Outpatient orthopaedic trauma surgery: is it safe?

Supplemental Digital Content is Available in the Text.

Abstract

Background:

This study describes outcomes after outpatient operative fracture management concurrently with implementation of the 2019 Orthopaedic Trauma Association Guidelines for Musculoskeletal Pain. Primary and revision orthopaedic trauma cases were both completed, with standard comprehensive pain management regimen modeled on OTA/AO guidelines. For each procedure, the senior surgeon was assisted by the same anesthesia team. Regional anesthesia was determined by the surgeon and the anesthesiologist.

Methods:

This was a retrospective case series of 88 patients older than 18 years undergoing surgery for fractures and complications at an outpatient facility associated with a Level 1 trauma center.

Results:

The primary outcome measures were complications, emergency department visits, and reoperations. Secondary outcomes included office calls, pain control complaints, and medication refills. All data were captured by retrospective chart review. Of the 88 study patients, 52% were male and the median age was 41 years. Seventy-seven percent (68/88) received regional anesthesia at the time of surgery. A total of 7 patients (8%, 10 encounters) presented to the emergency department within 30 days. 60% of these encounters (6/10) were related to their injury. Four patients presented for pain, but none were admitted. Most (76%) of the office calls for pain were secondary to medication refill in accordance with the guidelines.

Conclusions:

Patients can be cared for within the outpatient setting for many complex orthopaedic trauma conditions. Appropriate patient selection, utilization of regional anesthesia in selected patients, and proper pain management are necessary to ensure safety and comfort.

Level of Evidence:

Level III.

1. Introduction

Over the past 4 decades in the United States, there has been increased utilization of the outpatient setting for surgical procedures.¹ Much of this momentum was believed to have been due to Medicare responding to escalating health care costs by allowing for coverage to extend to ambulatory surgery centers.² Orthopaedics is the third most frequent user of ambulatory surgery centers (ASCs) according to the Centers for Disease Control (CDC).

Certain orthopaedic subspecialties have quickly adapted to expanding outpatient procedures, particularly during and after the COVID-19 pandemic.^3,4 It is common for shoulder, knee, and hip arthroplasty to take place in the outpatient setting.^5–7 Recently, several spine surgeries have been shown to be safe in the outpatient setting.^8,9 One orthopaedic subspecialty that has been slower to convert to more outpatient procedures is trauma. The limited literature on outpatient procedures for orthopaedic trauma mostly includes distal radius fractures, pediatric fractures, and nonunion.^3,10–14

Until recently, there has not been significant literature published regarding outpatient management of orthopaedic trauma injuries. The literature that existed only evaluated isolated ankle or distal radius fractures.^15–17 In 2020, Wolfstadt et al¹⁸ published a series operationalizing outpatient fracture management with immense success (eg, 94% reduction in length of stay, high patient satisfaction, and no safety concerns), leading others to consider treating more fracture cases in the outpatient setting. Recently, additional research found more complex lower extremity fractures to be safe in the outpatient setting further bolstering outpatient utilization.¹⁹ The goal of this study was to evaluate a variety of reconstructive and acute traumatic cases that used operative intervention in the outpatient setting with implementation of the 2019 Orthopaedic Trauma Association (OTA/AO) Guidelines for Musculoskeletal Pain. The hypothesis of this study was that patients could be appropriately cared for in the outpatient setting and would not have significant issues with pain or unscheduled urgent and emergency center visits. In addition, this study planned to evaluate the effect that regional or local anesthesia would have on outcomes.

