Background:
The prevalence of bone stimulator use among nonoperatively and operatively managed scaphoid nonunion patients is unknown. We hypothesize that bone stimulators are a relatively underutilized treatment for scaphoid nonunion patients.
Methods:
We used the 2009–2017 Truven Marketscan Research Databases to identify patients with closed scaphoid fractures and performed an analysis of variance test to determine resource utilization and bone stimulator use among these patients.
Results:
A total of 36,611 patients with scaphoid fractures were identified: 30,143 were managed nonoperatively and 6468 were managed operatively. Nonunion was diagnosed in 500 (1.66%) nonoperatively and in 1211 (19%) operatively managed patients. Bone stimulators were used in less than 2% of nonoperatively and operatively managed scaphoid nonunion patients.
Conclusion:
Lack of trust in the technology and heterogenous (and occasionally burdensome) requirements for insurance approval are barriers to bone stimulator use; however, surgeons should examine how this technology may fit into the treatment algorithm for these difficult cases.
Takeaways
Question: What is the role of bone stimulators among scaphoid fracture patients?
Findings: Bone stimulators are infrequently used among scaphoid fracture patients with nonunion.
Meaning: Surgeons should consider where bone stimulators can be used to adjunct healing among complicated scaphoid nonunion patients.
INTRODUCTION
The scaphoid is the most fractured wrist bone, comprising 50%–90% of all carpal fractures.1 The retrograde blood supply of the scaphoid makes it one of the most difficult bones to heal.2 Treatment is largely predicated on the location of fracture and whether displacement has occurred. Nondisplaced fractures of the waist are most common and often managed successfully with a thumb spica cast, whereas displaced fractures and those in the proximal pole typically require surgical fixation.3,4 Nonunion is a radiographic diagnosis, typically defined as the lack of bridging bone across a fracture site 6 months after an injury; the rate of scaphoid nonunion varies in the literature from 5% to 12% with some studies citing rates as high as 47%.5 Because scaphoid nonunion can lead to scaphoid nonunion advanced collapse and subsequent early onset osteoarthritis, increased attention is paid to the acute diagnosis.6–10
Early identification of scaphoid fractures, as well as minimizing risk factors for nonunion, results in better patient outcomes at decreased costs. Surgical intervention for scaphoid fractures is variable and depends on the location of the fracture and quality of the remaining bone. Interventions include open/percutaneous reduction and fixation, as well as open reduction and internal fixation (ORIF) with pedicled or free vascularized bone flaps. Despite the advances in surgical techniques, the overall cost to the healthcare system to care for scaphoid fracture patients can be tremendous with frequent imaging, occupational therapy needs, chronic pain medication consumption, and multiple procedures.11,12
Bone stimulators induce new bone formation, promote fracture healing, and can ignite healing in patients with nonunion.13 They have been successfully used to augment spinal fusion as well as femur, tibial, and radius fracture healing.14–16 Despite this, bone stimulator use for scaphoid nonunion remains controversial, and it has yet to be widely accepted. Insurance companies varying and arduous requirements for approval further limit its use.17 We investigated the use of electrical bone stimulators among nonoperatively and operatively managed scaphoid nonunion patients. We hypothesize that bone stimulators are relatively underutilized in the management of scaphoid nonunion patients.
METHODS
Data Source and Study Cohort
We used the Truven Health Marketscan Commercial and Supplemental Database from 2009 to 2017 to identify patients with closed scaphoid fractures. This database consists of patient-level medical records, reimbursements, and health care expenditures across inpatient and outpatient settings. The health care usage of over 90 million active employees, early retirees, Medicare-eligible retirees, and their dependents are represented by Marketscan each year. This extensive database includes over 30 billion health records representing inpatient and outpatient encounters, making it the most comprehensive employer-based collection of patient data in the United States; it includes health insurance claims from private insurers, Medicare, and Medicaid, and claims from the uninsured.
For the current study, we identified patients with International Classification of Diseases, Ninth Revision and Tenth Revision diagnosis codes for closed scaphoid fractures or scaphoid nonunion between January 1, 2009, and December 31, 2015. We required patients to be enrolled at least 1 month before the date of their fracture diagnosis to exclude patients with other acute hand/wrist trauma or open scaphoid fracture within this time frame. Additionally, we required 24 consecutive months of enrollment and excluded patients who were younger than 18 or older than 65. We divided the patient cohort into operative and nonoperative groups. Patients in the operative group were identified with specific scaphoid operative procedure CPT codes. Nonoperative patients were identified by CPT code for closed reduction and/or lack of an operative CPT code (Fig. 1). We split each group of patients into nonunion and routine healing subgroups based on ICD 9 or 10 codes. To better characterize the treatment options, operative scaphoid fracture patients were divided into ORIF, open reduction and internal fixation with bone graft (ORIF + bone graft), open reduction and internal fixation with vascularized bone graft (ORIF and VBG), and open reduction and internal fixation with free microvascular anastomosis (ORIF with MVA). We performed descriptive statistics based on surgical treatment and patient demographic variables (eg, gender, age, region of residence, smoking status, employment status, and insurance type). An analysis of the association between the previously described patient variables and the type of treatment received was conducted using a χ2 test.
