higher odds of non- 500%34. A variety of factors affect this home discharge (OR 1.43, 95%CI 1.3-1.57, dramatic increase in utilization of TSA p<0.001), 90-day readmission (OR 1.55, including the aging population,
95%CI 1.3-1.86, p<0.001), 90-day ED visit developments in implant design and
(OR 1.39, 95%CI 1.23-1.57, p<0.001), and operative technique, and progressive extended length of stay (≥3 days; OR 1.23, improvement in patient-reported and clinical
95%CI 1.12-1.36, p<0.001). outcomes11; 18; 29; 35. Given the exponential
increase in utilization, there is an increased Discussion and Conclusions: Depression need for the identification of factors prior to TSA is common and is associated associated with outcomes, complications, with increased risk of complications and and resource utilization following TSA1; 22; increased healthcare utilization following 33; 34. TSA. Determining whether this is a modifiable risk factor requires further Preoperative depression may be one investigation. such factor. The World Health Organization
estimates that within the United States, more Level of Evidence: Level III than 17 million people are depressed (5.9% Keywords: Total Shoulder Arthroplasty; of the population)7. However, compared to Depression; Outcomes; Healthcare the general population, patients undergoing Utilization; Complication; Cost shoulder, hip, and knee replacements have a
INTRODUCTION higher incidence of depression4; 8; 17; 28; 30. In
Between 2000 and 2011, the number patients undergoing hip and knee of total shoulder arthroplasties (TSA) arthroplasty, a preoperative diagnosis of performed nationally increased nearly depression is associated with higher risk for
P a g e | 128 medical complications8; 9; 25; 30; 32, infection5; depression on postoperative outcomes and
6; 25, readmission20, transfusion9, non-home healthcare utilization following total TSA. discharge9; 36, and increased length of stay23; We utilized the Truven Marketscan®
26. While a few studies have evaluated Commercial Claims and Encounters and depression and patient-reported outcomes Medicare Supplemental and Coordination of following TSA40; 42, there remains a paucity Benefit databases (Truven Health, Ann of evidence regarding the association Arbor, MI) to identify patients undergoing between depression, post-operative TSA. These databases include patients with complications, and healthcare utilization private insurance and those with private after TSA4; 28. Medicare insurance supplements. The
database began in 1995 and has grown Therefore, the purpose of this study substantially, now including 240+ million is to define the relationship between a patients. Unlike other large databases that preoperative diagnosis of depression and the include only inpatient data with short-term risk for post-operative complications and follow-up, the Truven database allows for healthcare utilization after TSA. We longitudinal follow-up of continuously hypothesize that depression will be enrolled patients. The database also reports associated with an increased risk for both claims data on inpatient stays, outpatient complications and increased healthcare visits, and pharmaceutical encounters. utilization following TSA.
We searched the Truven database METHODS Data Source and Included Patients from 2009 to 2017 for patients undergoing
This study is a retrospective cohort TSA. We identified these patients using study investigating the effect of preoperative Current Procedural Terminology code 23472
P a g e | 129
(total arthroplasty of glenohumeral joint and ICD-10 diagnosis codes: 3004, 30112, with glenoid and proximal humeral 3090, 3091, 311, 29682, 29620-6, 29630-6, replacement). This includes both anatomic F32, F33, F341, and F432. Table 1. We and reverse TSA. We did not include included patients who had a documented revision procedures and shoulder diagnosis of depression during a healthcare hemiarthroplasty procedures in this analysis. encounter within the one-year preoperative
Additionally, we used International period in the preoperative depression cohort.
Classification of Diseases (ICD) diagnosis We included those who did not in the No codes to identify patients undergoing Depression cohort. arthroplasty for fracture or avascular Following this, we collected baseline necrosis and excluded these patients. patient demographic data for each patient.
Finally, to be further included in analysis, We collected, recorded, and compared the we required patients to have continuous following between cohorts: age, sex, enrollment for one year pre- and geographic region of surgery location, postoperatively. We excluded patients insurance status, and medical comorbidities. without this two-year continuous We collected comorbidities using the enrollment. Elixhauser method16. We collected the
Patient Cohorts and Baseline Patient following comorbidities: obesity, chronic Information kidney disease (CKD), alcohol use disorder, After we established the final patient tobacco use, hypertension (HTN), coronary population using the above criteria, we split artery disease (CAD), congestive heart the patients into two cohorts: 1) Preoperative failure (CHF), rheumatic disease, and Depression and 2) No Depression. We diabetes mellitus (DM). After we calculated determined depression status using ICD-9
P a g e | 130 comorbidity incidences, we compared them insurance, not to be confused with charges) between patient cohorts. for the initial surgery and inpatient hospital
Outcomes of Interest: Complication, stay for each patient and compared the data Healthcare Utilization, and Costs Data between cohorts. We also collected complication data Statistical Analysis for each included patient. We collected the We performed all statistical analysis following complications for the 90-day in this study with SPSS version 25 statistical postoperative period using ICD-9 and ICD- software (IBM Corporation, Armonk, NY). 10 codes: Prosthetic Joint Infection (PJI), We considered a p-value <0.05 significant in Infection (PJI and superficial surgical site all comparisons. Prior to analysis of infection), wound complication, sepsis, and outcomes, we compared baseline patient thromboembolic event. We collected the demographic and comorbidity information need for revision surgery for one year between cohorts using chi-square analysis. following the index procedure. In addition to We subsequently used chi-square analysis to these complications, we also collected and compare the rates of complication and analyzed healthcare utilization factors. healthcare utilization between groups. These included extended length of stay Thereafter, we used a multivariate binomial (LOS) (≥3 days), non-home discharge, 90- logistic regression model to control for day readmission, 90-day all-cause potential confounders and to assess whether emergency department (ED) visit, 90-day or not depression is an independent risk pain-related ED visit, and the receipt of an factor for complication following TSA. This opioid prescription ≥6-months model controlled for all of the factors listed postoperatively. Finally, we collected net in Table 2. We used the same model to costs (actual amount of money paid by assess for an association of preoperative
P a g e | 131 depression with healthcare utilization and disorders (p=0.094), HTN (p=0.324), CHF inpatient care costs. (p=0.462), or DM (p=0.092). Table 2.
RESULTS Depression and Postoperative Baseline Patient Demographic Information Complications Following Primary TSA and Comorbidities Analysis of postoperative medical Using the above criteria, we complications demonstrated that while identified 22,623 patients from the database postoperative sepsis (0.5% v. 0.3%, for inclusion in the study. This included p=0.022) was more common in depressed 19,414 patients (86%) without depression patients, the occurrence of myocardial and 3,209 (14%) patients with a diagnosis of infarction (MI), stroke, venous preoperative depression. Depressed patients thromboembolism (VTE), and pneumonia were more frequently younger, female, had no significant difference between obese, tobacco users, and had higher rates of groups (p>0.05). We confirmed this finding clinically diagnosed anxiety and rheumatic with multivariate analysis where sepsis was disease (p<0.001). Patients with comorbid found to be twice as likely in depressed depression had lower rates of CAD patients (p=0.016). Table 3, 4. (p<0.001). Table 2. While we identified Surgical complications were statistical differences in patient geographic universally more common in the depressed region and insurance status, these were patient within the first year following small and unlikely to be clinically relevant, surgery. This included one-year revision especially given this all-insured cohort. We surgery (p<0.001), one-year PJI (p=0.025), did not identify any difference between return to operating room for irrigation and cohorts regarding CKD (p=0.9), alcohol use debridement (OR p<0.001), prosthetic
P a g e | 132 complication (p<0.001), and wound DISCUSSION complication (p=0.004). Table 3, 4. TSA utilization has markedly
Depression and Postoperative Resource increased over the past two decades34. As an Utilization Following Primary TSA elective procedure, there is a continued need
Univariate analysis demonstrated for the identification of risk factors for that depressed patients had higher rates of complications following surgery. One risk non-home discharge (27.3% v. 21.7%, factor that researchers have extensively p<0.001), 30-day readmission (2.7% v. studied in the hip and knee arthroplasty
1.4%, p<0.001), 90-day readmission (5.8% patient, with comparatively little attention in v. 3.5%, p<0.001), 90-day all-cause shoulder arthroplasty, is preoperative emergency room visit (13.7% v. 10%, depression4; 28; 38. Identifying risk factors for p<0.001), and extended LOS (22.9% v. adverse outcomes is critical for all aspects of
20%, p<0.001). Table 3. patient care. This knowledge may play a role
Multivariate analysis demonstrated in patient and provider decision-making and that these differences were preserved even can guide future attempts at risk reduction. when controlling for demographic and The purpose of this study was to evaluate comorbid conditions. Specifically, we found the relationship between preoperative depressed patients to have higher rates of depression and postoperative surgical non-home discharge (p<0.001), 30-day complications, medical complications, and readmission (p<0.001), 90-day readmission healthcare utilization following TSA.
(p<0.001), 90-day ED visit (p<0.001), and The results of this investigation extended LOS (≥3 days; p<0.001). Table 4. suggest that a preoperative diagnosis of
depression is an independent risk factor for
P a g e | 133 medical and surgical complications reported, but much higher than the 4.4% that following TSA. Specifically, we found that Bot et al. previously reported in their depressed patients had elevated odds of shoulder arthroplasty cohorts4. Of note, the sepsis, one-year revision surgery, one-year prevalence of depression found in our cohort
PJI, return to operating room for irrigation is much higher than the established general and debridement, prosthetic complication, population prevalence of 5.9%7. Our and wound complications. We additionally cohort’s higher rate of depression, when found that depressed patients had higher compared to prior literature, is likely rates of non-home discharge, 30-day multifactorial and may be attributed to readmission, 90-day readmission, 90-day differences in methodology, as well as the
ED visit, and extended LOS. We also found analysis of an all-insured population. For that postoperative rates of most medical instance, our study protocol utilized complications (including MI, stroke, VTE, diagnosis codes and included those with a and pneumonia) were equivalent between diagnosis within one year of surgery, groups. These findings are novel and contrasting prior work that used point-of- important to understand, especially as value- care surveys immediately preoperatively. based payment models emerge. Newly This all-insured population likely has greater developed risk adjustment models, therefore, exposure to the medical field and, therefore, should consider the above findings. these patients may be more likely to obtain a
diagnosis of depression. In support of this We found that depression was a notion, the incidence of depression found in preoperative comorbid condition in 14% of this cohort is similar to that reported in an our studied patients. This is similar to the all-insured revision lower extremity incidence of 12.4% that Mollon et al.28
P a g e | 134 arthroplasty population41. Still, given our Mollon et al. used a database which allows reliance on clinically documented the capture of only inpatient complications depression, we likely underestimate the true and lacks any long-term follow-up, limiting prevalence among this patient population. possible conclusions28. They, too, included a
Despite this, our cohort of depressed cohort consisting primarily of the early years patients was more likely to consist of young, of reverse TSA when utilization of this obese females – similar to prior work4; 8; 28; design was much lower than in our modern
30; 41. cohort28. In contrast, our cohort includes a
contemporary cohort with one-year These findings confirm prior continual follow-up, allowing for much literature and expand on previously reported stronger conclusions. With improvements in results in shoulder arthroplasty4; 28. Prior TSA implant design and technique over the studies examining the effect of past decade, it is important to understand the psychological illness and outcomes relationship between depression and following TSA are limited in their outcomes in a contemporary cohort1; 18; 35. methodology. For example, Bot et al. Our study additionally examined healthcare examined a large population of patients but utilization such as postoperative ED visits, included both hemiarthroplasty and TSA as an element prior work did not include. Still, well as a heterogenous group of psychiatric the current investigation corroborates the illnesses with only short-term follow-up4. findings of these prior studies – that Additionally, they reported on a dated preoperative depression is a risk factor for cohort (1990–2007) which included only infection28, non-home discharge4; 28, and four years post-Federal Drug Administration postoperative adverse events4. While studies approval of the reverse TSA. Similarly,
P a g e | 135 have shown that depression also negatively and increased revision surgery among influences patient-reported outcomes patients with preoperative depression. We
(specifically, that depression is an have also observed these findings following independent predictor of less improvement lower-extremity arthroplasty as well as those of the American Shoulder and Elbow in other surgical fields5; 6; 10; 12; 21; 41.