2. Methods

After obtaining institutional review board approval from Atrium Health Musculoskeletal Institute (05-19-10E; PI: Dr. Rachel B. Seymour), a level IV retrospective study of a prospectively implemented protocol at a single Level 1 trauma center was completed between November 2016 and June 2020. Any investigation involving human subjects or the use of patient data for research purposes was approved by the committee on research ethics at the institution in which the research was conducted in accordance with the Declaration of the World Medical Association and that any informed consent from human subjects was obtained as required. Each procedure was completed by 1 senior orthopaedic trauma faculty with the same anesthesia team. Patient selection for outpatient surgery was determined by the attending surgeon and defined criteria set forth by anesthesia to minimize risk. Appendix 1, http://links.lww.com/OTAI/A112 highlights exclusion criteria that the outpatient surgery center defined as too high risk for their facility. Patients underwent a variety of upper and lower extremity surgical procedures that included utilization of implants such as intramedullary fixation, plates, screws, wires, external fixators, and continuous compression devices. Both chronic and acute injuries were treated at the outpatient surgery center. Injuries treated included but were not limited to intra-articular distal humerus fractures, calcaneus fractures, subtalar fusions, intra-articular distal femur fractures, tibial plateau fractures, and pilon fractures in addition to other primary and revision surgeries. Hardware removal cases were excluded because those cases are routinely been performed on an outpatient basis. Deformity care cases were also excluded.

Throughout the entire perioperative period (preoperative, operating room, post-anesthesia care unit [PACU]), these patients followed the OTA/AO Guidelines for Musculoskeletal Pain.²⁰ It is important to report that while the OTA/AO guidelines were not published until 2019, this institution had been working on developing that protocol for years prior and formally presented at the Orthopaedic Trauma Association's Annual Meeting in 2017. This allowed the protocol to be implemented near the beginning of this study period. Most of the patients (75%) received regional or local anesthesia to assist with perioperative pain control. Intraoperative opioid medication was avoided. The ultimate decision whether to use regional anesthesia was dependent on the surgeon and the anesthesiologist. Regional anesthesia was offered to all patients with shared decision making with the anesthesiologist and the patient. Regional anesthesia was performed by an attending anesthesiologist preoperatively, whereas local anesthesia was administered by the operative team. Patients received 81 mg of aspirin twice daily for deep vein thrombosis (DVT) prophylaxis.²¹

Patients' postoperative pain regimens were selected to either major musculoskeletal injury procedure (operative fixation long bone, periarticular, complex injury, significant soft-tissue surgery) or minor musculoskeletal injury procedure (operative fixation small bone, simple fracture, small soft-tissue surgery) based on the recommendations set forth in the OTA/AO guidelines. Major surgery guidelines recommend hydrocodone/acetaminophen or oxycodone/acetaminophen q4hr pro re nata (PRN) for 1 week with taper over the next 4 weeks. In addition, patients should have received ibuprofen, gabapentin, and scheduled acetaminophen to provide multimodal pain control. Minor surgery guidelines include hydrocodone/acetaminophen q6hr PRN with 3 weeks' taper as well as ibuprofen, gabapentin, and scheduled acetaminophen. In the postoperative discussion with the patient and caregivers, the surgeon provided education about expecting pain once the regional anesthesia wears off and instructed patients to start using ice, transcutaneous electrical nerve stimulation (TENS), other nonpharmaceutical and cognitive strategies (ie, music, aromatherapy), and their prescribed medications as soon as they have return of any sensation to manage and prevent rebound pain. Patients were instructed to call back to the office at 1-week intervals for continuing prescribing of taper because of 1-week prescription duration limits set forth by the state of North Carolina. These calls are fielded by nursing staff in the clinic in the same manner as any other patient treated by this surgeon. There is not a different protocol or system for patients treated in outpatient facilities. The nurses initiated the pain medication prescriptions in accordance with the protocol in the OTA/AO guidelines, and an advanced practice provider (nurse practitioners or physician's assistants) or physician (resident, fellow, or attending) signed the prescription. Opioid prescriptions were monitored for 3 months to assess for chronicity of use.^22,23 All patients followed up as indicated by the surgeon.