Fig. 1.
Study flow demonstrating inclusion and exclusion criteria for the study groups.
Resource Utilization
Operative and nonoperative groups were followed up for 24 months to assess resource utilization. We calculated the mean number of clinic visits, imaging beyond the index radiograph, operating room costs, and occupational therapy session costs. We determined the costs associated with each patient based on their claim records. For example, the cost of a radiograph was determined by identifying the cost of a single claim record with the CPT code for the radiograph. The total charges for each clinic visit encounter in the patient’s record were used to determine the cost of an office visit. Procedural cost data are comprised of the total cost of the encounter and include primarily the charges for the surgeon, operating room, and anesthesia. We used all the charges (radiographs, clinic visits, occupational therapy visits, and surgery) to calculate a mean cost associated with a particular treatment. Using the analysis of variance test, we compared the resource utilization between nonunion and routine healing subgroups among the surgical and nonsurgical patients. Statistical analyses were performed using SAS 9.4 software (SAS Institute, Inc., Cary, N.C.) and R 4.0.3 software [R Core Team (2020)].
RESULTS
Nonoperatively Managed Scaphoid Fractures
A total of 36,611 patients with scaphoid fractures were identified; 30,143 were managed nonoperatively. Most patients were men (P < 0.001), young working-age individuals, and 18–34 years old (P = 0.0004). Among patients managed nonoperatively, nonunion was diagnosed in 500 (1.66%). A greater proportion of nonunion patients were smokers compared with those who went on the routine healing (12% versus 7.7%), P = 0.001 (Table 1). Bone stimulators were used in less than 2% of nonoperatively managed patients with nonunion. Examination of insurance type, employment status, and regional location did not demonstrate a significant difference in the development of nonunion among nonoperatively managed patients.
Table 1.
Demographic Data for Nonoperative Scaphoid Fractures
| Scaphoid Fractures Managed Nonoperatively | |||
|---|---|---|---|
| Characteristic | Routine Healing | Nonunion | P |
| Number | 29,643 (98.34) | 500 (1.66) | |
| Total | 30,143 | ||
| Sex | |||
| Male | 16,963 (57.22) | 377 (75.4) | <0.001 |
| Female | 12,680 (42.78) | 123 (24.6) | |
| Age | |||
| 18–34 | 12,863 (43.39) | 248 (49.6) | 0.004 |
| 35–44 | 4830 (16.29) | 84 (16.8) | |
| 45–54 | 6044 (20.39) | 97 (19.4) | |
| 55–64 | 5906 (19.92) | 71 (14.2) | |
| Insurance | |||
| PPO | 17,966 (60.61) | 306 (61.2) | 0.766 |
| Comprehensive | 528 (1.78) | 11 (2.2) | |
| HMO | 3727 (12.57) | 63 (12.6) | |
| POS | 2079 (7.01) | 39 (7.8) | |
| Other | 5343 (18.02) | 81 (16.2) | |
| Smoking status | |||
| Never smoker | 27,360 (92.30) | 440 (88) | 0.001 |
| Smoker | 2283 (7.70) | 60 (12) | |
| Employment status | |||
| Employed | 14,339 (48.37) | 236 (47.2) | 0.928 |
| Retiree | 1509 (5.09) | 24 (4.8) | |
| Disability | 63 (0.21) | 1 (0.2) | |
| Other/unemployed | 13,732 (46.32) | 239 (47.8) | |
| Region | |||
| Northeast | 6536 (22.05) | 107 (21.4) | 0.964 |
| North Central | 6178 (20.84) | 103 (20.6) | |
| South | 9822 (33.13) | 163 (32.6) | |
| West | 6521 (22.00) | 116 (23.2) | |
| Bone stimulator use | |||
| Invasive | 2 (0.003) | 2 (0.4) | |
| Noninvasive | 31 (0.10) | 6 (1.2) | |
Operatively Managed Scaphoid Fractures
Of the patients with scaphoid fractures, 6468 were managed operatively, and nonunion was diagnosed in 1211 (19%). There were 4783 patients who had records for the 24-month study period and were included in the in-depth analysis. A greater proportion of the surgical patients were men, P = 0.008 (Table 2). Most patients who underwent surgery were between ages 18 and 34 (<0.001). Smoking was not prohibitive for surgery and overall, and 10% of operative patients were identified as smokers (Table 2). When grouped together, smoking increased the chances of nonunion in a statistically significant manner among operatively managed scaphoid fracture patients, P = 0.034 (Table 2). Examination of the time lag between fracture diagnosis and surgery among patients who went on to routine healing and those who developed nonunion revealed a significant difference between groups (routine healing 28 days versus nonunion 183 days, P < 0.001) (Table 2). Bone stimulator use among the surgical groups with nonunion was also less than 2%.
Table 2.
