David Saper, MD

ABSTRACT

Delayed unions and nonunions of the scaphoid are most often treated by open reduction and internal fixation with bone grafting. We sought to evaluate a large consecutive series of nondisplaced or minimally displaced scaphoid nonunions and delayed unions treated by a compression screw without bone grafting by 2 fellowship trained hand surgeons. A total of 23 patients (19 males, 4 females) were identified who had fractures located at the distal third (2), the waist (18), and the proximal third (3). Of the 23 patients, 19 had a complete follow-up (mean follow-up period, 5.2 months) with evidence of radiographic union. There were no radiographic signs of arthrosis, osteonecrosis of the scaphoid, hardware-related complications, or reported revision surgeries. In conclusion, nonunions and delayed unions in nondisplaced or minimally displaced scaphoids without carpal malalignment can be successfully treated using compression screw fixation without bone grafting.

Continued to: Scaphoid nonunions or delayed unions with displacement...

Scaphoid nonunions or delayed unions with displacement, humpback deformities, or dorsal intercalated segmental instability deformities require open exposure with reduction of the fracture and autogenous bone grafting (structural or nonstructural and vascularized or nonvascularized).^1,2 However, in the absence of displacement or deformity, compression and internal fixation without bone grafting may be sufficient to achieve union.

Several reports have described the use of internal fixation alone in the management of scaphoid nonunions with both minimal and extensive bone loss.^3-7 These studies have shown that screw fixation alone affords less morbidity to the patient while allowing high rates of union.

Previous reports of internal fixation alone included limited numbers of patients for review. Therefore, we aim to review a large consecutive series of scaphoid delayed unions and nonunions without osteonecrosis or deformity managed by only internal fixation. Our hypothesis is that drilling combined with compression and rigid stabilization would allow for bony union in these cases

MATERIALS AND METHODS

After Institutional Review Board approval was obtained, a retrospective review of prospectively collected data was performed on consecutive patients with a delayed union or nonunion of the scaphoid. All injuries had failed conservative treatment of casting for at least 12 weeks and ultrasound stimulation, and were subsequently treated by compression screw fixation by 1 of 2 fellowship trained hand surgeons. The database comprised the data of patients who presented to a single, Level 1 trauma center between 2000 and 2012.  

Delayed unions and nonunions were defined as a lack of radiographic trabecular bridging and pain on clinical examination at 3 and 6 months, respectively. All fractures were nondisplaced or minimally displaced (<2 mm), and patients with carpal malalignment or humpback deformity (based on scapholunate angle on plain radiographs) were excluded. Clinical outcome measures included evidence of radiographic union, revision surgery, pain, and reported complications.

Continue to: Inclusion criteria were all patients who sustained...

Inclusion criteria were all patients who sustained a minimally displaced scaphoid fracture and were treated conservatively with casting for at least 12 weeks and ultrasound stimulation, and progressed to delayed unions or nonunions.

Patients younger than age 18 years or with radiographic evidence of arthrosis or humpback deformity were excluded. Any fracture with >2 mm of gapping on original injury radiographs was not considered as minimally displaced and was also excluded. Furthermore, patients with a previous ipsilateral scaphoid injury or hand surgery were also excluded.

Compression screw placement was recorded as being either central or eccentric based on Trumble and colleagues’⁸ criteria. Posteroanterior (PA), lateral, and scaphoid view radiographs were reviewed by the first author (DS) and the treating hand surgeon (AS). Central screw placement was substantiated if the screw was in the middle third of the proximal pole in all 3 views.

The final set of postoperative radiographs was reviewed for unions. Union was defined as bridging trabeculation with near or complete obliteration of the fracture line on PA, lateral, and scaphoid radiographic views. Computerized tomography (CT) was performed at the discretion of the treating surgeon, and its use was not required if there was near obliteration of the fracture line on the 3-view radiographs and in the absence of patient-reported pain. Patients with bone loss or sclerosis were included as long as no deformity existed.