Surgeons score), the methodology of the Despite these findings, one must current study does not allow for the analysis understand that the relationship between of functional outcomes40. depression and outcomes following
Follow-up patient evaluations have arthroplasty is likely complex. It is well- repeatedly demonstrated a relationship established that chronic pain is a risk factor between infection and depression following for psychological manifestation in the form arthroplasty procedures25; 28; 41. While this of depression and anxiety27. Importantly, relationship may seem obscure, while Werner and colleagues found that psychological distress produces a systemic patients with depression improved less than state of inflammation leading to those without depression, depressed patients preoccupation of the immune system and a still had significant improvement in clinical resultant susceptible host state19. Other work outcomes scores40. Post-arthroplasty pain has also demonstrated that depression may improvement has been associated with influence T-cell phenotype, further improvement in depression symptoms after propagating immune susceptibility14; 15. This total hip and knee arthroplasty2; 3; 13; 24; 37. likely explains our finding of increased PJI Therefore, patients with depression should and subsequent return to operating room for clearly have access to surgery, even with irrigation and debridement, increased sepsis, their increased risk profile. Ultimately, in an
P a g e | 136 era of shared decision-making, surgeons an improvement in depressive symptoms should discuss our results with patients who and manifestations31. have depression as a comorbid condition. While this study has many strengths, We recommend that surgeons use our results there are also multiple limitations, and we to more accurately counsel patients and to must interpret the results of the investigation better inform their decisions. Additionally, with these in mind. First, as is the case with while our results cannot comment directly any analysis of large databases, our cohort is on the modifiability of depression as a risk subject to the complete and accurate coding factor, it seems prudent to optimize a of depressed patients. While we identified a patient’s mental health to the greatest extent higher proportion of patients with possible prior to intervening surgically. depression than prior studies, we likely still Importantly, surgeons should likely tailor underestimate the true incidence of preoperative counseling of patients to better depression among the TSA patient serve this specific cohort. It is our opinion population. For instance, prior studies using that clearly establishing patient appropriate questionnaires as the diagnostic criteria for expectations with regard to pain, pain depression have reported higher prevalence control, and postoperative function may help of depression (50%) within a population of improve satisfaction. We may best patients undergoing shoulder surgery for accomplish this through a multidisciplinary instability39. Additionally, our examined approach. Again, surgeons should keep in patient population represents an all-insured mind that comorbid conditions may cohort. While this may limit the exacerbate or cause depression, and generalizability, it does allow for the improving this baseline issue could lead to elimination of what is likely significant
P a g e | 137 confounding between depression, results of this study highlight the impact of socioeconomic status, and insurance status28. depression on outcomes and have the
Further, given our use of a large database, potential to drive risk adjustment models to we lost some granularity in patient care. For ensure that providers are fairly assessed on instance, we are unable to delineate the quality metrics. timing and extent of clinical depression management that patients received. We were additionally unable to quantify the severity of depression. Lastly, the retrospective nature of the study limits our ability to make definitive conclusions regarding causality.
In conclusion, patients with a preoperative diagnosis of depression have higher risk for multiple surgical complications and increased healthcare utilization after TSA. Patient counseling within the depressed patient population should highlight the increased risk of post- operative complications in order to define appropriate expectations. Determining whether this is a modifiable risk factor will require further investigation. As we strive to provide value in healthcare services, the
P a g e | 138
Table 1. International Classification of Diseases (ICD) 9 and 10 diagnostic codes to identify depression Code Description 3004 Dysthymic disorder 30112 Chronic depressive personality disorder 3090 Adjustment disorder with depressed mood 3091 Prolonged depressive reaction 311 Depressive disorder, not elsewhere classified 29682 Atypical depressive disorder 29620 Major depressive affective disorder, single episode, unspecified 29621 Major depressive affective disorder, single episode, mild 29622 Major depressive affective disorder, single episode, moderate Major depressive affective disorder, single episode, severe, without mention of 29623 psychotic behavior Major depressive affective disorder, single episode, severe, specified as with psychotic 29624 behavior 29625 Major depressive affective disorder, single episode, in partial or unspecified remission 29626 Major depressive affective disorder, single episode, in full remission 29630 Major depressive affective disorder, recurrent episode, unspecified 29631 Major depressive affective disorder, recurrent episode, mild 29632 Major depressive affective disorder, recurrent episode, moderate Major depressive affective disorder, recurrent episode, severe, without mention of 29633 psychotic behavior Major depressive affective disorder, recurrent episode, severe, specified as with 29634 psychotic behavior Major depressive affective disorder, recurrent episode, in partial or unspecified 29635 remission 29636 Major depressive affective disorder, recurrent episode, in full remission F32 Major depressive disorder, single episode F33 Major depressive disorder, recurrent F341 Dysthymic disorder F432 Adjustment disorder with depressed mood
P a g e | 139
Table 2. Demographics and comorbidities of patients with and without depression No P- Demographic Depression Depression Value Age Group <55 1,652 (8.5%) 367 (11.4%) <0.001 5,807 1,252
55-64 (29.9%) (39.0%) 6,141
65-74 (31.6%) 967 (30.1%) 5,084
75-84 (26.2%) 549 (17.1%) 85+ 730 (3.8%) 74 (2.3%) Sex 10,106 1,039 Male (52.1%) (32.4%) <0.001 9,308 2,170
Female (47.9%) (67.6%) Region 3,090 Northeast (15.9%) 475 (14.8%) 0.036 6,571 1,111
Midwest (33.8%) (34.6%) 6,490 1,025
South (33.4%) (31.9%) 3,010
West (15.5%) 556 (17.3%) Insurance 5,442 Comprehensive (28.7%) 734 (23.6%) <0.001 9,774 1,642
Preferred Provider Organization (51.6%) (2.9%) Health Maintainence Organization 1,834 (9.7%) 399 (12.9%) Point-of-service 994 (5.2%) 168 (5.4%) High Deductible Health Plans 892 (4.7%) 162 (5.2%) Comorbidities 3,083 Obesity (15.9%) 713 (22.2%) <0.001 Chronic Kidney Disease 1,259 (6.5%) 210 (6.5%) 0.900 Alcohol Use Disorders 181 (0.9%) 69 (2.2%) 0.094 Tobacco Use 1,304 (6.7%) 296 (9.2%) <0.001 13,511 2,261 Hypertension (69.6%) (70.5%) 0.324
P a g e | 140
3,876 Coronary Artery Disease (20.0%) 532 (16.6%) <0.001 Anxiety 1,219 (6.3%) 443 (13.8%) <0.001 Congestive Heart Failure 1,106 (5.2%) 178 (5.2%) 0.462 Rheumatic Disease 1,171 (6.0%) 241 (7.5%) 0.001 4,417 3,209 Diabetes (22.8%) (21.5%) 0.092
P a g e | 141
Table 3. Univariate Analysis of Complications Complication No Depression Depression P-Value Surgical Complications Revision Surgery* 222 (1.1%) 3,133 (2.3%) <0.001 Prosthetic Joint Infection* 274 (1.4%) 60 (1.9%) 0.046 Return to OR for I & D** 67 (0.3%) 26 (0.8%) <0.001 Prosthetic Complication 561 (2.9%) 141 (4.4%) <0.001 Wound Complication 116 (0.6%) 32 (1.0%) 0.082 Medical Complications Sepsis 55 (0.3%) 17 (0.5%) 0.022 Myocardial Infarction 175 (0.9%) 34 (1.1%) 0.386 Stroke 336 (1.7%) 59 (1.8%) 0.666 Thromboembolic Event 483 (2.5%) 93 (2.9%) 0.172 Pneumonia 97 (0.5%) 14 (0.4%) 0.634 Resource Utilization Non-home Discharge 21.7% (4,064) 846 (27.3%) <0.001 30-Day Readmission 278 (1.4%) 87 (2.7%) <0.001 90-Day Readmission 680 (3.5%) 186 (5.8%) <0.001 ED Visit 1,931 (10.0%) 438 (13.7%) <0.001 Extended LOS (>2 Days) 3,876 (20.0%) 735 (22.9%) <0.001 * Revision surgery and PJI displayed with 1-year follow up, all other complications displayed for 90 days post-operative follow up; ** OR = Operating room, I&D = irrigation and debridement
P a g e | 142
Table 4. Adjusted Risks of Complications
Complication Odds Ratio P-Value Surgical Complications Revision Surgery* 1.92 (1.45-2.55) <0.001 Prosthetic Joint Infection* 1.41 (1.04-1.90) 0.025 Return to OR for I & D** 2.72 (1.67-4.42) <0.001 Prosthetic Complication 1.54 (1.26-1.88) <0.001 Wound Complication 1.84 (1.20-2.79) 0.004 Medical Complications Sepsis 2.04 (1.14-3.65) 0.016 Myocardial Infarction 1.23 (0.82-1.83) 0.314 Stroke 1.34 (0.99-1.80) 0.056 Thromboembolic Event 1.19 (0.93-1.51) 0.163 Pneumonia 0.76 (0.42-1.37) 0.358 Resource Utilization Non-home Discharge 1.429 (1.30-1.57) <0.001 30-Day Readmission 1.748 (1.35-2.27) <0.001 90-Day Readmission 1.55 (1.30-1.86) <0.001 ED Visit 1.39 (1.23-1.57) <0.001 Extended LOS (≥3 Days) 1.23 (1.12-1.36) <0.001 * Revision surgery and PJI displayed with 1-year follow up, all other complications displayed for 90 days post-operative follow up; ** OR = Operating room, I&D = irrigation and debridement
P a g e | 143
References:
1. Anakwenze OA, Yehyawi T, Dillon MT, Paxton E, Navarro R, Singh A. Effect of Age on Outcomes of Shoulder Arthroplasty. The Permanente journal 2017;21:16-056. 10.7812/tpp/16-056
2. Badura-Brzoza K, Zajac P, Kasperska-Zajac A, Brzoza Z, Matysiakiewicz J, Piegza M et al. Anxiety and depression and their influence on the quality of life after total hip replacement: preliminary report. Int J Psychiatry Clin Pract 2008;12:280-284. 10.1080/13651500802095012
3. Blackburn J, Qureshi A, Amirfeyz R, Bannister G. Does preoperative anxiety and depression predict satisfaction after total knee replacement? Knee 2012;19:522-524. 10.1016/j.knee.2011.07.008
4. Bot AG, Menendez ME, Neuhaus V, Ring D. The influence of psychiatric comorbidity on perioperative outcomes after shoulder arthroplasty. Journal of shoulder and elbow surgery 2014;23:519-527. 10.1016/j.jse.2013.12.006
5. Bozic KJ, Lau E, Kurtz S, Ong K, Berry DJ. Patient-related risk factors for postoperative mortality and periprosthetic joint infection in medicare patients undergoing TKA. Clin Orthop Relat Res 2012;470:130-137. 10.1007/s11999-011-2043-3
6. Bozic KJ, Lau E, Kurtz S, Ong K, Rubash H, Vail TP et al. Patient-related risk factors for periprosthetic joint infection and postoperative mortality following total hip arthroplasty in Medicare patients. J Bone Joint Surg Am 2012;94:794-800. 10.2106/JBJS.K.00072
7. Brody DJ, Pratt LA, Hughes JP. Prevalence of Depression Among Adults Aged 20 and Over: United States, 2013-2016. NCHS data brief 2018:1-8.