Demographic Data for Operatively Managed Scaphoid Fractures
| Scaphoid Fractures Managed Operatively | |||
|---|---|---|---|
| Characteristic | Routine Healing | Nonunion | P |
| Number | 5257 (81.28) | 1211 (18.72) | |
| Total | 6468 | ||
| Sex | |||
| Male | 4295 (81.70) | 1028 (84.89) | 0.008 |
| Female | 962 (18.30) | 183 (15.11) | |
| Age | |||
| 18–34 | 3676 (69.93) | 921 (76.05) | <0.001 |
| 35–44 | 653 (12.42) | 128 (10.57) | |
| 45–54 | 516 (9.82) | 114 (9.41) | |
| 55–64 | 412 (7.84) | 48 (3.96) | |
| Insurance | |||
| PPO | 3226 (61.37) | 752 (62.10) | 0.785 |
| Comprehensive | 70 (1.33) | 13 (1.07) | |
| HMO | 604 (11.49) | 133 (10.98) | |
| POS | 378 (7.19) | 79 (6.52) | |
| Other | 979 (18.62) | 234 (19.32) | |
| Smoking status | |||
| Never smoker | 4749 (90.34) | 1069 (88.27) | 0.034 |
| Smoker | 508 (9.66) | 142 (11.73) | |
| Employment status | |||
| Employed | 2736 (52.04) | 616 (50.87) | 0.434 |
| Retiree | 171 (3.25) | 31 (2.56) | |
| Disability | 6 (0.11) | 1 (0.08) | |
| Other/unemployed | 2344 (44.59) | 563 (46.49) | |
| Region | |||
| Northeast | 1018 (19.36) | 232 (19.16) | 0.833 |
| North Central | 1158 (22.03) | 271 (22.38) | |
| South | 1734 (32.98) | 404 (33.36) | |
| West | 1253 (23.83) | 288 (23.78) | |
| Bone stimulator use | |||
| Invasive | 2 (0.04) | 2 (0.17) | |
| Noninvasive | 42 (0.80) | 18 (1.49) | |
| Duration from initial fracture, mean ± SD | 28.4 ± 61.39 | 183.83 ± 202.57 | <0.001 |
The following operative procedures were performed for scaphoid fracture patients: ORIF of scaphoid fracture (ORIF), ORIF + bone graft, ORIF and VBG, and ORIF with MVA. Among the most performed surgical procedures for scaphoid fractures, the prevalence of nonunion after surgical management was as follows: ORIF, 8.2%; ORIF + bone graft, 23%; ORIF + VBG, 49%; and ORIF + MVA, 57% (Table 3). ORIF + VBG and ORIF + MVA are rarely utilized as the first operation in the treatment of a scaphoid fracture unless specific circumstances demand it. Delayed patient presentation (>3 months) and/or location of the fracture (proximal pole fractures) may explain these findings. The higher rates of nonunion among the patients who underwent more advanced procedures (VBG and MVA) underpin the difficulty of these cases and need for early indicators of delayed bone healing.
Table 3.
Demographic Characteristics based on Surgical Intervention
| Open Reduction with or without Internal Fixation (25,628) | Open Reduction with or without Internal Fixation (25,628) + Bone Graft (20,902 or 20,900) | Insertion of Vascular Pedicle into Carpal Bone (25,430) | Bone Graft with Microvascular Anastomosis (20,962) | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Characteristic | Routine Healing | Nonunion | P | Routine Healing | Nonunion | P | Routine Healing | Nonunion | P | Routine Healing | Nonunion | P |
| Number | 3736 | 335 | 289 | 90 | 134 | 131 | 21 | 28 | ||||
| Total | 4071 | 379 | 265 | 49 | ||||||||
| Sex | 0.115 | |||||||||||
| Male | 2992 (80.09) | 280 (83.58) | 243 (84.08) | 74 (82.22) | 0.679 | 123 (91.79) | 115 (87.79) | 0.280 | 17 (80.95) | 25 (89.29) | 0.412 | |
| Female | 744 (19.91) | 55 (16.42) | 46 (15.92) | 16 (17.78) | 11 (8.21) | 16 (12.21) | 4 (19.05) | 3 (10.71) | ||||
| Age | 0.001 | |||||||||||
| 18–34 | 2501 (66.94) | 252 (75.22) | 208 (71.97) | 68 (75.56) | 0.716 | 112 (83.58) | 110 (83.97) | 0.960 | 15 (71.43) | 23 (82.14) | 0.231 | |
| 35–44 | 499 (13.36) | 44 (13.13) | 29 (10.03) | 8 (8.89) | 12 (8.96) | 11 (8.4) | 3 (14.29) | 5 (17.86) | ||||
| 45–54 | 402 (10.76) | 26 (7.76) | 30 (10.38) | 10 (11.11) | 7 (5.22) | 8 (6.11) | 2 (9.52) | 0 (0) | ||||
| 55–64 | 334 (8.94) | 13 (3.88) | 22 (7.61) | 4 (4.44) | 3 (2.24) | 2 (1.53) | 1 (4.76) | 0 (0) | ||||