After surgical intervention, a short-arm cast was applied for 6 weeks, followed by a wrist splint for 4 to 8 weeks depending on patient comfort.

Continue to: SURGICAL TECHNIQUE...

SURGICAL TECHNIQUE

Either a 1-cm to 2-cm transverse incision distal to Lister’s tubercle or a longitudinal incision just ulnar was utilized. The extensor pollicis longus was identified and retracted. A longitudinal or an L-shaped capsulotomy was made to identify the proximal pole of the scaphoid. With the wrist flexed, a guide wire was inserted down the central axis of the scaphoid and confirmed by fluoroscopy. The measurement was made off the guidewire and 4 to 6 mm was subtracted. The scaphoid was then drilled, and the variable pitch compression screw (Acutrak Headless Compression Screw, Acumed) was inserted. Compression and position of the screw were confirmed by fluoroscopy before closure.

RESULTS

A total of 23 patients (19 males, 4 females) with acute scaphoid fractures who were treated conservatively with casting for at least 12 weeks and ultrasound stimulation, and progressed to delayed unions or nonunions were identified in this study. The ages of the patients ranged from 19 to 50 years. Of the 23 patients, 6 were smokers. The majority of patients sustained fractures in the scaphoid waist (18 patients) (Figure 1). Two patients had distal third fractures, and 3 had proximal third fractures.

The average time from the sustained injury to the surgical intervention was 8.2 months (range, 3.1-27.6 months). There were no patients with delayed diagnoses. Three fractures were identified as delayed unions with failure of union and pain after 3 months of conservative treatment, whereas the other 20 were identified as nonunions with at least 6 months of failed conservative treatment.

Of the 23 patients, 21 were found to have centrally placed variable compression screws based on Trumble and colleagues’⁸ criteria. Of the 23 patients, 19 had a complete follow-up course with radiographs at 6 months after surgery. All of these 19 patients had evidence of radiographic union defined as bridging trabeculation with near or complete obliteration of the fracture line on PA, lateral, and scaphoid radiographic views (Figure 2). Of the 6 smokers, 5 progressed to radiographic union and 1 patient had <6 months of postoperative return visits and could not be contacted. At the final clinic visit, all of the 19 patients denied wrist pain on direct palpation over the scaphoid tubercle, and no complications were reported. There were no repeat or revision surgical interventions.

Four patients had limited follow-up with <6 months of postoperative return visits. Their final set of radiographs did not demonstrate complete bridging trabeculation. One patient who moved away from the area was lost to follow-up but was contacted. The patient stated that he had a pain-free wrist with no further surgical interventions on his scaphoid. The other 3 patients could not be contacted.

DISCUSSION

The management of scaphoid nonunions and delayed unions has dramatically evolved over the past 20 years.^1,3-8 Historically, semi-rigid stabilization using Kirschner wires and casting afforded a 77% union rate in these cases.⁹ More recently, several authors have reported that stabilization without bone grafting can predictably unite scaphoid nonunions. Treating patients with uncomplicated scaphoid nonunions and delayed unions by internal fixation alone may be all that is required to achieve union.

The definitions of a scaphoid nonunion and delayed union are complex. The exact time when a scaphoid fracture heals varies between patients.^2,5,10 However, the majority of hand surgeons believe that failure to see clear signs of healing (in waist fractures) after 3 months from the injury would suggest a failure to heal and a “delayed” union, whereas failure after 6 months from the injury and without clear signs of healing indicate a nonunion.^5,6,10,11 Any resorption at the fracture site suggests that the fracture will not heal by continued immobilization alone and will require surgery.¹⁰

Continue to: Hand surgeons have several surgical options...

Hand surgeons have several surgical options when managing scaphoid injuries. Mahmoud and Koptan⁴ used a volar approach to percutaneously deliver a headless compression screw into 27 nonunions. Postoperative CT scans demonstrated fracture union in all 27 patients, and no patient underwent revision surgery. Interestingly, 14 of their patients had extensive preoperative resorption (but no deformity) of >5 mm.