8. Browne JA, Sandberg BF, D'Apuzzo MR, Novicoff WM. Depression is associated with early postoperative outcomes following total joint arthroplasty: a nationwide database study. The Journal of arthroplasty 2014;29:481-483. 10.1016/j.arth.2013.08.025
9. Buller LT, Best MJ, Klika AK, Barsoum WK. The influence of psychiatric comorbidity on perioperative outcomes following primary total hip and knee arthroplasty; a 17-year
P a g e | 144
analysis of the National Hospital Discharge Survey database. J Arthroplasty 2015;30:165-170. 10.1016/j.arth.2014.08.034
10. Chang SM, Parney IF, McDermott M, Barker FG, 2nd, Schmidt MH, Huang W et al. Perioperative complications and neurological outcomes of first and second craniotomies among patients enrolled in the Glioma Outcome Project. J Neurosurg 2003;98:1175- 1181. 10.3171/jns.2003.98.6.1175
11. Dillon MT, Chan PH, Inacio MCS, Singh A, Yian EH, Navarro RA. Yearly Trends in Elective Shoulder Arthroplasty, 2005-2013. Arthritis care & research 2017;69:1574- 1581. 10.1002/acr.23167
12. Doering LV, Cross R, Vredevoe D, Martinez-Maza O, Cowan MJ. Infection, depression, and immunity in women after coronary artery bypass: a pilot study of cognitive behavioral therapy. Altern Ther Health Med 2007;13:18-21.
13. Duivenvoorden T, Vissers MM, Verhaar JA, Busschbach JJ, Gosens T, Bloem RM et al. Anxiety and depressive symptoms before and after total hip and knee arthroplasty: a prospective multicentre study. Osteoarthritis Cartilage 2013;21:1834-1840. 10.1016/j.joca.2013.08.022
14. Elenkov IJ. Systemic stress-induced Th2 shift and its clinical implications. Int Rev Neurobiol 2002;52:163-186.
15. Elenkov IJ, Chrousos GP. Stress Hormones, Th1/Th2 patterns, Pro/Anti-inflammatory Cytokines and Susceptibility to Disease. Trends Endocrinol Metab 1999;10:359-368.
16. Elixhauser A, Steiner C, Harris DR, Coffey RM. Comorbidity measures for use with administrative data. Med Care 1998;36:8-27.
17. Ellis HB, Howard KJ, Khaleel MA, Bucholz R. Effect of psychopathology on patient- perceived outcomes of total knee arthroplasty within an indigent population. The Journal of bone and joint surgery American volume 2012;94:e84. 10.2106/jbjs.K.00888
18. Gerber C, Canonica S, Catanzaro S, Ernstbrunner L. Longitudinal observational study of reverse total shoulder arthroplasty for irreparable rotator cuff dysfunction: results after 15
P a g e | 145
years. Journal of shoulder and elbow surgery 2018;27:831-838. 10.1016/j.jse.2017.10.037
19. Ghoneim MM, O'Hara MW. Depression and postoperative complications: an overview. BMC Surg 2016;16:5. 10.1186/s12893-016-0120-y
20. Gold HT, Slover JD, Joo L, Bosco J, Iorio R, Oh C. Association of Depression With 90- Day Hospital Readmission After Total Joint Arthroplasty. J Arthroplasty 2016;31:2385- 2388. 10.1016/j.arth.2016.04.010
21. Gordon RJ, Weinberg AD, Pagani FD, Slaughter MS, Pappas PS, Naka Y et al. Prospective, multicenter study of ventricular assist device infections. Circulation 2013;127:691-702. 10.1161/CIRCULATIONAHA.112.128132
22. Groh GI, Groh GM. Complications rates, reoperation rates, and the learning curve in reverse shoulder arthroplasty. Journal of shoulder and elbow surgery 2014;23:388-394. 10.1016/j.jse.2013.06.002
23. Halawi MJ, Chiu D, Gronbeck C, Savoy L, Williams VJ, Cote MP. Psychological Distress Independently Predicts Prolonged Hospitalization After Primary Total Hip and Knee Arthroplasty. J Arthroplasty 2019;34:1598-1601. 10.1016/j.arth.2019.03.063
24. Jones AR, Al-Naseer S, Bodger O, James ETR, Davies AP. Does pre-operative anxiety and/or depression affect patient outcome after primary knee replacement arthroplasty? Knee 2018;25:1238-1246. 10.1016/j.knee.2018.07.011
25. Klement MR, Nickel BT, Penrose CT, Bala A, Green CL, Wellman SS et al. Psychiatric disorders increase complication rate after primary total knee arthroplasty. Knee 2016;23:883-886. 10.1016/j.knee.2016.05.007
26. March MK, Harmer AR, Dennis S. Does Psychological Health Influence Hospital Length of Stay Following Total Knee Arthroplasty? A Systematic Review. Arch Phys Med Rehabil 2018;99:2583-2594. 10.1016/j.apmr.2018.03.020
27. Merskey H. Psychological aspects of pain. Curr Med Res Opin 1974;2:515-520. 10.1185/03007997409115252
P a g e | 146
28. Mollon B, Mahure SA, Ding DY, Zuckerman JD, Kwon YW. The influence of a history of clinical depression on peri-operative outcomes in elective total shoulder arthroplasty: a ten-year national analysis. The bone & joint journal 2016;98-b:818-824. 10.1302/0301- 620x.98b6.37208
29. Norris TR, Iannotti JP. Functional outcome after shoulder arthroplasty for primary osteoarthritis: a multicenter study. Journal of shoulder and elbow surgery 2002;11:130- 135.
30. Pan X, Wang J, Lin Z, Dai W, Shi Z. Depression and Anxiety Are Risk Factors for Postoperative Pain-Related Symptoms and Complications in Patients Undergoing Primary Total Knee Arthroplasty in the United States. The Journal of arthroplasty 2019. 10.1016/j.arth.2019.05.035
31. Rasouli MR, Menendez ME, Sayadipour A, Purtill JJ, Parvizi J. Direct Cost and Complications Associated With Total Joint Arthroplasty in Patients With Preoperative Anxiety and Depression. J Arthroplasty 2016;31:533-536. 10.1016/j.arth.2015.09.015
32. Saltzman BM, Chalmers PN, Gupta AK, Romeo AA, Nicholson GP. Complication rates comparing primary with revision reverse total shoulder arthroplasty. Journal of shoulder and elbow surgery 2014;23:1647-1654. 10.1016/j.jse.2014.04.015
33. Schairer WW, Nwachukwu BU, Lyman S, Craig EV, Gulotta LV. National utilization of reverse total shoulder arthroplasty in the United States. Journal of shoulder and elbow surgery 2015;24:91-97. 10.1016/j.jse.2014.08.026
34. Simovitch RW, Friedman RJ, Cheung EV, Flurin PH, Wright T, Zuckerman JD et al. Rate of Improvement in Clinical Outcomes with Anatomic and Reverse Total Shoulder Arthroplasty. The Journal of bone and joint surgery American volume 2017;99:1801- 1811. 10.2106/jbjs.16.01387
35. Stundner O, Kirksey M, Chiu YL, Mazumdar M, Poultsides L, Gerner P et al. Demographics and perioperative outcome in patients with depression and anxiety undergoing total joint arthroplasty: a population-based study. Psychosomatics 2013;54:149-157. 10.1016/j.psym.2012.08.009
P a g e | 147
36. Tarakji BA, Wynkoop AT, Srivastava AK, O'Connor EG, Atkinson TS. Improvement in Depression and Physical Health Following Total Joint Arthroplasty. J Arthroplasty 2018;33:2423-2427. 10.1016/j.arth.2018.03.051
37. Wagner ER WK, Chang MJ, Woodmass JM, Borque KA, Warner JP. Common Pain Generators After Total Shoulder Arthroplasty and Their Pesky Relatives. Seminars in Arthroplasty 2017;28:166-174.
38. Weekes DG, Campbell RE, Giunta NJ, Pepe MD, Tucker BS, Ciccotti MG et al. Prevalence of Clinical Depression among Patients after Shoulder Stabilization Repair: A Prospective Study. Orthopaedic journal of sports medicine 2019;7:2325967119S2325900259. 10.1177/2325967119S00259
39. Werner BC, Wong AC, Chang B, Craig EV, Dines DM, Warren RF et al. Depression and Patient-Reported Outcomes Following Total Shoulder Arthroplasty. The Journal of bone and joint surgery American volume 2017;99:688-695. 10.2106/jbjs.16.00541
40. Wilson JM FK, Erens G, Bradbury TL, Guild GN. Preoperative Depression Is Associated with Increased Risk Following Revision Total Joint Arthroplasty. Journal of Arthroplasty 2019;IN PRESS.
41. Wong SE, Colley AK, Pitcher AA, Zhang AL, Ma CB, Feeley BT. Mental health, preoperative disability, and postoperative outcomes in patients undergoing shoulder arthroplasty. Journal of shoulder and elbow surgery 2018;27:1580-1587. 10.1016/j.jse.2018.02.066
P a g e | 148
Opioid Use Following Total Shoulder Arthroplasty: Who Requires Refills and for How Long?
Jeremiah Pflederer, MD; Corey Spencer, Jacob M Wilson, MD, Alex Dawes, Charles Daly, MD; Michael B Gottschalk, MD; Eric R Wagner, MD
P a g e | 149
Abstract only included if they were greater than 18 years
Introduction: Pain control following a total old, were not being treated for a fracture, and had shoulder arthroplasty (TSA) is directly related to continuous enrollment in the database for 6 successful recovery and post-surgical outcomes. months preoperatively and 1 year postoperatively.