| Insurance | 0.462 | |||||||||||
| PPO | 2264 (60.6) | 197 (58.81) | 182 (62.98) | 53 (58.89) | 0.257 | 77 (57.46) | 90 (68.7) | 0.081 | 9 (42.86) | 17 (60.71) | 0.212 | |
| Comprehensive | 50 (1.34) | 3 (0.90) | 5 (1.73) | 3 (3.33) | 2 (1.49) | 0 (0) | 1 (4.76) | 1 (3.57) | ||||
| HMO | 439 (11.75) | 33 (9.85) | 28 (9.69) | 8 (8.89) | 13 (9.7) | 14 (10.69) | 4 (19.05) | 1 (3.57) | ||||
| POS | 277 (7.41) | 28 (8.36) | 15 (5.19) | 9 (10) | 18 (13.43) | 8 (6.11) | 3 (14.29) | 1 (3.57) | ||||
| Other | 706 (18.90) | 74 (22.09) | 59 (20.42) | 20 (22.22) | 24 (17.91) | 19 (14.5) | 4 (19.05) | 8 (28.57) | ||||
| Smoking Status | 0.581 | |||||||||||
| Never smoker | 3370 (90.20) | 299 (89.25) | 269 (93.08) | 82 (91.11) | 0.541 | 120 (89.55) | 113 (86.26) | 0.410 | 17 (80.95) | 22 (78.57) | 0.838 | |
| Smoker | 366 (9.80) | 36 (10.75) | 20 (6.92) | 8 (8.89) | 14 (10.45) | 18 (13.74) | 4 (19.05) | 6 (21.43) | ||||
| Employment status | 0.463 | |||||||||||
| Employed | 1946 (52.09) | 165 (49.25) | 163 (56.40) | 40 (44.44) | 0.127 | 77 (57.46) | 63 (48.09) | 0.223 | 10 (47.62) | 15 (53.57) | 0.680 | |
| Retiree | 127 (3.40) | 9 (2.69) | 4 (1.38) | 1 (1.11) | 3 (2.24) | 2 (1.53) | 0 (0) | 0 (0) | ||||
| Disability | 4 (0.11) | 0 (0) | 0 (0) | 0 (0) | 1 (0.75) | 0 (0) | 0 (0) | 0 (0) | ||||
| Other/unemployed | 1659 (44.40) | 161 (48.06) | 122 (42.21) | 49 (54.44) | 53 (39.55) | 66 (50.38) | 11 (52.38) | 13 (46.43) | ||||
| Region | 0.751 | |||||||||||
| Northeast | 720 (19.27) | 73 (21.79) | 68 (23.53) | 17 (18.89) | 0.702 | 35 (26.12) | 30 (22.9) | 0.351 | 6 (28.57) | 3 (10.71) | 0.277 | |
| North Central | 823 (22.03) | 74 (22.09) | 61 (21.11) | 19 (21.11) | 28 (20.9) | 21 (16.03) | 4 (19.05) | 10 (35.71) | ||||
| South | 1260 (33.73) | 106 (31.64) | 89 (30.8) | 26 (28.89) | 45 (33.58) | 41 (31.3) | 7 (33.33) | 11 (39.29) | ||||
| West | 869 (23.26) | 78 (23.28) | 68 (23.53) | 26 (28.89) | 23 (17.16) | 36 (27.48) | 4 (19.05) | 3 (10.71) | ||||
| Bone stimulator use | ||||||||||||
| Invasive | 1 (0.03) | 0 (0) | 1 (0.35) | 0 (0) | 0 (0) | 0 (0) | 0 (0) | 0 (0) | ||||
| Noninvasive | 24 (0.64) | 5 (1.49) | 2 (0.69) | 1 (1.11) | 2 (1.49) | 6 (4.58) | 0 (0) | 0 (0) | ||||
Resource Utilization among Operatively and Nonoperatively Managed Scaphoid Fractures
We individually characterized resource utilization among nonoperatively and operatively managed patients with scaphoid fractures who went on to routine healing or nonunion. As expected, nonunion patients had a greater number of clinical visits across all treatments when compared with those who went on to routine healing (P < 0.001) (Table 4). Patients who underwent ORIF + VBG with routine healing had the lowest mean number of clinic visits at 6.93 visits compared with those who underwent ORIF with bone graft, mean number of clinic visits at 14.7 visits (P < 0.001) (Table 4). Patients with routine healing after nonoperative management had the lowest mean number of plain radiographs (three) and CTs (one) compared with those who underwent free vascularized bone graft who had a mean 13 plain radiographs and four CTs during their treatment course (P < 0.001) (Table 4). Occupational therapy is used variably for scaphoid fracture patients managed operatively and nonoperatively; there was no statistically significant difference in the mean number of sessions, but those who were treated with ORIF + MVA had the longest duration of OT, 420 days (P < 0.001) (Table 4). Overall, it is resource intensive to provide care for scaphoid fracture patients; the mean costs to care for these patients ranged from $10,967 for a nonoperative patient with routine healing compared with $33,166 for a patient who underwent vascularized bone graft with nonunion (Fig. 2).
Table 4.