Although volar percutaneous approaches for internal fixation have been cited to provide high rates of union and high patient satisfaction in acute scaphoid fracture fixation, this study utilized a dorsal approach. Both Wozasek and Moser¹² and Haddad and Goddard¹³ reported excellent results and high union rates using a volar approach in consecutive acute scaphoid fractures. Despite these results, there are concerns that using a volar approach may damage the scaphotrapezial joint and may be prone to eccentric placement of compression screws.^8,14

Slade and colleagues³ did utilize the dorsal approach with arthroscopic assistance to deliver a compression screw into scaphoid nonunions in 15 consecutive patients without any evidence of deformity, sclerosis, or resorption. Similar to our investigation, they treated patients with both delayed unions and nonunions. CT scans were used to confirm unions in all their patients. Using a dorsal approach, Yassaee and Yang¹⁵ treated 9 consecutive patients using a compression screw without bone grafting for both delayed and nonunion scaphoid injuries. Other authors have used both volar and dorsal approaches in 12 consecutive delayed and nonunion scaphoid injuries and found that 11 of the 12 injuries progressed to unions.⁶

Although these authors and several others advocate the use of CT scans to assess unions, our investigation used bridging trabeculation obliteration of the fracture line on 3 standard radiographic views to confirm unions in addition to the absence of pain clinically.^16,17 CT scans expose the patient to increased radiation that, in our experience, does not alter the postoperative clinical course.¹⁸ If there is clear evidence of bridged callus and no pain on physical examination, a CT scan performed to reconfirm the union affords little benefit to clinical management.¹⁹

Continue to: All these previous studies have demonstrated...

All these previous studies have demonstrated excellent union rates but using a limited series of patients. We reviewed a large number of consecutive patients with scaphoid delayed unions and nonunions treated by screw fixation without bone grafting. Our hospital is a safety net institution for a large urban catchment area and had complete radiographic and clinical data for 19 of our 23 patients. One patient was contacted by telephone and he reported no pain and no revision surgical interventions.

The limitations of this study include not only its retrospective design but also its limited secondary outcome measures. However, our primary outcomes of union, pain, and complications are of utmost importance to clinicians and patients alike. Similar to other authors, we used radiographs to confirm unions. Although bridging trabeculation in radiographs has been demonstrated as soon as 1 month after the injury, there may be problems with interobserver reliability.^{4,13,15,20,21}

Patients being lost to follow-up is not uncommon in the orthopedic trauma literature and can influence results.^22,23 It is speculative to infer that the 3 patients who did not complete a follow-up course did not return because their pain had mitigated.

CONCLUSION 

Like several fractures, the lack of stability and the absence of micro-motion are believed to contribute to fibrous nonunions in scaphoid fractures.¹³ This study provides a large consecutive cohort of patients with minimally displaced scaphoid delayed unions and nonunions that were successfully treated by rigid internal fixation without bone grafting. These results confirm previous reports that bone grafting is not required to provide predictable unions for the majority of scaphoid nonunions.

This paper will be judged for the Resident Writer’s Award.

ABSTRACT

Delayed unions and nonunions of the scaphoid are most often treated by open reduction and internal fixation with bone grafting. We sought to evaluate a large consecutive series of nondisplaced or minimally displaced scaphoid nonunions and delayed unions treated by a compression screw without bone grafting by 2 fellowship trained hand surgeons. A total of 23 patients (19 males, 4 females) were identified who had fractures located at the distal third (2), the waist (18), and the proximal third (3). Of the 23 patients, 19 had a complete follow-up (mean follow-up period, 5.2 months) with evidence of radiographic union. There were no radiographic signs of arthrosis, osteonecrosis of the scaphoid, hardware-related complications, or reported revision surgeries. In conclusion, nonunions and delayed unions in nondisplaced or minimally displaced scaphoids without carpal malalignment can be successfully treated using compression screw fixation without bone grafting.