The current standard of care includes the Opioid data was collected using National Drug utilization of a multimodal analgesic approach Codes (NDC) from outpatient pharmacy claims. including nerve blockade, anti-inflammatories, Only opioid naïve patients, defined as zero opioid and breakthrough prescription opioid medication prescriptions filled within 6 months in an effort to provide postoperative analgesia. preoperatively, were included. Opioid
While this original opioid prescription is prescriptions filled between the 6 weeks before sufficient for the majority of patients, some the procedure and 2 weeks after were designated patients go on to require additional refills and a as original prescriptions and assumed to be few end up having prolonged opioid use. Our related to their operation. Patients were then study investigated patient risk factors associated grouped into one of three cohorts based on with opioid refill post-surgery. We hypothesized postoperative opioid use: 1) Patients with no that current tobacco or alcohol use, clinical additional refills, 2) Patients with at minimum of depression, and diabetes will all increase the risk one additional refill up through 6 months of additional opioid refills. postoperatively, and 3) Patients with additional
Methods: : Using the Truven Marketscan® refills and continued opioid use past 6 months. national insurance database, all patients who One-way ANOVA with post-hoc comparison and underwent a TSA from 2010 to 2017 were multivariate regression were run for statistical identified using the common Current Procedural analysis. Odds ratios with 95% confidence
Terminology (CPT) code 23472. Patients were
P a g e | 150 intervals were calculated for variables with p depression, tobacco use, diabetes, and rheumatic value <0.05. disease. Depressed patients were at the highest
risk for prolonged opioid use (OR 1.59, CI 1.38- Results: A total of 17,706 opioid naïve patients 1.84, p<0.001) followed by current tobacco users were included, 10,882 (61.5%) did not have any (OR 1.51, CI 1.23-1.85, p<0.001) (Table 2) additional refills, 4,473 (25.3%) had an additional prescription within 6 months after surgery, and Discussion: We observed a dose-dependent
2,351 (13.3%) had prolonged opioid use after 6 relationship between original opioid prescription months postoperatively. Upon comparison of data and number of additional refills needed, original opioid prescription data it was found that suggesting that initial overprescribing may lead to the prolonged use group were initially prescribed prolonged dependency. We also found several an average of 12.9 more pills than the no refill patient demographic factors and comorbidities group and 7.36 more than the refill group that significantly increase the odds of requiring
(p<0.001). This same trend was also seen when additional opioid refills following TSA and comparing total OME with the prolonged use recommend medical optimization of group being prescribed an average of 150 more comorbidities and cessation of modifiable risk
OMEs than those patients who had no refills and factors prior to surgery. Complete discussion with nearly 100 more OMEs than those in the refill patients regarding their specific risk, along with group. (Table 1). On multivariate analysis, appropriate opioid prescribing are necessary to patients older than 75 were determined to be at help minimize the need for additional refills. half the risk of prolonged use compared to those Level of Evidence: Prognostic, Level III between 18 and 54 years old (p<0.001). Female Introduction gender was shown to increase odds for additional The opioid epidemic in the United States refills along with coronary artery disease, clinical is well established and has become a source of
P a g e | 151 extensive medical research, but despite this risk factors for opiate refill after TSA. Pain intense scrutiny it has received, the opioid control following total shoulder arthroplasty epidemic continues.1-3 Recent epidemiology (TSA) is directly related to successful recovery studies have demonstrated a near 3-fold increase and postoperative outcomes.14 The current in the number of people with an opioid use standard of care includes multimodal analgesia disorder.4-7 Orthopedic surgeons are among the including nerve blockade, anti-inflammatories, leading prescribers of opiate medications, and and a breakthrough prescription of opiate pain therefore orthopedists have a substantial medication.15 While the majority of patients opportunity for improved opioid stewardship.8 achieve sufficient pain control with their original
Elective orthopedic procedures can be extremely prescription, some go on to require additional effective pain-relieving interventions, however, refills or progress to prolonged opioid use.16,17 breakthrough narcotic pain medications remain Identifying these patients is critical as a single the standard for management of acute postsurgical refill has been shown to increase the postoperative pain. Unfortunately, for the opiate potential of misuse by more than 40%.10 naïve patient, orthopedic surgery, and its Preoperative opiate use is well established associated opioid exposure can be a gateway to as a risk factor for prolonged post-operative prolonged opiate use.9,10 Therefore, developing a opiate use throughout a range of orthopedic procedure- and patient-specific understanding of procedures, including total shoulder risk factors for opiate misuse is essential to guide arthroplasty.4,17-19 Previous studies in hip and prescribing practices. knee arthroplasty, as well as those in the general
Despite the fact that the incidence of TSA surgery literature, have identified younger age, in the United States is increasing exponentially11- smoking, alcohol abuse, mental health disorders
13, there remains a paucity of data investigating (depression/anxiety) and larger initial prescription
P a g e | 152 as risk factors for prolonged opiate use after This is a retrospective review of the surgery in the opiate naïve patient.6,20,21 However, Truven Health MarketScan Commercial Claims the risk factors for prolonged use following and Encounters and Medicare Supplemental and primary TSA are not well-defined , as the Coordination of Benefit databases (Truven majority of focus in the literature has been on Health, Ann Arbor, MI). The database was non-arthroplasty hand and upper extremity queried for the years 2010 through 2017. The procedures.22-24 Stratifying patient risk prior to Truven database is a commercially available, surgery is essential so that prescription strategies nationwide insurance database that includes may be optimized to decrease the risk of information on over 240 million patients and iatrogenic opioid dependence. offers the ability to track patients who have
continual enrollment longitudinally. Additionally, In this study, we sought to identify the Truven database is advantageous in that it patient-specific risk factors associated with includes information on outpatient opioid refill and prolonged postoperative use pharmaceutical claims. following TSA. We additionally examined the relationship between the magnitude of initial We queried the database for patients opiate prescription and incidence of refill. We undergoing Total Shoulder Arthroplasty (TSA) hypothesized that current tobacco or alcohol use, using the Current Procedural Terminology clinical depression, and comorbid medical (CPT)code 23472 (Arthroplasty, glenohumeral conditions would increase the risk of subsequent joint; total shoulder (glenoid and proximal opioid refills. humeral replacement). Patients <18 years old and
those undergoing TSA for the treatment of a Methods fracture were excludedfrom the study. To allow Data Source and Cohort Selection sufficient time to collect preoperative data,
P a g e | 153 comorbidities, and postoperative prescription an original opioid prescription were excluded data, only patients who were continuously from the study population. A complete selection enrolled for at least 6 months preoperatively and algorithm is outlined in Figure 2.
12 months postoperatively were included. This Baseline Patient Data 18-month study period was broken down into 4 Baseline patient demographic separate periods: 1) pre-operative period, 2) information, comorbidities, and medication data surgical period, 3) refill period, and 4) prolonged were collected for the preoperative period. use period. These study periods are shown to Demographics, including age and sex, along with scale and thoroughly defined in Figure 1. the following comorbidities were recorded and Given the aim of the study, we only analyzed: obesity (defined as BMI ≥ 30), chronic included patients who were “opioid naïve” which kidney disease, alcohol use disorders, tobacco we defined as any patient with no filled opioid use, hypertension, hyperlipidemia, coronary prescriptions during the pre-operative period (6 artery disease, congestive heart failure, diabetes, months to 6 weeks prior to surgery). Opioid rheumatic disease, depression, and anxiety. prescriptions filled during the surgical period (6 Each patient’s opioid pain medication weeks before surgery through the 2 weeks after) data for the entire 18-month study period was were designated as original prescriptions and it collected using the National Drug Codes (NDCs) was assumed that the intended use of these included in the Truven database. The use of prescriptions was for procedure-related pain NDCs in large database studies has been widely control. To control for the possibility of patients utilized.25,26 After isolating the opioid therapeutic not filling their prescription through their class (i.e. hydrocodone, oxycodone, morphine, insurance plan (eg. Using a discount pharmacy or hydromorphone, codeine, etc.), we recorded the rebate program (GoodRx)), those patients without prescription dose and quantity. We then used this
P a g e | 154 information to calculate oral morphine above), we used chi-square analysis to compare equivalents (OME) according to publicly baseline characteristics and comorbidities available opioid conversion tables.27 between groups. We then performed multiple 2-
sample t-tests to compare original opioid Postoperative Outcomes prescription data including quantity of pills and Patients were divided into one of three OMEs between the cohorts. Binomial logistic separate cohorts based on their opioid use during regression was then used to assess independent the postoperative periods defined in Figure 1. The risk factors associated with additional opioid first cohort, designated as “No Refill”, included refills, while controlling for the baseline those patients who did not fill an opioid demographic and comorbid data included in prescription outside of the surgical period. The Table 1. For this analysis, we compared the refill second, “Refill”, cohort included patients who and prolonged use groups to the no refill cohort filled at least one additional opioid prescription as reference. We are reporting results as odds during the refill period, lasting up to 6 months ratios (OR) with 95% confidence intervals (CI). after surgery. The third, “Prolonged Use” cohort A p-value of <0.05 was selected as representing included those patients who filled at least one significance for this study. All statistical analysis opioid prescription during both the refill period were done using SAS version 9.4 (SAS, Cary, and prolonged use period, thus indicating NC) statistical software. continued opioid use for longer than 6 months Results after TSA. Baseline Demographic Data and Comorbidities Statistical Analysis From 2010 to 2017 we identified 17,706 After allotting patients to their respective patients meeting our specified inclusion and cohorts (based on postoperative opioid use, as exclusion criteria from the Truven database
P a g e | 155
(Figure 2). Those patients were then separated prescribed increased with additional refills, with into the three cohorts as described above and the an average difference of 12.9 pills between the no distribution of patients is shown in Table 1. The refill group and the prolonged use group majority (61.5%) of patients who underwent TSA (p<0.001). This same trend was also seen when did not require any additional opioid refill after comparing total OME. The prolonged use group their original prescription. was prescribed an average of 150 more OMEs
than those patients who had no refills and nearly Upon chi-squared analysis, multiple 100 more OMEs than those in the refill group. differences in baseline demographics and Table 2. comorbidities were noted between groups. The proportion of females in each group increased Risk Factor Analysis with additional prescriptions while average Subsequently, we performed multivariate patient age slightly decreased. There were also analysis, controlling for the variables listed in significant differences in several examined Table 1, in order to determine independent risk comorbidities including obesity, tobacco use, factors associated with additional opioid refills. coronary artery disease, diabetes, rheumatic Age was determined to decrease risk with disease, depression, and anxiety. While some of patients 75 years old or older shown to be at half these differences were small, the rate of each the risk of prolonged use compared to those condition was universally higher in the prolonged between 18 and 54 years old (p<0.001). Female use group. Table 1. gender was shown to increase odds for additional
Original Opioid Prescription refills (Refill group OR 1.20, CI 1.12-1.29,
p<0.001; Prolonged use group OR 1.37, CI 1.25- We then compared original opioid 1.51, p<0.001). Comorbidities associated with prescription data filled during the surgical period increased odds of requiring an additional refill up between the groups. The average quantity
P a g e | 156 through 6 months postoperatively were coronary and the majority of these prescriptions are not artery disease and clinical depression (OR 1.10, provided by orthopedists.30,31 However, in the
CI 1.00-1.21, p=0.042; OR 1.29, CI 1.14-1.45, opioid naïve patient, surgery, and its associated p<0.001; respectively). These comorbidities were opioid exposure is known to be a potential also risk factors for prolonged opioid use, along gateway to prolonged opioid use in the setting of with tobacco use, diabetes, and rheumatic inappropriate prescribing habits.9 This iatrogenic disease. Depressed patients were at the highest opioid dependence is particularly concerning and risk for prolonged opioid use (OR 1.59, CI 1.38- provides a clinical opportunity for prevention. In
1.84, p<0.001) followed by current tobacco users this study, we sought to identify the rate of and
(OR 1.51, CI 1.23-1.85, p<0.001). Table 3. risk factors for prolonged opioid use following
Discussion TSA in the opioid naïve patient. Better
The opioid epidemic in the United States delineating patient-specific risk factors for is well documented and addressing this epidemic prolonged opioid use should assist surgeons in has been resource demanding.1,2 While the the early identification of the at-risk patient and medical management of chronic pain is difficult, provide opportunity for early clinical evidence suggests that management with chronic intervention. opioids is ineffective and potentially harmful.28 The results of this investigation
Elective orthopedic procedures, such as TSA, are demonstrated a dose dependent relationship known to be effective pain-relieving procedures between original opioid prescription and the need that provide good functional outcomes.29 Given for an opioid refill as well as for prolonged opioid the nature of these procedures and the pathology use. On average, patients that obtained a refill they address, many patients presenting for and went on to prolonged use were initially arthroplasty procedures have, unfortunately, prescribed 6 and 13 more pills respectively. already been prescribed opioids preoperatively Historically, surgeons have based opiate
P a g e | 157 prescriptions on anecdotal experience and, at orthopedic and general surgery procedures and times, prescriber convenience. The common analyzed the impact of initial prescription assumption has been that larger initial duration on probability of a refill. They found prescriptions will decrease the burden of patient that following musculoskeletal procedures, as phone calls and refill requests32,33. Our results prescription duration increased past 15 days refill contradict this assumption, suggesting that rates increased incrementally.6 Another recent patients who are given more opiates will request database study with a cohort of 1,372 opiate and utilize more opiates. We also identified naïve total knee arthroplasty patients found that several patient-specific risk factors that were patients who required a refill initially received a associated with increased odds of requiring 46.7% larger opioid prescription.20 The mounting additional opioid refills following TSA. evidence that patients who are given more opiates
Specifically, younger patient age, female gender, will request more opiates highlights the coronary artery disease, and depression each importance of managing patient expectations for independently increased the odds of refill and pain management preoperatively and to provide prolonged use. Additionally, tobacco use, counseling to minimize the risk of iatrogenic diabetes, and rheumatic disease significantly dependence.34 Furthermore, studies increase the odds of prolonged opiate use. demonstrating that magnitude of opioids provided
This is the first study to identify a dose does not influence the likelihood of a refill should dependent relationship between original also free surgeons to prescribe conservatively prescription and subsequent opioid refills among without undue concern for the burden of refill opiate naïve patients undergoing primary TSA. requests.21,35
Our findings are similar to prior literature across Surgeons performing TSA should strive to a range of surgical procedures. Scully et al prescribe the minimum quantity of opiates investigated opiate refill rates across 13 common required to achieve appropriate pain control.