Resource Utilization of Scaphoid Fracture Patients
| Scaphoid Fracture Management | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| Nonoperatively Managed | Open Reduction with or without Internal Fixation (25,628) | Open Reduction with or without Internal Fixation (25,628) + Bone Graft (20,902 or 20,900) 379 | Insertion of Vascular Pedicle into Carpal Bone (25,430) | Bone Graft with Microvascular Anastomosis (20,962) | |||||||
| Medical Services | Routine Healing | Nonunion | RoutineHealing | Nonunion | Routine Healing | Nonunion | Routine Healing | Nonunion | Routine Healing | Nonunion | P |
| Total | 29,643 | 500 | 3736 | 335 | 289 | 90 | 134 | 131 | 21 | 28 | |
| Clinic visit (99,211–99,215) | |||||||||||
| Number, mean ± SD | 9.83 ± 11.24 | 14.43 ± 14.94 | 7.39 ± 8.72 | 13.28 ± 16.24 | 7.56 ± 8.47 | 14.71 ± 21.14 | 6.93 ± 6.97 | 12.62 ± 16.97 | 11.2 ± 15.82 | 11.14 ± 10.38 | <0.001 |
| Costs per session (USD), mean ± SD | 87.35 ± 53.92 | 89.78 ± 77.43 | 87.67 ± 48.45 | 83.34 ± 45.56 | 88.62 ± 41.48 | 83.31 ± 37.92 | 85.09 ± 56.01 | 91.52 ± 55.42 | 88.18 ± 26.10 | 80.12 ± 34.96 | 0.091 |
| Imaging | |||||||||||
| XR (73,100 and 73,110) | |||||||||||
| Number, mean ± SD | 3.38 ± 2.53 | 5.95 ± 4.41 | 6.07 ± 3.44 | 12.04 ± 7.38 | 6.83 ± 3.67 | 11.82 ± 8.47 | 6.56 ± 3.56 | 11.82 ± 7.82 | 8.4 ± 6.29 | 13.14 ± 8.80 | <0.001 |
| Costs per session (USD), mean ± SD | 50.78 ± 62.95 | 56.95 ± 103.81 | 51.21 ± 67.61 | 51.16 ± 59.19 | 49.96 ± 83.29 | 53.70 ± 69.52 | 46.42 ± 45.80 | 49.06 ± 51.19 | 40.66 ± 40.03 | 59.26 ± 62.38 | 0.214 |
| Percentage of having XR | 84.84 | 92.6 | 97.64 | 99.70 | 98.62 | 98.89 | 98.51 | 100 | 95.24 | 100 | |
| MRI (73,218, 73,219, and 73,220) | |||||||||||
| Number, mean ± SD | 1.38 ± 0.76 | 1.86 ± 1.07 | 1.25 ± 0.53 | 1.86 ± 1.07 | 1 ± 0 | 2.25 ± 1.26 | 1 | 0 | 1 | 0 | 0.65 |
| Costs per session (USD), mean ± SD | 550.71 ± 578.17 | 826.75 ± 886.56 | 508.27 ± 349.44 | 758.22 ± 698.52 | 917.53 ± 246.23 | 996.39 ± 715.04 | 1014.49 | N/A | 441 | N/A | 0.096 |
| Percentage of having MRI | 1.18 | 1.4 | 0. 64 | 2.09 | 0.69 | 4.44 | 0.75 | 0 | 4.76 | 0 | |
| CT (73,200, 73,201, and 73,202) | |||||||||||
| Number, mean ± SD | 1.69 ± 0.93 | 2.37 ± 2.02 | 1.95 ± 1.23 | 3.33 ± 2.68 | 2.16 ± 1.67 | 2.66 ± 2.00 | 2.11 ± 1.31 | 3.62 ± 2.96 | 2 ± 0.82 | 4.33 ± 3.81 | <0.001 |
| Costs per session (USD), mean ± SD | 336.82 ± 372.28 | 371.96 ± 392.30 | 365.90 ± 430.42 | 379.99 ± 375.82 | 369.49 ± 371.21 | 415.09 ± 429.88 | 340.66 ± 307.88 | 315.36 ± 306.03 | 479.57 ± 841.28 | 438.81 ± 379.81 | <0.001 |
| Percentage of having CT | 13.92 | 40.8 | 32.09 | 68.96 | 50.17 | 71.11 | 55.22 | 75.57 | 47.62 | 75 | |
| Occupational therapy | |||||||||||
| Number, mean ± SD | 6.69 ± 11.74 | 6 ± 8.40 | 6.25 ± 11.36 | 7.86 ± 13.66 | 6.69 ± 13.47 | 5.29 ± 9.39 | 3.90 ± 5.39 | 4.90 ± 7.29 | 1.25 ± 0.5 | 13.7 ± 20.32 | 0.377 |
| Costs per session (USD), mean ± SD | 52.03 ± 151.38 | 43.84 ± 50.78 | 52.03 ± 50.47 | 58.67 ± 52.95 | 47.89 ± 59.62 | 51.14 ± 47.35 | 55.66 ± 42.22 | 43.26 ± 51.29 | 36.7 ± 26.79 | 41.63 ± 23.90 | 0.802 |
| Duration from initial fracture, mean ± SD | 240.67 ± 216.03 | 274.86 ± 213.94 | 168.27 ± 182.66 | 330.30 ± 203.29 | 259.79 ± 303.64 | 320.91 ± 206.89 | 149.77 ± 60.65 | 282.13 ± 181.92 | 157.4 ± 164.22 | 420.26 ± 251.64 | <0.001 |
| Procedural costs (USD), mean ± SD | N/A | N/A | 2893.82 ± 3111.98 | 2828.39 ± 3305.40 | 938.79 ± 653.68 | 861.11 ± 1060.98 | 1087.71 ± 2274.51 | 1022.85 ± 1148.80 | 5088.05 ± 4856.8 | 7734.88 ± 11096.69 | <0.001 |
| Bone stimulator use | |||||||||||
| Cost of therapy | 132.45 ± 253.76 | 226.13 ± 453.79 | 60.15 ± 61.65 | 102.02 ± 134.88 | 92.97 ± 105.01 | N/A | 89.14 ± 2.39 | 107.78 ± 103.61 | N/A | N/A | |
Fig. 2.