Continued to: Scaphoid nonunions or delayed unions with displacement...

Scaphoid nonunions or delayed unions with displacement, humpback deformities, or dorsal intercalated segmental instability deformities require open exposure with reduction of the fracture and autogenous bone grafting (structural or nonstructural and vascularized or nonvascularized).^1,2 However, in the absence of displacement or deformity, compression and internal fixation without bone grafting may be sufficient to achieve union.

Several reports have described the use of internal fixation alone in the management of scaphoid nonunions with both minimal and extensive bone loss.^3-7 These studies have shown that screw fixation alone affords less morbidity to the patient while allowing high rates of union.

Previous reports of internal fixation alone included limited numbers of patients for review. Therefore, we aim to review a large consecutive series of scaphoid delayed unions and nonunions without osteonecrosis or deformity managed by only internal fixation. Our hypothesis is that drilling combined with compression and rigid stabilization would allow for bony union in these cases

MATERIALS AND METHODS

After Institutional Review Board approval was obtained, a retrospective review of prospectively collected data was performed on consecutive patients with a delayed union or nonunion of the scaphoid. All injuries had failed conservative treatment of casting for at least 12 weeks and ultrasound stimulation, and were subsequently treated by compression screw fixation by 1 of 2 fellowship trained hand surgeons. The database comprised the data of patients who presented to a single, Level 1 trauma center between 2000 and 2012.  

Delayed unions and nonunions were defined as a lack of radiographic trabecular bridging and pain on clinical examination at 3 and 6 months, respectively. All fractures were nondisplaced or minimally displaced (<2 mm), and patients with carpal malalignment or humpback deformity (based on scapholunate angle on plain radiographs) were excluded. Clinical outcome measures included evidence of radiographic union, revision surgery, pain, and reported complications.

Continue to: Inclusion criteria were all patients who sustained...

Inclusion criteria were all patients who sustained a minimally displaced scaphoid fracture and were treated conservatively with casting for at least 12 weeks and ultrasound stimulation, and progressed to delayed unions or nonunions.

Patients younger than age 18 years or with radiographic evidence of arthrosis or humpback deformity were excluded. Any fracture with >2 mm of gapping on original injury radiographs was not considered as minimally displaced and was also excluded. Furthermore, patients with a previous ipsilateral scaphoid injury or hand surgery were also excluded.

Compression screw placement was recorded as being either central or eccentric based on Trumble and colleagues’⁸ criteria. Posteroanterior (PA), lateral, and scaphoid view radiographs were reviewed by the first author (DS) and the treating hand surgeon (AS). Central screw placement was substantiated if the screw was in the middle third of the proximal pole in all 3 views.

The final set of postoperative radiographs was reviewed for unions. Union was defined as bridging trabeculation with near or complete obliteration of the fracture line on PA, lateral, and scaphoid radiographic views. Computerized tomography (CT) was performed at the discretion of the treating surgeon, and its use was not required if there was near obliteration of the fracture line on the 3-view radiographs and in the absence of patient-reported pain. Patients with bone loss or sclerosis were included as long as no deformity existed.

After surgical intervention, a short-arm cast was applied for 6 weeks, followed by a wrist splint for 4 to 8 weeks depending on patient comfort.

Continue to: SURGICAL TECHNIQUE...

SURGICAL TECHNIQUE

Either a 1-cm to 2-cm transverse incision distal to Lister’s tubercle or a longitudinal incision just ulnar was utilized. The extensor pollicis longus was identified and retracted. A longitudinal or an L-shaped capsulotomy was made to identify the proximal pole of the scaphoid. With the wrist flexed, a guide wire was inserted down the central axis of the scaphoid and confirmed by fluoroscopy. The measurement was made off the guidewire and 4 to 6 mm was subtracted. The scaphoid was then drilled, and the variable pitch compression screw (Acutrak Headless Compression Screw, Acumed) was inserted. Compression and position of the screw were confirmed by fluoroscopy before closure.