The primary outcomes were complications, emergency department (ED) visits, and reoperations. All records available through the health care system's electronic health record were reviewed for 1 year postoperatively. ED and urgent care (UC) visits were further stratified to assess for visits related to the orthopaedic injury/surgery versus unrelated utilization. Patients were deemed to have been evaluated for orthopaedic issues if they had pain in injured extremity, concern for infection, or splint/cast concerns. Nonorthopaedic visits included but were not limited to unrelated medical concerns, mental health, or additional motor vehicle crashes. Patient charts were thoroughly reviewed for general communication with the orthopaedic trauma team. All patient calls to the office were logged. The content of these calls was evaluated by the primary author to determine the reason for the call (ie, pain, concern for infection, or other [including general wound care, follow-up, or clerical work]). Calls for pain were subdivided further into calls for medication refill (to follow state law) and issues with significant pain to clarify patients who were following the scheduled taper or were having significant breakthrough pain.

Descriptive statistics were used to characterize our patient population regarding demographics, original mechanism of injury, OTA/AO fracture classification, and operative characteristics, using SAS software version 9.4 (SAS Institute Inc., Cary, NC).

3. Results

Eighty-eight patients underwent orthopaedic procedures at an outpatient facility during the study period (Table 1). Of these, 83 (94.3%) were discharged on the same calendar day as the procedure. One patient was admitted after surgery because of lack of transportation, 1 patient was admitted for poor mobilization in PACU, 2 patients were observed because of patient preference, and 1 patient was observed because of unclear reasons. Fifty-two percent (n = 46) of the patients were male. Most of the patients had surgery for acute fracture fixation (55.7%). An additional 36% of patients had surgery for a fracture-related complication (ie, nonunion, infection, or wound complications). The remaining patients had surgery for soft-tissue management, bullet removal, heterotopic ossification removal, or I&D. All cases with grafting were performed with bone graft substitute.

Table 1.

Baseline characteristics (N = 88).

Age: Median (IQR)	40.5 (31, 52)
Male	46 (52.27%)
Patients with ED visits	27 (30.68%)
EBL: Median (IQR)	75 (25,150)
Surgery type and indication
Acute fracture fixation	49 (55.7%)
Internal fixation, n=40
Plate, n=26
Screw/wire/pin, n=8
IMN, n=4
IMN and plate, n=2
Fusion, n=8
Internal fixation and allograft, n=1
Other acute fracture management	3 (3.4%)
Wound I&D and flap/skin graft, n=2
Ballistic fracture—bullet removal, n=1
Treatment of fracture complication	32 (36.4%)
Nonunion, n=23
Internal fixation, n=17
Plate, n=12
IMN, n=1
IMN and screw, n=2
Screw, n=1
TTC nail, n=1
Internal fixation with grafting, n=5
IMN and allograft, n=1
Plate and allograft, n=3
Plate and massive bone grafting, n=1*
Internal fixation with antibiotic spacer, n=1
Deep SSI, n=5
I&D, n=2
Internal fixation (IMN), n=2
External fixation, n=1
Infection and nonunion, n=3
I&D and tissue transfer, n=1
Bone graft, n=2*
Open wound, n=1
Skin graft, n=1
Other	4 (4.5%)
Sequelae of necrotizing fasciitis, skin grafting, n=1
Osteomyelitis at bone graft donor site, I&D, n=1
Hematogenous osteomyelitis, I&D, and internal fixation (IMN and screw), n=1
Removal heterotopic ossification, n=1
Total	88
OTA/AO class
Upper extremity	26 (29.55%)
Proximal humerus (11)	3 (3.41%)
Diaphyseal humerus (12)	6 (6.82%)
Distal humerus (13)	6 (6.82%)
Diaphyseal radius (2R2)	3 (3.41%)
Distal radius (2R3)	5 (5.68%)
Proximal ulna (2U1)	2 (2.27%)
Diaphyseal ulna (2U2)	1 (1.14%)
Lower extremity	35 (39.77%)
Diaphyseal femur (32)	3 (3.41%)
Distal femur (33)	3 (3.41%)
Proximal tibia (41)	10 (11.36%)
Diaphyseal tibia (42)	5 (5.68%)
Distal tibia (43)	13 (14.77%)
Diaphyseal fibula (4F2)	1 (1.14%)
Foot/ankle	18 (20.45%)
Malleolus (44)	5 (5.68%)
Talus (81)	1 (1.14%)
Calcaneus (82)	12 (13.64%)
N/A	9 (10.23%)
Total	88
Injury location
Upper extremity	27 (30.68%)
Lower extremity	38 (43.18%)
Foot/ankle	22 (25.0%)
Pelvis	1 (1.14%)
Total	88