P a g e | 158
Clear guidelines have not been established, prolonged opioid use in patients undergoing soft however, multimodal pain control has been tissue procedures of the hand, upper extremity, shown to effectively reduce opiate demand after and shoulder22-24,37, but few have focused on
TSA.15 Martusiewicz et al prospectively analyzed TSA. Chatha el al reported that TSA patients on the opiate consumption of 50 patients after TSA preoperative narcotics were 3.5 times more likely and found that with the use of an interscalene to develop postoperative dependence.17 Similarly, block catheter for 3 days, the majority of the in cohort of 12,038 TSA patients Khazi et al patient’s consumed <25 5mg oxycodone tablets. identified chronic preoperative opioid use as the
The same patients additionally reported strongest risk factor for ongoing use at 12 months satisfaction with their postoperative pain postoperatively, with chronic lung disease, age control.16 Interestingly, the average initial <65, chronic pain disorders, psychiatric prescription for the 17,706 patients in our study diagnoses, and EtOH abuse also reaching was greater than 90 pills. This snapshot of statistical significance.18 However, given that national prescribing practices suggests that these prior studies included patients on current practice is likely excessive, placing preoperative opioids, it is difficult to determine patients at risk for prolonged use and contributing risk factors for those who are preoperatively a large number of pills at risk for diversion to opioid naïve. To our knowledge, this is the first society.36 study to isolate analysis to opiate naïve TSA
Understanding patient specific risk factors patients. This is important given iatrogenic that predispose to opioid misuse after surgery is opioid dependence is of significant concern. Our critical to inform safe and responsible data suggests that in opioid naïve patients, prescribing. Certainly, this has become a central younger patients, females and patients with active focus of recent orthopedic literature. Prior tobacco use, coronary artery disease, diabetes, studies have investigated risk factors for rheumatic disease, or depression are at higher risk
P a g e | 159 prolonged opioid use after TSA. Therefore, at employee sponsored medical insurance or those risk patients with modifiable comorbidities, such with Medicare supplemental insurance. as smoking, may benefit from preoperative Uninsured patients, those with other private optimization. Additionally, surgeons should alert insurance plans not included by the Marketscan at risk patients of their elevated risk and counsel database, those with Medicare advantage, and them extensively on the importance of expedient those with Medicaid would not be included in this opioid cessation. We would also recommend that analysis, potentially limiting the generalizability in high risk patients, surgeons should tailor a pain to these specific patient cohorts. management strategy to minimize risk of patient Despite these limitations, the Truven harm. Given our results this should include a Marketscan database represents a strength of the reduction in the number of pills initially current investigation. The database allows for prescribed. analysis of a large number of patients and unlike
While the results of this investigation are many administrative national databases (like the informative, there are multiple limitations to this National Inpatient Sample) includes information investigation and the results of this study should from both inpatient and outpatient encounters. be interpreted with these in mind. First, as with Additionally, the database allows for longitudinal the analysis of any large database, we are reliant follow-up as long as the patient remains enrolled on accurate ICD and CPT coding within patient in their healthcare plan. Last, the database tracks records. Second, we used NDC codes to identify administered prescriptions. Therefore, we were preoperative opioid prescriptions. While there is able to track preoperative opioid prescriptions for significant precedence in the literature for this,38- the 6-month preoperative period and through a
41 we are again reliant on accurate coding. Last, year postoperatively. This is much more than the the Truven Marketscan database contains 30-90 days allowed by most other databases.42 information only on patients with private,
P a g e | 160
Conclusions
In conclusion, among opiate naïve patients undergoing TSA, we found that young age, female sex, and multiple preoperative comorbidities as being associated with prolonged postoperative opioid use. We recommend that surgeons utilize this information to counsel patients on their preoperative risk and consider tailoring a patient-specific pain management strategy for at risk patients. Furthermore, we identified a dose dependent relationship between initial opiate prescription and risk of subsequent refill and prolonged use.
Therefore, while perhaps counterintuitive, we recommend that providers limit the size of initial opioid prescriptions in order to lower the rate of need for patient refill and prolonged opioid use. Further study is needed to determine the optimal quantity of narcotic pills to control pain after TSA, while minimizing risk of dependence.
P a g e | 161
Table 2Table3. Adjusted 1. Demographics Odds of Risk and Factors Comorbidities for Post -ofOperative Post-Operative Opioid OpioidUse1 Use Groups PostPost-Operative-Operative Opioid Opioid Use Use Groups P-Value No Refill GroupsRefill Prolonged Use Refill Prolonged Use Total, n (%) 10,882 (61.5) 4,473 (25.3) 2,351 (13.3) Risk FactorAve Age, (SD) Odds Ratio67.98 (9.79) P-Value66.38 (9.83)Odds Ratio65.92 (9.95)P -Value<0.001 Age GroupAge Group 1 18-54 18-54 [Reference]925 (8.5) - 418 (9.3)[Reference] 268 (11.4) - <0.001 55-64 55-64 1.06 (0.933,245-1.21) (29.8) 0.3811,602 (35.8)0.84 (0.72-0.99)863 (36.7) 0.036 65-74 65-74 0.84 (0.733,573-0.96) (32.8) 0.0091,442 (32.2)0.58 (0.49-0.68)712 (30.3) <0.001 75+ 75+ 0.65 (0.563,139-0.75) (28.8) <0.0011,011 (22.6)0.46 (0.39-0.55)508 (21.6) <0.001 Sex Sex1 Male Male [Reference]6,125 (56.3) - 2,375 (53.1)[Reference] 1,182 (50.3) - <0.001 Female 1.20 (1.12-1.29) <0.001 1.37 (1.25-1.51) <0.001 Female 4,757 (43.7) 2,098 (46.9) 1,169 (49.7) Comorbidities1 Obesity 1,322 (12.1) 583 (13.0) 347 (14.8) 0.002 Renal Disease 554 (5.1) 250 (5.6) 128 (5.4) 0.416 Alcohol Abuse 103 (0.9) 33 (0.7) 28 (1.2) 0.167 Tobacco Use 395 (3.6) 192 (4.3) 135 (5.7) <0.001 Hypertension 7,117 (65.4) 2,924 (65.4) 1,591 (67.7) 0.095 Hyperlipidemia 5,975 (54.9) 2,473 (55.3) 1,321 (56.2) 0.518 Coronary Artery 2,112 (19.4) 874 (19.5) 527 (22.4) 0.003 Disease Congestive Heart 434 (4.0) 191 (4.3) 114 (4.8) 0.156 Failure Diabetes 2,127 (19.5) 919 (20.5) 585 (24.9) <0.001 Rheumatic Disease 508 (4.7) 200 (4.5) 154 (6.6) <0.001 Depression 850 (7.8) 470 (10.5) 317 (13.5) <0.001 Anxiety 326 (3.0) 170 (3.8) 104 (4.4) 0.001 1Presented as n (% of opioid use group) (95% confidence interval)
Table 2. Comparison of Original Opioid Prescription Data Post-Operative Opioid Use Groups No Refill Refill Prolonged Use Quantity 89.49 (87.3-91.6) 94.98 (92.1-97.9) 102.34 (98.6-106.1) Prescribed1 Average Difference 0 5.5 (1.7-9.3) 12.9 (7.9-17.8) P-Value2 - 0.005 <0.001 Total OME1 741.25 (722.9-759.5) 796.08 (769.9-822.2) 891.18 (855.3-927.0) Average Difference 0 54.8 (21.8-87.9) 149.9 (107.2-192.7) P-Value2 - 0.001 <0.001 1Presented as average (95% confidence interval); 2P-value when compared to no refill group; OME: Oral Morphine Equivalents
P a g e | 162
Comorbidities Obesity 0.98 (0.88-1.10) 0.777 1.05 (0.92-1.19) 0.517 Renal Disease 1.16 (0.99-1.36) 0.063 1.03 (0.84-1.26) 0.802 Alcohol Abuse 0.68 (0.46-1.02) 0.061 1.02 (0.66-1.57) 0.933 Tobacco Use 1.16 (0.97-1.38) 0.105 1.51 (1.23-1.85) <0.001 Hypertension 1.03 (0.95-1.11) 0.531 1.10 (0.99-1.22) 0.080 Hyperlipidemia 0.99 (0.92-1.07) 0.849 0.97 (0.88-1.07) 0.514 Coronary Artery 1.10 (1.00-1.21) 0.042 1.36 (1.21-1.53) <0.001 Disease Congestive Heart 1.08 (0.90-1.30) 0.391 1.09 (0.88-1.37) 0.434 Failure Diabetes 1.07 (0.98-1.17) 0.132 1.35 (1.21-1.51) <0.001 Rheumatic Disease 0.91 (0.76-1.07) 0.248 1.34 (1.11-1.62) 0.003 Depression 1.29 (1.14-1.45) <0.001 1.59 (1.38-1.84) <0.001 Anxiety 1.10 (0.91-1.33) 0.343 1.09 (0.87-1.38) 0.454 1Compaired to patients with no post-operative refill; Significant values highlighted in bold.
P a g e | 163
18 month study period
Pre-operative period Surgical period Refill period Prolonged use period (6 months to 6 weeks pre- (6w pre-op to (2 weeks to 6 months post- (6 to 12 months post-op) 2w post-op) op) op) Figure 1. Timeline of defined study periods represented to scale.