RESULTS

A total of 23 patients (19 males, 4 females) with acute scaphoid fractures who were treated conservatively with casting for at least 12 weeks and ultrasound stimulation, and progressed to delayed unions or nonunions were identified in this study. The ages of the patients ranged from 19 to 50 years. Of the 23 patients, 6 were smokers. The majority of patients sustained fractures in the scaphoid waist (18 patients) (Figure 1). Two patients had distal third fractures, and 3 had proximal third fractures.

The average time from the sustained injury to the surgical intervention was 8.2 months (range, 3.1-27.6 months). There were no patients with delayed diagnoses. Three fractures were identified as delayed unions with failure of union and pain after 3 months of conservative treatment, whereas the other 20 were identified as nonunions with at least 6 months of failed conservative treatment.

Of the 23 patients, 21 were found to have centrally placed variable compression screws based on Trumble and colleagues’⁸ criteria. Of the 23 patients, 19 had a complete follow-up course with radiographs at 6 months after surgery. All of these 19 patients had evidence of radiographic union defined as bridging trabeculation with near or complete obliteration of the fracture line on PA, lateral, and scaphoid radiographic views (Figure 2). Of the 6 smokers, 5 progressed to radiographic union and 1 patient had <6 months of postoperative return visits and could not be contacted. At the final clinic visit, all of the 19 patients denied wrist pain on direct palpation over the scaphoid tubercle, and no complications were reported. There were no repeat or revision surgical interventions.

Four patients had limited follow-up with <6 months of postoperative return visits. Their final set of radiographs did not demonstrate complete bridging trabeculation. One patient who moved away from the area was lost to follow-up but was contacted. The patient stated that he had a pain-free wrist with no further surgical interventions on his scaphoid. The other 3 patients could not be contacted.

DISCUSSION

The management of scaphoid nonunions and delayed unions has dramatically evolved over the past 20 years.^1,3-8 Historically, semi-rigid stabilization using Kirschner wires and casting afforded a 77% union rate in these cases.⁹ More recently, several authors have reported that stabilization without bone grafting can predictably unite scaphoid nonunions. Treating patients with uncomplicated scaphoid nonunions and delayed unions by internal fixation alone may be all that is required to achieve union.

The definitions of a scaphoid nonunion and delayed union are complex. The exact time when a scaphoid fracture heals varies between patients.^2,5,10 However, the majority of hand surgeons believe that failure to see clear signs of healing (in waist fractures) after 3 months from the injury would suggest a failure to heal and a “delayed” union, whereas failure after 6 months from the injury and without clear signs of healing indicate a nonunion.^5,6,10,11 Any resorption at the fracture site suggests that the fracture will not heal by continued immobilization alone and will require surgery.¹⁰

Continue to: Hand surgeons have several surgical options...

Hand surgeons have several surgical options when managing scaphoid injuries. Mahmoud and Koptan⁴ used a volar approach to percutaneously deliver a headless compression screw into 27 nonunions. Postoperative CT scans demonstrated fracture union in all 27 patients, and no patient underwent revision surgery. Interestingly, 14 of their patients had extensive preoperative resorption (but no deformity) of >5 mm.

Although volar percutaneous approaches for internal fixation have been cited to provide high rates of union and high patient satisfaction in acute scaphoid fracture fixation, this study utilized a dorsal approach. Both Wozasek and Moser¹² and Haddad and Goddard¹³ reported excellent results and high union rates using a volar approach in consecutive acute scaphoid fractures. Despite these results, there are concerns that using a volar approach may damage the scaphotrapezial joint and may be prone to eccentric placement of compression screws.^8,14

Slade and colleagues³ did utilize the dorsal approach with arthroscopic assistance to deliver a compression screw into scaphoid nonunions in 15 consecutive patients without any evidence of deformity, sclerosis, or resorption. Similar to our investigation, they treated patients with both delayed unions and nonunions. CT scans were used to confirm unions in all their patients. Using a dorsal approach, Yassaee and Yang¹⁵ treated 9 consecutive patients using a compression screw without bone grafting for both delayed and nonunion scaphoid injuries. Other authors have used both volar and dorsal approaches in 12 consecutive delayed and nonunion scaphoid injuries and found that 11 of the 12 injuries progressed to unions.⁶