Costs of fracture healing. A, Mean total costs ($) associated with for scaphoid fractures with routine healing based on treatment choice. B, Mean total costs ($) associated with care for scaphoid fractures with nonunion based on treatment choice.
DISCUSSION
In the current study, we found that bone stimulator use among nonoperatively and operatively managed scaphoid nonunion patients was less than 2%. This technology has been around for some time; in 1979, the FDA approved the first electrical bone stimulation device for the treatment of nonunion and congenital pseudarthroses.15 Since then, multiple noninvasive external devices have been developed that deliver different modes of energy including electric and ultrasonic stimulation. Several companies manufacture FDA-approved bone stimulators that deliver electromagnetic energy including Zimmer Biomet (devices OsteGen and OsteoGenM) and Don Joy Global (device CMF OL1000). Among the ultrasonic devices, Exogen manufactured the first FDA-approved device, and more recently, Orthofix received approval for Accelstim, an additional low-intensity ultrasound bone stimulator.18
Bone stimulators have been used to treat scaphoid nonunion since 1982 with one study demonstrating a 75% healing rate among patients whose previous surgery failed.19 After this, Bora et al20 demonstrated a 71% healing rate after 12 weeks of noninvasive bone stimulator use among scaphoid nonunion patients. In a recent randomized study, the authors examined the impact of one dose of intraoperatively administered electrical bone stimulation following ORIF with nonvascularized bone graft among scaphoid nonunion patients; the addition of bone stimulator resulted in a greater proportion of healed patients by 12 weeks postoperatively.21 Bone stimulators have also been used to accelerate bone healing among acute scaphoid fractures; Mayr et al22 compared the effects of adjunctive bone stimulation (ultrasound) on acute fracture healing and found radiographic healing in the experimental group at 43 days compared with 62 days in the control group; this was statistically significant. Hannemann et al23,24 demonstrated similar findings among acute scaphoid fractures treated with pulsed electromagnetic bone stimulation. In addition, they examined the cost-effectiveness of pulsed electromagnetic field bone stimulators among acute scaphoid fracture patients and found a reduction in total working days lost from 12.9 to 9.8 days.23,24 Despite its use to treat acute fractures, current indications for the use of bone stimulators in the US include failed bone graft procedures, patient refusal of surgery, or patient who is unsafe for surgery. Extrapolating from this, bone stimulators may play a role in faster clinical and radiographic healing among nonunion patients, and that may translate to earlier return to activities. Once initiated, patients should anticipate at least 3–4 months (sometimes longer) of treatment to determine its effect.
Bone stimulators have been used as an adjunct to healing among femur fracture nonunions as well as spinal fusion nonunions. Hughes and Anglen25 surgically implanted 121 direct current bone stimulators for nonunion fractures and arthrodesis patients and found that 85% had radiographic and/or clinical healing of their nonunion site at 7.1 months. Nolte et al26 examined the effect of low-intensity ultrasound bone stimulators among a heterogenous group of fracture nonunion patients (tibia, femur, radius, and scaphoid) and found 86% of patients healed, with an average healing time of 152 days.
The low prevalence of bone stimulator use for scaphoid nonunion in the current study is perhaps due in part to increased costs of this technology as well as the stringent, burdensome, and variable requirements for approval by insurance companies. Some surgeons may be unaware of the appropriate CPT codes (20974: electrical stimulation to aid bone healing, noninvasive and 20979: low-intensity ultrasound stimulation to aid bone healing, noninvasive) and Healthcare Common Procedural Coding System codes (E0747: osteogenesis stimulator, electrical, noninvasive, and other than spinal applications and E0760: osteogenesis stimulator, low-intensity ultrasound, and noninvasive) that are crucial for approval and necessary for claim submission.27 Huang et al17 surveyed medical directors representing over 119 million insured patients regarding a hypothetical clinical scenario of delayed bone healing and request for electrical bone stimulator. They found that 84% of insurance companies required specific time frames of delayed healing, 76% required serial plain radiographs, 44% required documentation about lack of infection, 36% required documentation of the size of the fracture gap, and 17% required physicians to document clinical signs of nonunion. With the demands of a busy clinical practice, many surgeons have difficulty fulfilling these requirements, limiting the approval and use of these devices.