Figure 2. Study cohort selection of patients undergoing total shoulder arthroplasty (TSA) using Truven MarketScan database, 2010-2017
References
P a g e | 164
1. Seth P, Scholl L, Rudd RA, Bacon S. Overdose Deaths Involving Opioids, Cocaine, and Psychostimulants - United States, 2015-2016. MMWR Morb Mortal Wkly Rep. 2018;67(12):349-358. 2. Vowles KE, McEntee ML, Julnes PS, Frohe T, Ney JP, van der Goes DN. Rates of opioid misuse, abuse, and addiction in chronic pain: a systematic review and data synthesis. Pain. 2015;156(4):569-576. 3. Singh GK, Kim IE, Girmay M, et al. Opioid Epidemic in the United States: Empirical Trends, and A Literature Review of Social Determinants and Epidemiological, Pain Management, and Treatment Patterns. Int J MCH AIDS. 2019;8(2):89-100. 4. Morris BJ, Mir HR. The opioid epidemic: impact on orthopaedic surgery. J Am Acad Orthop Surg. 2015;23(5):267-271. 5. Sabatino MJ, Kunkel ST, Ramkumar DB, Keeney BJ, Jevsevar DS. Excess Opioid Medication and Variation in Prescribing Patterns Following Common Orthopaedic Procedures. J Bone Joint Surg Am. 2018;100(3):180-188. 6. Scully RE, Schoenfeld AJ, Jiang W, et al. Defining Optimal Length of Opioid Pain Medication Prescription After Common Surgical Procedures. JAMA Surg. 2018;153(1):37-43. 7. Sullivan MD, Edlund MJ, Fan MY, Devries A, Brennan Braden J, Martin BC. Trends in use of opioids for non-cancer pain conditions 2000-2005 in commercial and Medicaid insurance plans: the TROUP study. Pain. 2008;138(2):440-449. 8. Volkow ND, McLellan TA, Cotto JH, Karithanom M, Weiss SR. Characteristics of opioid prescriptions in 2009. Jama. 2011;305(13):1299-1301. 9. Edlund MJ, Martin BC, Russo JE, DeVries A, Braden JB, Sullivan MD. The role of opioid prescription in incident opioid abuse and dependence among individuals with chronic noncancer pain: the role of opioid prescription. Clin J Pain. 2014;30(7):557- 564. 10. Brat GA, Agniel D, Beam A, et al. Postsurgical prescriptions for opioid naive patients and association with overdose and misuse: retrospective cohort study. BMJ. 2018;360:j5790. 11. Kim SH, Wise BL, Zhang Y, Szabo RM. Increasing incidence of shoulder arthroplasty in the United States. J Bone Jt Surg Ser A. 2011;93(24):2249-2254. 12. Issa K, Pierce CM, Pierce TP, et al. Total Shoulder Arthroplasty Demographics, Incidence, and Complications-A Nationwide Inpatient Sample Database Study. Surg Technol Int. 2016;29:240-246.
P a g e | 165
13. Wagner ER FK, Wilson JM, Higgins I, Daly C, Gottschalk MB. The Incidence of Shoulder Arthroplasty: Rise and Future Projections Compared to Hip and Knee Arthroplasty. Journal of shoulder and elbow surgery. 2020. 14. Clarke H, Bonin RP, Orser BA, Englesakis M, Wijeysundera DN, Katz J. The prevention of chronic postsurgical pain using gabapentin and pregabalin: A combined systematic review and meta-analysis. Anesthesia and Analgesia. 2012;115(2):428- 442. 15. McLaughlin DC, Cheah JW, Aleshi P, Zhang AL, Ma CB, Feeley BT. Multimodal analgesia decreases opioid consumption after shoulder arthroplasty: a prospective cohort study. J Shoulder Elbow Surg. 2018;27(4):686-691. 16. Martusiewicz A, Khan AZ, Chamberlain AM, Keener JD, Aleem AW. Outpatient narcotic consumption following total shoulder arthroplasty. JSES Int. 2020;4(1):100- 104. 17. Chatha K, Borroto W, Goss L, Ghisa C, Gilot G, Sabesan VJ. How orthopedic surgeons can impact opioid use and dependence in shoulder arthroplasty. JSES Int. 2020;4(1):105-108. 18. Khazi ZM, Lu Y, Patel BH, Cancienne JM, Werner B, Forsythe B. Risk factors for opioid use after total shoulder arthroplasty. J Shoulder Elbow Surg. 2020;29(2):235- 243. 19. Morris BJ, Zumsteg JW, Archer KR, Cash B, Mir HR. Narcotic Use and Postoperative Doctor Shopping in the Orthopaedic Trauma Population. J Bone Joint Surg Am. 2014;96(15):1257-1262. 20. Wise KL, McCreary DL, Parikh HR, et al. Factors Associated with a Second Opioid Prescription Fill in Total Knee Arthroplasty. J Arthroplasty. 2020;35(6S):S163-S167. 21. Sekhri S, Arora NS, Cottrell H, et al. Probability of Opioid Prescription Refilling After Surgery: Does Initial Prescription Dose Matter? Ann Surg. 2018;268(2):271- 276. 22. Johnson SP, Chung KC, Zhong L, et al. Risk of Prolonged Opioid Use Among Opioid-Naïve Patients Following Common Hand Surgery Procedures. J Hand Surg Am. 2016;41(10):947-957.e943. 23. Gil JA, Gunaseelan V, DeFroda SF, Brummett CM, Bedi A, Waljee JF. Risk of Prolonged Opioid Use Among Opioid-Naive Patients After Common Shoulder Arthroscopy Procedures. Am J Sports Med. 2019;47(5):1043-1050.
P a g e | 166
24. Leroux TS, Saltzman BM, Sumner SA, et al. Elective Shoulder Surgery in the Opioid Naive: Rates of and Risk Factors for Long-term Postoperative Opioid Use. Am J Sports Med. 2019;47(5):1051-1056. 25. Blevins Peratikos M, Weeks HL, Pisansky AJB, Yong RJ, Stringer EA. Effect of Preoperative Opioid Use on Adverse Outcomes, Medical Spending, and Persistent Opioid Use Following Elective Total Joint Arthroplasty in the United States: A Large Retrospective Cohort Study of Administrative Claims Data. Pain Med. 2020;21(3):521-531. 26. Wilson JM FK EG, Bradbury TL, Guild GN. Preoperative Opioid Use is a Risk Factor for Complication Following Revision Total Hip Arthroplasty. Hip Int. 2020. 27. Services CfMM. Opioid Oral Morphine Milligram Equivalent (MME) Conversion Factors. https://www.cms.gov/Medicare/Prescription-Drug- Coverage/PrescriptionDrugCovContra/Downloads/Opioid-Morphine-EQ-Conversion- Factors-Aug-2017.pdf. Published 2017. Accessed2020. 28. Boudreau D, Von Korff M, Rutter CM, et al. Trends in long-term opioid therapy for chronic non-cancer pain. Pharmacoepidemiol Drug Saf. 2009;18(12):1166-1175. 29. Norris TR, Iannotti JP. Functional outcome after shoulder arthroplasty for primary osteoarthritis: a multicenter study. J Shoulder Elbow Surg. 2002;11(2):130-135. 30. Wilson JM, Farley KX, Aizpuru M, Wagner ER, Bradbury TL, Guild GN. The Impact of Preoperative Opioid Use Disorder on Complications and Costs following Primary Total Hip and Knee Arthroplasty. Adv Orthop. 2019;2019:9319480. 31. Namba RS, Paxton EW, Inacio MC. Opioid Prescribers to Total Joint Arthroplasty Patients Before and After Surgery: The Majority Are Not Orthopedists. J Arthroplasty. 2018;33(10):3118-3124 e3113. 32. Stanek JJ, Renslow MA, Kalliainen LK. The effect of an educational program on opioid prescription patterns in hand surgery: a quality improvement program. J Hand Surg Am. 2015;40(2):341-346. 33. Kim N, Matzon JL, Abboudi J, et al. A Prospective Evaluation of Opioid Utilization After Upper-Extremity Surgical Procedures: Identifying Consumption Patterns and Determining Prescribing Guidelines. J Bone Joint Surg Am. 2016;98(20):e89. 34. Farley KX, Anastasio AT, Kumar A, Premkumar A, Gottschalk MB, Xerogeanes J. Association Between Quantity of Opioids Prescribed After Surgery or Preoperative Opioid Use Education With Opioid Consumption. Jama. 2019;321(24):2465-2467.
P a g e | 167
35. Choo KJ, Grace TR, Khanna K, Barry J, Hansen EN. A Goal-directed Quality Improvement Initiative to Reduce Opioid Prescriptions After Orthopaedic Procedures. J Am Acad Orthop Surg Glob Res Rev. 2019;3(9):e109. 36. Bicket MC, Long JJ, Pronovost PJ, Alexander GC, Wu CL. Prescription Opioid Analgesics Commonly Unused After Surgery: A Systematic Review. JAMA Surg. 2017;152(11):1066-1071. 37. Nicholson AD, Kassam HF, Steele JL, Passarelli NR, Blaine TA, Kovacevic D. Development of a clinical risk calculator for prolonged opioid use after shoulder surgery. J Shoulder Elbow Surg. 2019;28(11):2225-2231. 38. Cozowicz C, Olson A, Poeran J, et al. Opioid prescription levels and postoperative outcomes in orthopedic surgery. Pain. 2017;158(12):2422-2430. 39. Weick J, Bawa H, Dirschl DR, Luu HH. Preoperative Opioid Use Is Associated with Higher Readmission and Revision Rates in Total Knee and Total Hip Arthroplasty. J Bone Joint Surg Am. 2018;100(14):1171-1176. 40. Smith SR, Bido J, Collins JE, Yang H, Katz JN, Losina E. Impact of Preoperative Opioid Use on Total Knee Arthroplasty Outcomes. J Bone Joint Surg Am. 2017;99(10):803-808. 41. Waljee JF, Cron DC, Steiger RM, Zhong L, Englesbe MJ, Brummett CM. Effect of Preoperative Opioid Exposure on Healthcare Utilization and Expenditures Following Elective Abdominal Surgery. Ann Surg. 2017;265(4):715-721. 42. Bohl DD, Singh K, Grauer JN. Nationwide Databases in Orthopaedic Surgery Research. J Am Acad Orthop Surg. 2016;24(10):673-682.