Although these authors and several others advocate the use of CT scans to assess unions, our investigation used bridging trabeculation obliteration of the fracture line on 3 standard radiographic views to confirm unions in addition to the absence of pain clinically.^16,17 CT scans expose the patient to increased radiation that, in our experience, does not alter the postoperative clinical course.¹⁸ If there is clear evidence of bridged callus and no pain on physical examination, a CT scan performed to reconfirm the union affords little benefit to clinical management.¹⁹

Continue to: All these previous studies have demonstrated...

All these previous studies have demonstrated excellent union rates but using a limited series of patients. We reviewed a large number of consecutive patients with scaphoid delayed unions and nonunions treated by screw fixation without bone grafting. Our hospital is a safety net institution for a large urban catchment area and had complete radiographic and clinical data for 19 of our 23 patients. One patient was contacted by telephone and he reported no pain and no revision surgical interventions.

The limitations of this study include not only its retrospective design but also its limited secondary outcome measures. However, our primary outcomes of union, pain, and complications are of utmost importance to clinicians and patients alike. Similar to other authors, we used radiographs to confirm unions. Although bridging trabeculation in radiographs has been demonstrated as soon as 1 month after the injury, there may be problems with interobserver reliability.^{4,13,15,20,21}

Patients being lost to follow-up is not uncommon in the orthopedic trauma literature and can influence results.^22,23 It is speculative to infer that the 3 patients who did not complete a follow-up course did not return because their pain had mitigated.

CONCLUSION 

Like several fractures, the lack of stability and the absence of micro-motion are believed to contribute to fibrous nonunions in scaphoid fractures.¹³ This study provides a large consecutive cohort of patients with minimally displaced scaphoid delayed unions and nonunions that were successfully treated by rigid internal fixation without bone grafting. These results confirm previous reports that bone grafting is not required to provide predictable unions for the majority of scaphoid nonunions.

This paper will be judged for the Resident Writer’s Award.

ABSTRACT

Delayed unions and nonunions of the scaphoid are most often treated by open reduction and internal fixation with bone grafting. We sought to evaluate a large consecutive series of nondisplaced or minimally displaced scaphoid nonunions and delayed unions treated by a compression screw without bone grafting by 2 fellowship trained hand surgeons. A total of 23 patients (19 males, 4 females) were identified who had fractures located at the distal third (2), the waist (18), and the proximal third (3). Of the 23 patients, 19 had a complete follow-up (mean follow-up period, 5.2 months) with evidence of radiographic union. There were no radiographic signs of arthrosis, osteonecrosis of the scaphoid, hardware-related complications, or reported revision surgeries. In conclusion, nonunions and delayed unions in nondisplaced or minimally displaced scaphoids without carpal malalignment can be successfully treated using compression screw fixation without bone grafting.

Continued to: Scaphoid nonunions or delayed unions with displacement...

Scaphoid nonunions or delayed unions with displacement, humpback deformities, or dorsal intercalated segmental instability deformities require open exposure with reduction of the fracture and autogenous bone grafting (structural or nonstructural and vascularized or nonvascularized).^1,2 However, in the absence of displacement or deformity, compression and internal fixation without bone grafting may be sufficient to achieve union.

Several reports have described the use of internal fixation alone in the management of scaphoid nonunions with both minimal and extensive bone loss.^3-7 These studies have shown that screw fixation alone affords less morbidity to the patient while allowing high rates of union.