Examination of resource utilization among scaphoid fracture patients revealed that operatively and nonoperatively managed patients with nonunion were costly to health care system with mean costs to care for nonunion scaphoid fracture patients that were consistently greater across all surgical treatments (>$10,000) compared with care scaphoid fractures with routine healing (P < 0.001). These costs are due in part to the greater number of clinic (outside the global period) and occupational therapy visits, increased frequency of plain radiographs, and use of more advanced imaging [computerized tomography/magnetic resonance imaging (CT/MRI)], as well as the performance of more complex procedures in these groups. Less frequent, advanced imaging at critical intervals can be an avenue to decrease health care expenses.28–31 External bone stimulator costs vary depending on manufacturer and insurance coverage, but range between $500 and $5000.32 Button et al33 performed a systematic review of the literature focusing on the economic analysis of bone stimulator use and found that among tibial fracture patients with delayed healing and/or nonnuion, a total cost savings of $15,00 per case (20-40% reduction) was achieved. Examining our data on resource utilization, it seems feasible to consider the use of bone stimulators in one’s treatment algorithm rather than obtaining repeated costly imaging such as MRI or CT, that is, $500–1000/study in patients with nonunion. When considering the health care expenditure among scaphoid nonunion patients, surgeons may consider bone stimulator use in place of repeat CT and/or MRI.
This study has some limitations. Similar to any retrospective administrative data study, this analysis is dependent on the accuracy of the data reported; however, Marketscan is a well-established medical database with proven credibility. The mechanism and location of the scaphoid fracture are unknown in this study group; however, we followed up patients for 24 months and minimized confounders by excluding patients with open fractures and other hand/wrist injuries. Because this is a database study, we do not have the perioperative discussions and clinical goals of the surgeon and patients in the decision to pursue nonoperative or operative treatment. In addition, we do not have data on the patient’s clinical examination; although they may have developed nonunion, they may be clinically asymptomatic or “well.” Because the Marketscan database provides code-specific data rather than individual patient notes, we were unable to identify how, when, and if patients who used bone stimulators improved. Extrapolating from the randomized control trials in the literature on the use of bone stimulators for scaphoid fractures, we have found that bone stimulators led to some clinical or radiographic improvement among 70%–75% of treated patients which is promising for the patients in whom it was used.19,20
As medicine continues to advance and technologies emerge, surgeons must determine how these adjuncts complement traditional treatments and factor into the overall management strategy. The use of bone stimulators to treat scaphoid nonunion patients is relatively low despite some evidence in the literature to suggest its ability to promote bone healing in these circumstances. This underutilization may be related to surgeons’ awareness and understanding of the technology, trust in its efficacy, and/or willingness to pursue insurance approval. As more evidence about its usefulness emerges, surgeons may consider this treatment adjunct rather than additional imaging in patients with established nonunion whose previous surgery failed or who have exhausted all operative interventions.
Footnotes
Published online 26 January 2023.
Disclosure: The authors have no financial interest to declare in relation to the content of this article.
Related Digital Media are available in the full-text version of the article on www.PRSGlobalOpen.com.
REFERENCES
- 1.Alshryda S, Shah A, Odak S, et al. Acute fractures of the scaphoid bone: systematic review and meta-analysis. Surgeon. 2012;10:218–229. [DOI] [PubMed] [Google Scholar]
- 2.Fowler JR, Hughes TB. Scaphoid fractures. Clin Sports Med. 2015;34:37–50. [DOI] [PubMed] [Google Scholar]
- 3.Gelberman RH, Menon J. The vascularity of the scaphoid bone. J Hand Surg Am. 1980;5:508–513. [DOI] [PubMed] [Google Scholar]
- 4.Kawamura K, Chung KC. Treatment of scaphoid fractures and nonunions. J Hand Surg. 2008;33:988–997. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Ko JH, Pet MA, Khouri JS, et al. Management of scaphoid fractures. Plast Reconstr Surg. 2017;140:333e–346e. [DOI] [PubMed] [Google Scholar]
- 6.Buijze GA, Ochtman L, Ring D. Management of scaphoid nonunion. J Hand Surg Am. 2012;37:1095–100. [DOI] [PubMed] [Google Scholar]
- 7.Meier R. Behandlungsoptionen beim karpalen Kollaps nach Skaphoidfraktur [Treatment options for scaphoid nonunion advanced collapse]. Unfallchirurg. 2019;122:211–218. [DOI] [PubMed] [Google Scholar]
- 8.Little CP, Burston BJ, Hopkinson-Woolley J, et al. Failure of surgery for scaphoid non-union is associated with smoking. J Hand Surg Br. 2006; 31:252–255. [DOI] [PubMed] [Google Scholar]
- 9.Tada K, Ikeda K, Okamoto S, et al. Scaphoid fracture—overview and conservative treatment. Hand Surg. 2015;20:204–209. [DOI] [PubMed] [Google Scholar]
- 10.Huang YC, Liu Y, Chen TH. Long-term results of scaphoid nonunion treated by intercalated bone grafting and Herbert’s screw fixation—a study of 49 patients for at least five years. Int Orthop. 2009;33:1295–1300. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Schuind F, Moungondo F, Kazzi EE. Prognostic factors in the treatment of scaphoid non-unions. Eur J Orthop Surg Traumatol. 2017;27:3–9. [DOI] [PubMed] [Google Scholar]
- 12.Singh HP, Taub N, Dias JJ. Management of displaced fractures of the waist of the scaphoid: meta-analyses of comparative studies. Injury. 2012;43:933–939. [DOI] [PubMed] [Google Scholar]
- 13.Uhl RL. The use of electricity in bone healing. Orthop Rev. 1989;18:1045–1050. [PubMed] [Google Scholar]
- 14.Frykman GK, Taleisnik J, Peters G, et al. Treatment of nonunited scaphoid fractures by pulsed electromagnetic field and cast. J Hand Surg Am. 1986;11:344–349. [DOI] [PubMed] [Google Scholar]
- 15.Cook JJ, Summers NJ, Cook EA. Healing in the new millennium: bone stimulators an overview of where we’ve been and where we may be heading. Clin Podiatr Med Surg. 2015;32:45–59. [DOI] [PubMed] [Google Scholar]
- 16.Forsted DL, Dalinka MK, Mitchell E, et al. Radiologic evaluation of the treatment of nonunion of fractures by electrical stimulation. Radiology. 1978;128:629–634. [DOI] [PubMed] [Google Scholar]
- 17.Huang AJ, Gemperli MP, Berghold L, et al. Health plan coverage determinants for technology-based interventions: the case of electrical bone growth stimulation. Am J Manag Care. 2004;10:957–962. [PubMed] [Google Scholar]
- 18.FDA Executive Summary. Prepared September 8-9, 2020 Meeting of the Orthopedic and Rehabilitation Devices Panel. fda.gov. 2020. Available at https://www.fda.gov/media/141850/download. Accessed October 12, 2022.