^*Bone graft substitute

A total of 7 patients (8%, 10 encounters) presented to the emergency department within 30 days. 60% of these encounters (6/10) were related to their injury (Table 2). Twenty-seven patients (31%) visited the emergency department or urgent care 41 times within 1 year of indexed procedure. Of the 41 patient encounters, 10 (24.4%) were associated with the index surgery (eg, pain, concern for infection, cast concern), while 31 (75.6%) were unrelated—26 (63.4%) were for medical reasons, 2 (4.9%) unrelated motor vehicle crashes, 1 (2.4%) fall, and 2 (4.9%) “other” visits (mosquito bites and brief paresthesia). For the 5 encounters for pain (12.2%), 1 occurred within the first 24 hours, 1 within the first 72 hours, and 1 more within the first month. None of the patients were admitted for further pain control issues.

Table 2.

ED visits.

	Total within 12 mo	Days from surgery
		≤30 d	>30 d
Patients with ED visits	27 (30.68%)	7	20
Total ED visits	41	10	31
Number of ED visits*: Median (range)	1 (1–4)	1 (1–3)	1 (1–4)
Reasons for ED visits
Medical	26 (63.41%)	4 (40.0%)	22 (70.96%)
Pain	5 (12.19%)	3 (30.0%)	2 (6.45%)
MVC	2 (4.87%)	0	2 (6.45%)
Fall	1 (2.44%)	0	1 (3.22%)
Concern for infection	3 (7.32%)	0	3 (9.68%)
Other†	4 (9.76%)	3 (30.0%)	1 (3.22%)
		2 cast concerns; 1 paresthesia	1 mosquito bites

^*For patients with ED visits.

^†Cast concern (n = 2), paresthesia (n = 1), mosquito bites (n = 1).

The complication rate from this series was rather high (n = 23, 26%) (Table 3). There were 11 patients (48% with all complications) who required unplanned reoperations (we excluded staged procedures and planned secondary surgeries). Seventeen percent of patients with complications (n = 4) experienced deep surgical site infection. Four patients (17% of patients with complications) had wound complications such as dehiscence or superficial surgical site infections. Other complications included DVT, hardware failure, readmissions, nonunion, malunion, renal failure, nerve injury, avascular necrosis, arthritis, and antibiotic complications.

Table 3.

Complications after outpatient fracture surgery (N = 88).

	Total
Major complication	23 (26.14%)
Deep infection	4 (4.55%)
Superficial infection	3 (3.41%)
DVT	1 (1.14%)
Hardware failure	3 (3.41%)
Reoperation	11 (12.5%)
Readmission	3 (3.41%)
Nonunion	4 (4.55%)
Malunion	0
Dehiscence	1 (1.14%)
Renal failure	0
Nerve injury	1 (1.14%)
Avascular necrosis	0
Arthritis	2 (2.27%)
Abx complication	0

There was no difference (P = 0.11) in opioid prescribing trends from patients who received regional anesthesia and those who did not (Table 4). For the group that received regional anesthesia, 13% (n = 9) continued receiving opioids beyond 1 month; for those without regional anesthesia, 10% (n = 2) continued receiving opioids beyond 1 month.

Table 4.

Opioid prescribing trends (N = 88).

Blocks	Patients (N = 88)	No opioids prescribed	Opioids prescribed in first 30 d postoperatively	Opioids prescribed 1–3 mo postoperatively*	P
Regional/local block	68 (77.3%)	10 (14.7%)	58 (85.3%)	9 (13.2%)	0.11
No block	20 (22.7%)	0	20 (100%)	2 (10.0%)

^*Patients can be in both categories; does not sum to total number of patients.