Previous reports of internal fixation alone included limited numbers of patients for review. Therefore, we aim to review a large consecutive series of scaphoid delayed unions and nonunions without osteonecrosis or deformity managed by only internal fixation. Our hypothesis is that drilling combined with compression and rigid stabilization would allow for bony union in these cases

MATERIALS AND METHODS

After Institutional Review Board approval was obtained, a retrospective review of prospectively collected data was performed on consecutive patients with a delayed union or nonunion of the scaphoid. All injuries had failed conservative treatment of casting for at least 12 weeks and ultrasound stimulation, and were subsequently treated by compression screw fixation by 1 of 2 fellowship trained hand surgeons. The database comprised the data of patients who presented to a single, Level 1 trauma center between 2000 and 2012.  

Delayed unions and nonunions were defined as a lack of radiographic trabecular bridging and pain on clinical examination at 3 and 6 months, respectively. All fractures were nondisplaced or minimally displaced (<2 mm), and patients with carpal malalignment or humpback deformity (based on scapholunate angle on plain radiographs) were excluded. Clinical outcome measures included evidence of radiographic union, revision surgery, pain, and reported complications.

Continue to: Inclusion criteria were all patients who sustained...

Inclusion criteria were all patients who sustained a minimally displaced scaphoid fracture and were treated conservatively with casting for at least 12 weeks and ultrasound stimulation, and progressed to delayed unions or nonunions.

Patients younger than age 18 years or with radiographic evidence of arthrosis or humpback deformity were excluded. Any fracture with >2 mm of gapping on original injury radiographs was not considered as minimally displaced and was also excluded. Furthermore, patients with a previous ipsilateral scaphoid injury or hand surgery were also excluded.

Compression screw placement was recorded as being either central or eccentric based on Trumble and colleagues’⁸ criteria. Posteroanterior (PA), lateral, and scaphoid view radiographs were reviewed by the first author (DS) and the treating hand surgeon (AS). Central screw placement was substantiated if the screw was in the middle third of the proximal pole in all 3 views.

The final set of postoperative radiographs was reviewed for unions. Union was defined as bridging trabeculation with near or complete obliteration of the fracture line on PA, lateral, and scaphoid radiographic views. Computerized tomography (CT) was performed at the discretion of the treating surgeon, and its use was not required if there was near obliteration of the fracture line on the 3-view radiographs and in the absence of patient-reported pain. Patients with bone loss or sclerosis were included as long as no deformity existed.

After surgical intervention, a short-arm cast was applied for 6 weeks, followed by a wrist splint for 4 to 8 weeks depending on patient comfort.

Continue to: SURGICAL TECHNIQUE...

SURGICAL TECHNIQUE

Either a 1-cm to 2-cm transverse incision distal to Lister’s tubercle or a longitudinal incision just ulnar was utilized. The extensor pollicis longus was identified and retracted. A longitudinal or an L-shaped capsulotomy was made to identify the proximal pole of the scaphoid. With the wrist flexed, a guide wire was inserted down the central axis of the scaphoid and confirmed by fluoroscopy. The measurement was made off the guidewire and 4 to 6 mm was subtracted. The scaphoid was then drilled, and the variable pitch compression screw (Acutrak Headless Compression Screw, Acumed) was inserted. Compression and position of the screw were confirmed by fluoroscopy before closure.

RESULTS

A total of 23 patients (19 males, 4 females) with acute scaphoid fractures who were treated conservatively with casting for at least 12 weeks and ultrasound stimulation, and progressed to delayed unions or nonunions were identified in this study. The ages of the patients ranged from 19 to 50 years. Of the 23 patients, 6 were smokers. The majority of patients sustained fractures in the scaphoid waist (18 patients) (Figure 1). Two patients had distal third fractures, and 3 had proximal third fractures.

The average time from the sustained injury to the surgical intervention was 8.2 months (range, 3.1-27.6 months). There were no patients with delayed diagnoses. Three fractures were identified as delayed unions with failure of union and pain after 3 months of conservative treatment, whereas the other 20 were identified as nonunions with at least 6 months of failed conservative treatment.