P a g e | 168
Fully Threaded Sacroiliac Lag Screws have Higher Load to Failure when Compared to Partially Threaded Screws: A Biomechanical Study
Huai Ming Phen, MBBS; Brent Wise, MD; Daniel Thompson, MS; Jason Nascone, MD; Adam Boissonneault, MBChB; Michael Maceroli, MD
P a g e | 169
Abstract: displacement curves, elastic, and plastic
deformation of each construct was recorded. The purpose of this study is to compare Fully threaded constructs had a 428% higher biomechanical properties of fully and yield force, 61% higher stiffness, 125% partially threaded iliosacral screws. We higher ultimate force, and 66% lower yield hypothesise that fully threaded screws will deformation (p<0.05). The average plastic have a higher yield force, and less deformation for partially threaded constructs deformation than partially threaded screws was 336% higher than fully threaded following axial loading. Twenty sawbone constructs (p=0.071), the final elastic blocks were uniformly divided to simulate deflection was 10% higher (p=0.248), and vertical sacral fractures. Ten blocks were the average total movement was 21% higher affixed with fully threaded iliosacral screws (p=0.107). We conclude from this in an over-drilled, lag-by-technique fashion biomechanical study that fully threaded, lag- whilst the remaining ten were fixed with by-technique iliosacral screws can withstand partially threaded lag-by-design screws. All significantly higher axial loads to failure screws measured 7.3-mm x 145mm, and than partially threaded screws. In addition, were inserted to a 70% of calculated fully threaded screws trended towards maximal insertional torque, ensuring exhibiting a significantly lower plastic uniform screw placement throughout across deformation following cyclical loading. models. Continuous axial loads were applied to 3 constructs of each type to Introduction failure to determine baseline characteristics. Pelvic ring fractures account for 3% of all Five hundred loading cycles of 500N at 1Hz skeletal fractures and can be associated with were applied to 4 constructs of each type, significant functional deficits if unstable and then axially loaded to failure. Force injuries progress to malunion.1 Surgical
P a g e | 170 fixation may minimize the risk of long term cannulated 7.3mm core diameter screw, pain and gait disturbance in unstable pelvic which is most commonly ring fractures.2-4 Percutaneous iliosacral (IS) utilised in the reduction and fixation of screw fixation, typically with a partially sacroiliac joint disruption. The purpose of threaded, cannulated screw, for stabilisation this biomechanical study is to compare the of the sacroiliac (SI) joint has been maximal torque, load to failure, and cyclical identified as a favourable method of loading properties of fully and partially posterior ring stabilisation and requires a threaded iliosacral screws in a sawbones minimally invasive approach.5 model. We hypothesise that a fully threaded
Although the utility of using a partially lag by technique 7.3mm screw will achieve threaded screw arises from only requiring a a higher insertion torque, higher load to single drill hole thereby eliminating the failure, and less displacement with cyclical need to over-drill through the cancellous loading when compared to a 7.3mm bone of the posterior superior iliac crest, partially threaded screw. partially threaded designs have been shown Methods: to have reduced initial screw stiffness and Ten 7.3mm x 145mm partially threaded yield load when compared to fully threaded screws (32mm threads), and ten fully screws when used for fixation of ankle threaded 7.3mm x 145mm screws (model fractures.6 Furthermore, the incidence of IS 209.745, 209.945; DePuy Synthes, Paoli screw loosening has been observed as at a PA) were obtained for analysis. 20 rate as high as 17.3%, with up to 11.8% of sawbones constructs were created to patients requiring revision surgery.7 simulate the sacroiliac joint. The blocks No studies to date have investigated or (model SKU:1522-02; Pacific Research compared the biomechanical properties of a Laboratories, WA) which were originally fully threaded against a partially threaded 130x180x40mm in dimension were cut into
P a g e | 171 two pieces, of 25 mm and 155 mm, for removed prior to biomechanical testing. testing. A 15pcf density was utilised in (Figure. 1) concordance with the American Society for Biomechanical testing was performed using Testing and Materials International standard an MTS 858 Mini Bionix (MTS Systems for biomechanical testing of osseous screws Corporation, Minneapolis MN). in cancellous bone. As our sawbones did not Torque simulate a cortical bone surface, we created a washer using 3/16 inch sheet aluminium to An insertion torque of approximately 70% prevent the screws from subsiding into the of maximum is recommended to prevent
8,9 sawbones during insertion. screws from stripping during insertion. In order to provide a standardised insertion Screws were placed by a single surgeon to torque for axial load testing, we performed a limit variation in technique. A 2.8mm maximal insertion torque (MIT) to failure threaded guidewire was inserted into the analysis using one screw of each type. The sawbones blocks in perpendicular MTS was utilised to insert the screw whilst orientation to the simulated SI joint. A torque readings (N-m) were continuously 5.0mm pilot hole was drilled across the SI obtained. MIT was defined as the maximum joint for both fully threaded and partially torque reading obtained prior to failure, threaded screws. An overdrilled glidehole which was defined as the highest achievable of 7.3mm was created for the fully threaded torque prior to a reduction in torque screw samples up to the level of the SI joint. readings, which was assumed to be The screw was inserted using a cannulated secondary to the screw stripping. Following hexagonal screwdriver over the guide wire, obtention of this value, all subsequent along with our manufactured and standard constructs were inserted at 70% of the MIT washers. The guidewire was subsequently for the designated screw.
P a g e | 172
To simulate an environment similar to that Failure was defined as a maximum in the operating room, we proceeded to displacement of 3.0cm, as this was felt to be insert the same screws manually to these the largest displacement that would be designated values. The construct was observed prior to the construct physically clamped to the baseplate of the MTS and the breaking, and would take into account the screw was inserted manually with a regular compression of the testing material itself. cannulated screw driver, which allowed us Force displacement graphs were obtained to obtain continuous torque readings. We for each tested construct. Stiffness (N/mm) found that it was impossible to insert the was calculated by calculating the gradient of fully threaded screw to the insertional the initial linear region of each plot. The torque previously designated by the MTS, yield force (N) was obtained by identifying as the magnitude of torque was too high. As the force value at the intersection of the plot a result, we opted to utilise the largest, most and a 0.05 mm parallel offset line to the reproducible torque value of 6.58N-m for initial linear region. Ultimate force (N) was the fully threaded screws. defined as the maximum force obtained at
Load to Failure any point during testing, prior to failure.
Yield deflection (mm) was defined as the To determine baseline characteristics of total deformation of the screw, at the point each of the screw types, an axial load to prior to yielding. failure was performed for 3 screws of each type. Constructs were randomised to a Cyclical Loading consecutive order for testing to minimise Cyclical load testing was performed for four consistency bias using MATLAB® screws of each type. Again, constructs were (MathWorks). An axial shearing load was randomised to a consecutive order for applied perpendicular to the construct at a testing. A 500N axial load was applied for rate of 0.25mm/s, until failure was achieved. 500 cycles at a frequency of 1Hz.10,11
P a g e | 173
Hysteresis curves of force against Maximal Insertional Torque displacement were obtained. Plastic Utilising the MTS, the fully threaded screw deformation, i.e. the relative deformation of achieved an MIT of 11.22N-m, compared to the construct after which it did not return to 2.15N-m for the partially threaded screw, its original, pre-loaded form, was calculated which in turn returned insertional torque as the difference between displacements values of 7.85N-m for fully threaded observed between the initial and final cycles screws, and 1.50N-m for partially threaded of loading. The final elastic deflection of screws (Figure. 2). Again, we were unable each screw was defined as the displacement to manually reproduce an insertional torque observed within the final cycles of loading, of 7.85N-m for the fully threaded screws essentially representing a construct that had and therefore 6.58N-m was used. been deformed by prior loading cycles, but Pre-cyclical loading: was now consistently returning to its original form. The total displacement of In axial compression, the fully threaded each construct was defined as the difference construct had significantly higher yield in displacement from neutral, i.e. 0, and the force (1727.60N vs 326.96N, p=0.0092), last cycle. (Figure. 3) Each construct was stiffness (761.55N/mm vs 473.77N/mm, again axially loaded to failure after cyclical p=0.0061), ultimate force (3489.92N vs loading. 1549.53N, p=0.019), and yield deformation
(2.50mm vs 0.86mm, p=0.025) when Student’s 2 tailed t-test was performed in compared to partially threaded constructs. MATLAB® using the “ttest” function to (Figure. 4) detect differences between the properties of the two types of screw, at a significance Post-cyclical loading: level of 0.05. During cyclical loading of the fully threaded
Results: construct, the average plastic deformation
P a g e | 174 was 0.041mm ±0.005. The average plastic force of the partially threaded screws deformation observed for partially threaded increased significantly. construct was 336% higher at 0.136mm Discussion ±0.087; however, this difference did not It is accepted that early operative treatment reach statistical significance (p=0.070). The of a displaced posterior pelvic ring allows final elastic deflection between the two for early rehabilitation and reduces types of construct was not statistically morbidity.12 The ideal technique and significant (0.84mm vs 1.013mm, p=0.25). construct for sacro-iliac or transiliac- The mean total movement of partially transsacral screw fixation remains unknown threaded constructs was 21% higher due to the variety of posterior pelvic (1.013mm, ±0.16 vs 0.84mm, ±0.097, fracture patterns and significant forces p=0.25). (Figure. 5) transmitted through the SI joints.7,13,14 The Following cyclical loading, fully threaded current literature regarding fully and constructs had a significantly higher average partially threaded SI screws investigates the yield force (1923.05N vs 608.21N, incidence of iatrogenic nerve injury. There p=0.00011), ultimate force (4038.20N vs is no indication that a compression screw 1658.48N, p=0.00037), and yield puts the neurovascular structures at risk.15,16 deformation (2.65mm vs 1.07mm, However, late displacement or loss of p=0.0016) (Table. 1). The stiffness of fully fixation can have significant functional threaded constructs and partially threaded consequences.1,4,17 No studies to date have constructs was not significantly different compared the biomechanical properties of a following cyclical loading (802.57N/mm vs fully threaded lag-by-technique posterior 787.49N/mm, p=0.3). When comparing pelvis screw against a lag-by-design constructs that had or had not been partially threaded screw. cyclically loaded, the stiffness and yield
P a g e | 175
Screw loosening typically occurs early in compared to those sustaining VS fractures the rehabilitation period following treatment with zone-1 sacral fractures, or simple of unstable pelvic ring injuries, as early sacroiliac joint dislocation (43.5% vs 7.1%, mobilisation after surgery is encouraged if p=0.019). These findings were similar to no other injuries preclude weight bearing. those observed by Griffin et al. in their
Kim et al. in 2016 found a 17.3% incidence retrospective review of 62 patients of sacroiliac screw loosening at a mean of sustaining VS type fractures, treated with a
25.3 days (range, 10-70 days), and revision partially threaded iliosacral screw, who rate of 11.8% in their retrospective review found that a vertical sacral fracture pattern of 110 pelvic ring injuries that received permeated an excessive risk of failure of percutaneous fixation with a 7.0mm 13%.14 Within their cohort, all of the partially threaded design.7 The incidence of failures occurred within 3 weeks of the loosening was found to be significantly initial surgery. The morphology of the higher in vertical shear (VS) type fractures injury as well as surrounding bone stock
(29.7%), when compared to anterior- also contributes towards construct failure, posterior compression (11.1%), or lateral- thus underscoring the need to maximize compression (10.9%) type fractures fixation strength and stability.