Patients called the office frequently after outpatient surgery, most often for pain (n = 96/166 calls; 57.8%) (Table 5). However, most (76%; 73/96) of the pain phone calls were simply to discuss pain medication refill according to the OTA/AO guidelines and state legislation. As more time passed from their date of surgery, calls from patients decreased. Medication refill was not only limited to traditional opioid medication but also other adjuncts for pain such as gabapentin, nonsteroidal anti-inflammatory drugs, and acetaminophen, which are encouraged as longer term options for pain control at our institution and within the OTA/AO guidelines.

Table 5.

Office phone calls (N = 166).

Time	Call types
	Total calls to the office	Calls related to pain		Calls for infection concerns	Others
		Pain report	Medication refill
All	166	23	73	16	54
First week	108 (65.1%)	14 (60.9%)	51 (69.9%)	6 (37.5%)	37 (68.5%)
After first week	58 (34.9%)	9 (39.1%)	22 (30.1%)	10 (62.5%)	17 (31.5%)

4. Discussion

As with any shift in patient care, the first question should always be, “Is it safe for the patients?” This series found a low rate of return to the hospital within 30 days (7/88, 8%). The readmission rate overall was only 3%. Patients most often called the office the first week to obtain planned pain medication refills. These results suggest patients can be safely cared for within the outpatient setting for many complex orthopaedic trauma conditions.

Although the literature for outpatient trauma surgery for more complex injuries such as intra-articular calcaneal, peri-articular lower extremity, and complex upper extremity fractures is minimal, there are data regarding injuries, such as ankle fractures, managed in the outpatient setting. Qin et al¹⁵ reported that outpatient management of closed ankle fractures were associated with less urinary tract infections (UTIs), postoperative pneumonia, venous thromboembolism, and reduced 30-day medical morbidity. These findings were consistent with other lower extremity fracture literature which stated that 90-daycomplications for outpatient fracture management was less than those managed inpatient.¹⁶ In addition, shifting lower extremity arthroplasty to the outpatient setting is not associated with any increase in 30-day readmissions or adverse events.^5–7 Postoperative UTIs or pneumonias were not tracked; however, this study's patients had 1 postoperative DVT and 3 readmissions. This rate was lower than what has been reported in orthopaedic literature.²⁴

Another of the main concerns with treating patients in the outpatient setting was these patients returning to local emergency departments for medical complications, pain control issues, or events related to their recovery. This study evaluated every ED and UC visit and found a relatively low rate of visits related to the indexed procedure. Around 12% of patients visit the ED or UC within 1 year of their surgery because of concern for pain or infection. However, upon closer investigation, it was reported that only 1 patient returned to the emergency department within 24 hours and only 1 more within 72 hours. This 24-hour bounce-back rate of ∼1% returning to the ED for care within a day was consistent with other orthopaedic literature.^17,22 When looking at overall ED usage over the year, 41 ED visits in 27 patients were documented among the 88 patients, and most of the visits (28/41 visits, 68%) were unrelated to the study surgery (eg, medical or new trauma). On the surface, this number seemed high, but when compared with the general American population, it provides appropriate context. According to the CDC, each year in the United States there are 40 ED visits for every 100 Americans, a comparable rate to this study.²⁵ When looking at the 30 days after surgery, there were 10 ED visits in 7 patients (8%). Only records in this health care system's electronic medical record (EMR) were able to be reviewed. Although this large health care system has a wide footprint and a large number of emergency departments and urgent care facilities, it is possible the rates reported were an underestimate if these patients presented to facilities outside this system.