- 19.Divelbiss BJ, Adams BD. Electrical and ultrasound stimulation for scaphoid fractures. Hand Clin. 2001;17:697–701. [PubMed] [Google Scholar]
- 20.Bora FW, Osterman AL, Woodbury DF, et al. Treatment of nonunion of the scaphoid by direct current. Orthop Clin North Am. 1984;15:107–112. [PubMed] [Google Scholar]
- 21.Muhldorfer-Fodor M, Wagner M, Kottmann T, et al. Comparison of scaphoid reconstruction with a non-vascularized bone graft, with and without shock waves; preliminary results. Handchir Mikrochir Plast Chir. 2020;52:404–412. [DOI] [PubMed] [Google Scholar]
- 22.Mayr E, Rudzki MM, Rudzki M, et al. Does low intensity, pulsed ultrasound speed healing of scaphoid fractures? Handchir Mikrochir Plast Chir. 2000;32:115–122. [DOI] [PubMed] [Google Scholar]
- 23.Hannemann PF, Essers BA, Schots JP, et al. Functional outcome and cost-effectiveness of pulsed electromagnetic fields in the treatment of acute scaphoid fractures: a cost-utility analysis. BMC Musculoskelet Disord. 2015;16:84–90. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Hannemann PF, Mommers EH, Schots JP, et al. The effects of low-intensity pulsed ultrasound and pulsed electromagnetic fields bone growth stimulation in acute fractures: a systematic review and meta-analysis of randomized controlled trials. Arch Orthop Trauma Surg. 2014;134:1093–1106. [DOI] [PubMed] [Google Scholar]
- 25.Hughes MS, Anglen JO. The use of implantable bone stimulators in nonunion treatment. Orthopedics. 2010;33:1–7. [DOI] [PubMed] [Google Scholar]
- 26.Nolte PA, van der Krans A, Patka P, et al. Low intensity pulsed ultrasound treatment of nonunion. J Trauma. 2001;51:693–702. [DOI] [PubMed] [Google Scholar]
- 27.United Healthcare Commercial Medical Policy. Electrical and ultrasound bone growth stimulators. Policy #2021T0561O. Effective January 1, 2021. Available at https://www.uhcprovider.com/content/dam/provider/docs/public/policies/comm-medical-drug/electrical-ultrasound-bone-growth-stimulators.pdf.
- 28.Yin Z, Zhang J, Gong K. Cost-effectiveness of diagnostic strategies for suspected scaphoid fractures. J Orthop Trauma. 2015;29:e245–e252. [DOI] [PubMed] [Google Scholar]
- 29.Brooks S, Cicuttini FM, Lim S, et al. Cost effectiveness of adding magnetic resonance imaging to the usual management of suspected scaphoid fractures. Br J Sports Med. 2005;39:75–79. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Hansen TB, Petersen RPB, Barckman J, et al. Cost-effectiveness of MRI in managing suspected scaphoid fractures. J Hand Surg Eur. 2009;34:627–630. [DOI] [PubMed] [Google Scholar]
- 31.Paulus MC, Braunstein J, Merenstein D, et al. Variability in orthopedic surgeon treatment preferences for non-displaced scaphoid fractures: a cross-sectional survey. J Orthop. 2016;13:337–342. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Howmuchisit.org Staff. How much does a bone growth stimulator cost? 2018. https://www.howmuchisit.org/bone-growth-stimulator-cost/. Accessed December 13, 2021.
- 33.Button ML, Gharsaa O, Latouche S, et al. Economic evaluation of bone stimulation modalities: a systematic review of the literature. Indian J Orthop. 2009;43:168–174. [DOI] [PMC free article] [PubMed] [Google Scholar]