(p=0.014). Screw fixation within the middle Anecdotally, the senior author (MM) has 1/3 of the S1 body had a higher screw noted subjective increase in screw purchase loosening when compared to screws and insertional torque when placing fully abutting the anterior cortex of the S1 body threaded screws in vivo. We found a higher (23.4% vs 8.7%, p=0.044). Patients with VS maximal insertion torque prior to failure that fractures and concomitant zone-2 sacral was more than five times greater for the fractures were also significantly more likely fully threaded lag screw when compared to to experience screw loosening when the partially threaded screw. The two types
P a g e | 176 of stainless steel screws compared in this osteoporotic groups, found that the pullout study have the identical thread and core strength for both normal and osteoporotic diameters, 7.3mm and 5.0mm respectively, bone was greatest at 50% of the MIT, and with a thread pitch of 2.75mm. The not statistically significant from pullout torsional strength of a screw, i.e. how much strengths at 70-90% of the MIT. For torque it can withstand prior to plastic osteopenic bone, the pullout strength was deformation, is proportional to the cube of greatest at 70% of MIT but the absolute the core diameter as well as the material value itself was not significantly different from which it is made.18 from the pullout strengths at 50% and 90%
of MIT.20 The aforementioned studies did Interestingly, the insertional torque of a not introduce any micromotion within the screw has not been correlated to the pullout construct prior to testing the pullout strength. Ricci et al. investigated the effect strength, which has been shown to inhibit of insertional torque on the pullout strength the formation of bone, instead becoming of cortical screws with varying pitches enveloped in a fibrous tissue which is also (range 1.0mm – 1.75m) within a cancellous evident radiographically by a visible bone surrogate. The MIT obtained for larger “halo”.21 pitch screws was significantly higher for those obtained with screws with a pitch of The utility of the screws used for sacroiliac
1.2mm or less, however, their findings fixation are to prevent axial displacement concluded that screw pitch, as well as the and rotation.13 The results of our axial screw’s ability to generate a higher initial loading tests are similar to those seen by insertional torque were not predictive of Downey et al, with the fully threaded screws pullout strength.19 Tankard et al. in a resisting much higher deforming forces cadaveric study of 15 pairs of humeri prior to failure when compared to the stratified to normal, osteopenic, or partially threaded screws.6 The authors
P a g e | 177 found no difference between the stiffness of holding strength of cancellous screws in the two screws. Our findings demonstrate biomechanical models, as these contribute that prior to any loading, the partially directly to the thread-bone interface, threaded screw exhibited a significantly through which failure occurs by shear. 25,26 lower stiffness, which increased following Given that the two screws are manufactured cyclical loading. We propose that this from the same material and the only phenomenon is due to the bone surrogate differing factor is the length of threads, we essentially settling around the screw, as the propose that the increased length of thread stiffness of the two screws following engagement led to the difference in forces cyclical loading was strikingly similar, seen, and that the initial micromotion upon which is to be expected given the identical initial loading may contribute to ultimate material properties of the screws. Despite screw failure in vivo. the differences in plastic deformation seen There were several limitations in the present between the two constructs, there was no study, consistent with biomechanical difference in the total displacement, which modelling. The torque results in our study we believe is due to the aforementioned were obtained with the utilisation of a reason, as the hysteresis curves for the final manufactured washer, as we found that cycles are visibly identical. during screw insertion the head and washer
Prior studies investigating screw loosening would subside into the testing material. under dynamic loading conditions have Furthermore, the MIT obtained for the fully utilised plate constructs, which is not threaded screw with use of this washer far applicable in the management of sacroiliac exceeded the torque able to be generated disruption.22-24 The length of thread manually. For this reason, statistical testing engagement, as well as the diameter of the was not performed for torque values as the screw have also been correlated with the results obtained cannot be extrapolated into
P a g e | 178 clinical practice. However, the same washer The present biomechanical study supported was utilised for all constructs, and although the hypothesis that fully threaded lag screws torques obtained may not be representative were able to withstand significantly higher of that obtained in vivo, they are still axial displacement loads and insertional consistent and reliable for the purpose of torque when compared to a partially this study. The orientation of the screw threaded screw in an SI joint model. In insertion may have differed between addition, fully threaded lag screws exhibited constructs, which may explain the large a trend toward smaller plastic deformation, standard deviation seen in the plastic i.e. irreversible deformation, when deformation of the screws. This compared to partially threaded screws, phenomenon could be reduced with an although this did not reach significance. infinitely larger sample size, far beyond Given the extreme forces placed across the what would be possible in vivo. sacroiliac joint with weight bearing, the
Furthermore, our testing setup did not allow ideal fixation construct minimizes us to identify whether or not the displacement, This biomechanical model displacement was occurring at the implant indicates that the increased thread or the compression of the testing material engagement length in fully threaded lag itself; however, our intention for this study screws may be an advantage over more was not to obtain absolute yield force and commonly used partially threaded screws. displacement characteristics, but rather These promising results underscore the need provide a quantified comparison of loading for well modelled cadaveric studies and characteristics between the two types of subsequent clinical trials to further elucidate screw within a consistent medium. the ideal screw construct in unstable pelvic
ring injuries. Conclusion:
P a g e | 179
Figure. 1
A B
P a g e | 180
Figure. 2
P a g e | 181
Figure 3
Total displacement
P a g e | 182
Figure. 4:
P a g e | 183
Figure. 5:
P a g e | 184
Figure Captions: Figure. 1: Photograph demonstrating methodology used to obtain maximum insertional torque readings. Clamp apparatus can be seen that was utilised to hold construct to base and prevent rotation on baseplate. Cannulated screwdriver power attachment was extended from MTS and utilised over guidewire to obtain screw specific torque insertion. Screwdriver power attachment was substituted with regular manual cannulated screwdriver when constructs were created B)
Photograph demonstrating apparatus and construct setup, with fabricated and manufactured washer prior to mechanical testing. A cage was constructed and utilised to prevent anterior and posterior tilting during testing, and retain the axial force vector applied by MTS.
Figure. 2: Graph demonstrating maximum torque readings obtained with screw rotation. The step-offs seen on each graph line were as a result of the MTS only being able to turn 270 degrees, requiring the system to be reset and thus dropping the observed torque to 0.
Figure. 3: Hysteresis curve demonstrating the first, and final ten cycles of axial loading from partially threaded screw number 2. The plastic deformation, final elastic deflection, and total displacement are calculated as above. The graph is representative of an axial load being applied, causing an initial deformation. The construct is then unloaded, resulting in an elastic return to its original shape. The plastic deformation can be interpreted as permanent construct deformation, i.e. the construct is no longer returning back to its original state as it has changed morphology.
Of note, the final 10 cycles can be seen to be overlapping each other directly, demonstrating that the construct has essentially “settled”, representing elastic changes during cyclical loading.
Figure. 4: Histograms with standard error bars demonstrating ultimate force (N), stiffness
(N/mm), yield force (N), and yield deflection (mm), for partially and fully threaded screws before and after cyclical loading.
P a g e | 185
References:
1. Grotz MRW, Allami MK, Harwood P, Pape HC, Krettek C, Giannoudis PV. Open pelvic fractures: Epidemiology, current concepts of management and outcome. Injury 2005;36:1-13.
2. Gänsslen A, Pohlemann T, Paul C, Lobenhoffer P, Tscherne H. Epidemiology of pelvic ring injuries. Injury 1996;27:13-20.
3. Henderson RC, Nepola JV, Claverie JG. Nonoperatively Treated Major Traumatic Pelvic
Disruptions: an Evaluation of the Long-term Results. The Iowa Orthopaedic Journal 1986;6:100.
4. McLaren AC, Rorabeck CH, Halpenny J. Long-term pain and disability in relation to residual deformity after displaced pelvic ring fractures. Canadian Journal of Surgery
1990;33:492-4.
5. Comstock C, van der Meulen M, Goodman S. Biomechanical Comparison of Posterior
Internal Fixation Techniques for Unstable Pelvic Fractures. Journal of Orthopaedic Trauma
1996;10:517-22.
6. Downey MW, Kosmopoulos V, Carpenter BB. Fully Threaded Versus Partially Threaded
Screws: Determining Shear in Cancellous Bone Fixation. Journal of Foot and Ankle Surgery
2015;54:1021-4.
7. Kim JW, Oh CW, Oh JK, et al. The incidence of and factors affecting iliosacral screw loosening in pelvic ring injury. Arch Orthop Trauma Surg 2016;136:921-7.
8. Dinah A, Mears S, Knight T, Soin S, Campbell J, Belkoff S. Inadvertent screw stripping during ankle fracture fixation in elderly bone. Geriatric Orthopaedic Surgery & Rehabilitation
2011;2.
9. Cleek T, Reynolds K, Hearn T. Effect of Screw Torque Level on Cortical Bone Pullout
Strength. J Orthop Trauma 2007;21.
P a g e | 186
10. Pel JJ, Spoor CW, Pool-Goudzwaard AL, Hoek van Dijke GA, Snijders CJ.
Biomechanical analysis of reducing sacroiliac joint shear load by optimization of pelvic muscle and ligament forces. Ann Biomed Eng 2008;36:415-24.
11. Wu T, Ren X, Cui Y, et al. Biomechanical study of three kinds of internal fixation for the treatment of sacroiliac joint disruption using biomechanical test and finite element analysis. J
Orthop Surg Res 2018;13:152.
12. Latenser B, Gentilello L, Tarver A, Thalgott J, Batdorf J. Improved Outcomes with Early
Fixation of Skeletally Unstable Pelvic Fractures. Journal of Trauma 1991;31.
13. Beaule PE, Antoniades J, Matta JM. Trans-sacral fixation for failed posterior fixation of the pelvic ring. Arch Orthop Trauma Surg 2006;126:49-52.
14. Griffin D, Starr A, Reinert C, Jones A, Whitlock S. Vertically Unstable Pelvic Fractures
Fixed with Percutaneous Iliosacral Screws: Does Posterior Injury Pattern Predict Fixation
Failure? Journal of Orthopaedic Trauma 2003;17.
15. Herman A, Keener E, Dubose C, Lowe JA. Zone 2 sacral fractures managed with partially-threaded screws result in low risk of neurologic injury. Injury 2016;47:1569-73.
16. Vaccaro A, Kim D, Brodke D, et al. Diagnosis and management of sacral spine fractures.
Instr Course Lect 2004.
17. Kellam JF. Long-term functional prognosis of posterior injuries in high-energy pelvic disruption. Journal of Orthopaedic Trauma1998:150-1.
18. Ansall R, Scales J. A Study of Some Factors Which Affect the Strength of Screws and their Insertion and Holding Power in Bone. J Biomechanics 1968;1.
19. Ricci W, Tornetta III P, Petteys T, et al. A Comparison of Screw Insertion Torque and
Pullout Strength. Journal of Orthopaedic Trauma 2010;24.
P a g e | 187
20. Tankard S, Mears S, Marsland D, Langdale E, Belkoff S. Does Maximum Torque Mean
Optimal Pullout Strength of Screws? Journal of Orthopaedic Trauma 2013;27.
21. Schatzker J, JG H, Sumner-Smith G. The Effect of Movement on the Holding Power of
Screws in Bone. Clinical Orthopaedics and Related Research 1975.
22. Graham A, Hess D, Stephens H. Screw Loosening in an In Vitro Mid Fibular Osteotomy
Model under Dynamic Loading Conditions. Foot & Ankle International 2000;21.
23. Gesensway D, Putnam M, Mente P, JL L. Design and biomechanics of a plate for the distal radius. Journal of Hand Surgery 1995;20.
24. Panagiotopoulos E, Fortis A, Millis Z, Lambiris E, Kostopoulos V, Vellios L. Pattern of screw loosening in fractures fixed with conventional and functional plates. Injury 1994;25.
25. Chapman J, Harrington R, Lee K, Anderson P, Tencer A, Kowalski D. Factors Affecting the Pullout Strength of Cancellous Bone Screws. Journal of Biomechanical Engineering
1996;118.
26. Hearn T, Schatzker J, Wolfson N. Extraction Strength of Cannulated Cancellous Bone
Screws. Journal of Orthopaedic Trauma 1993;7.
P a g e | 188
P a g e | 189
P a g e | 190
P a g e | 191
P a g e | 192
Orthopaedic Surgery & Research Faculty
Adult Reconstruction Steven Presciutti, Assistant Professor Thomas Bradbury, Associate Professor John Rhee, Professor Greg Erens, Assistant Professor S. Tim Yoon, Professor Oren Goltzer, Assistant Professor George Guild III, Assistant Professor Spine & Neurosurgery James Roberson, Robert P. Kelly Professor Matthew Gary, Assistant Professor Keerat Singh, Assistant Professor Daniel Refai, Associate Professor Gerald Rodts, Professor Foot & Ankle Jason Bariteau, Assistant Professor Sports Medicine Sam Labib, Associate Professor Kyle Hammond, Assistant Professor John Louis-Ugbo, Assistant Professor Ed Jackson II, Assistant Professor Spero Karas, Associate Professor Hand & Upper Extremity Greg Lee, Assistant Professor Charles Daly, Assistant Professor T. Scott Maughon, Assistant Professor Michael Gottschalk, Assistant Professor Mathew Pombo, Assistant Professor Clifton Meals, Assistant Professor Bryan Whitfield, Assistant Professor Diane Payne, Assistant Professor John Xerogeanes, Professor Eric Wagner, Assistant Professor Trauma Musculoskeletal Oncology Jared Allen, Assistant Professor David Monson, Assistant Professor Roberto Hernandez-Irizarry, Asst. Professor Shervin Oskouei, Assistant Professor Mike Maceroli, Assistant Professor Nickolas Reimer, Assistant Professor Tom Moore Jr., Assistant Professor Will Reisman, Assistant Professor Pediatric Orthopaedics Mara Schenker, Assistant Professor Richard Thomas, Assistant Professor Robert Bruce, Associate Professor
Nicholas Fletcher, Associate Professor Research Spine Hicham Drissi, PhD, Prof. & Vice-Chair Svenja Illien-Junger, PhD, Asst. Professor Dheera Ananthakrishnan, Asst. Professor Nick Willett, PhD, Assistant Professor Scott Boden, Professor & Chair
Eli Garrard, Assistant Professor John Heller, Baur Professor Keith Michael, Assistant Professor Andrew Milby, Assistant Professor