Part of transitioning traditionally inpatient surgery to the ambulatory setting revolved around appropriate pain control, following the OTA/AO Guidelines for Musculoskeletal Pain and leveraging regional anesthesia.^{20,24,26–28} In monitoring appropriate pain control, we collected calls to the office and the already-described ED visits. However, patients did not call the office just because of pain. Patients frequently reported wound issues, concern for infection, discussed nonpharmacologic prescriptions such as physical therapy, and requested Workers' Compensation claim paperwork. Thus, documenting the reason for all calls allowed a broader understanding of what patients needed acutely after surgery. It was found that due to state law and implementation of the OTA/AO guidelines, most calls to the office were solely to have medication refilled as planned, which was viewed as a success. Although patients frequently called the office, we were unable to compare how the volume of phone calls compares with calls from patients treated in the inpatient setting.

This finding highlights a key component to any successful outpatient trauma surgery program—communication. With patients no longer in a hospital bed, the questions that are usually answered in house before discharge now needed to be answered in the emergency department or clinic, before discharge from the recovery room, or remotely. Recently, there has been the greatest shift in delivering care remotely. During the study period, the COVID-19 pandemic started, changing how patients receive care and their experience of remote medicine. Patients and surgeons seemed to be similarly satisfied with care delivered virtually and have no changes in patient-reported outcomes in regard to pain and function.^29,30 Because of this, patient concerns could continue to be addressed electronically through the EMR or through virtual platforms to help answer questions and minimize UC and ED visits as well as time-consuming office visits.

Although cost was not directly addressed, it must be discussed to provide a thorough review of inpatient versus outpatient orthopaedic surgery. Minimal data currently exist in the complex orthopaedic trauma world regarding true cost analysis of inpatient verse outpatient surgery because this is a relatively novel concept. However, other orthopaedic subspecialties have addressed this issue well and have found that ankle, hip, and knee arthroplasty performed outpatient costs up to 30% less than those completed inpatient.^7,31,32 These findings were consistent with this system's foot and ankle colleagues who reported significant cost savings and better outcomes when moving patients to the outpatient setting for isolated ankle fractures.¹⁶ This was especially important to our patients. They continue to bear more of the cost of surgery as deductibles have skyrocketed colinearly to an increase in cost containment and catastrophic-only policies.^33,34 It is prudent for surgeons to recognize these challenges that our patients are facing.

An obvious limitation of this study was that patients did not fill out satisfaction scores throughout their care. Although it is the overall impression that patients were satisfied with their care, this finding was not able to be objectively reported. In addition, this study was limited to the outcomes of a single surgeon at 1 institution with 1 anesthesia team. However, there are very reproducible steps that can be completed to achieve a similar setup at all institutions. Now that feasibility and safety have been documented in a single-surgeon series, the dissemination process can proceed to additional surgeons and/or institutions. Finally, although the lead surgeon does not have defined criteria for selecting patients for outpatient surgery, anesthesia does (Appendix 1, http://links.lww.com/OTAI/A112). This could help self-select patients who are more likely to do well from outpatient management of their injuries. Another limitation is the lack of data regarding similar cases performed in the inpatient setting. Comparing outcomes among these 2 groups is an important next step.

Ultimately, implementation of an outpatient model in orthopaedic trauma required 3 key steps. First was willingness to change the current model. This required not only orthopaedic surgeon involvement but also the emergency department and anesthesia willing to embrace this shift in delivering care. Second was education. Patients needed to be educated from the moment they arrived in the emergency department with their initial injury regarding plans for treatment. They also then needed to be continually educated about how to address pain control postoperatively, how to mobilize safely, and what would be concerning in their recovery that would prompt return to clinic or emergency department. Third was proper patient selection. Although this study did not establish well-defined criteria for patient selection outside of anesthesia requirements documented in the Appendix, http://links.lww.com/OTAI/A112, it was important to consider mobility concerns, living situation (ie, support at home and distance from hospital), and medical comorbidities as part of one's algorithm for patient selection. By transitioning these patients to the outpatient setting, a safe alternative was found that could help deliver better care while relieving patients and hospitals of long inpatient stays that have become all too frequent with staffing issues, patient overflow, and ongoing constraints of inpatient facilities.