Of the 23 patients, 21 were found to have centrally placed variable compression screws based on Trumble and colleagues’⁸ criteria. Of the 23 patients, 19 had a complete follow-up course with radiographs at 6 months after surgery. All of these 19 patients had evidence of radiographic union defined as bridging trabeculation with near or complete obliteration of the fracture line on PA, lateral, and scaphoid radiographic views (Figure 2). Of the 6 smokers, 5 progressed to radiographic union and 1 patient had <6 months of postoperative return visits and could not be contacted. At the final clinic visit, all of the 19 patients denied wrist pain on direct palpation over the scaphoid tubercle, and no complications were reported. There were no repeat or revision surgical interventions.

Four patients had limited follow-up with <6 months of postoperative return visits. Their final set of radiographs did not demonstrate complete bridging trabeculation. One patient who moved away from the area was lost to follow-up but was contacted. The patient stated that he had a pain-free wrist with no further surgical interventions on his scaphoid. The other 3 patients could not be contacted.

DISCUSSION

The management of scaphoid nonunions and delayed unions has dramatically evolved over the past 20 years.^1,3-8 Historically, semi-rigid stabilization using Kirschner wires and casting afforded a 77% union rate in these cases.⁹ More recently, several authors have reported that stabilization without bone grafting can predictably unite scaphoid nonunions. Treating patients with uncomplicated scaphoid nonunions and delayed unions by internal fixation alone may be all that is required to achieve union.

The definitions of a scaphoid nonunion and delayed union are complex. The exact time when a scaphoid fracture heals varies between patients.^2,5,10 However, the majority of hand surgeons believe that failure to see clear signs of healing (in waist fractures) after 3 months from the injury would suggest a failure to heal and a “delayed” union, whereas failure after 6 months from the injury and without clear signs of healing indicate a nonunion.^5,6,10,11 Any resorption at the fracture site suggests that the fracture will not heal by continued immobilization alone and will require surgery.¹⁰

Continue to: Hand surgeons have several surgical options...

Hand surgeons have several surgical options when managing scaphoid injuries. Mahmoud and Koptan⁴ used a volar approach to percutaneously deliver a headless compression screw into 27 nonunions. Postoperative CT scans demonstrated fracture union in all 27 patients, and no patient underwent revision surgery. Interestingly, 14 of their patients had extensive preoperative resorption (but no deformity) of >5 mm.

Although volar percutaneous approaches for internal fixation have been cited to provide high rates of union and high patient satisfaction in acute scaphoid fracture fixation, this study utilized a dorsal approach. Both Wozasek and Moser¹² and Haddad and Goddard¹³ reported excellent results and high union rates using a volar approach in consecutive acute scaphoid fractures. Despite these results, there are concerns that using a volar approach may damage the scaphotrapezial joint and may be prone to eccentric placement of compression screws.^8,14

Slade and colleagues³ did utilize the dorsal approach with arthroscopic assistance to deliver a compression screw into scaphoid nonunions in 15 consecutive patients without any evidence of deformity, sclerosis, or resorption. Similar to our investigation, they treated patients with both delayed unions and nonunions. CT scans were used to confirm unions in all their patients. Using a dorsal approach, Yassaee and Yang¹⁵ treated 9 consecutive patients using a compression screw without bone grafting for both delayed and nonunion scaphoid injuries. Other authors have used both volar and dorsal approaches in 12 consecutive delayed and nonunion scaphoid injuries and found that 11 of the 12 injuries progressed to unions.⁶

Although these authors and several others advocate the use of CT scans to assess unions, our investigation used bridging trabeculation obliteration of the fracture line on 3 standard radiographic views to confirm unions in addition to the absence of pain clinically.^16,17 CT scans expose the patient to increased radiation that, in our experience, does not alter the postoperative clinical course.¹⁸ If there is clear evidence of bridged callus and no pain on physical examination, a CT scan performed to reconfirm the union affords little benefit to clinical management.¹⁹

Continue to: All these previous studies have demonstrated...

All these previous studies have demonstrated excellent union rates but using a limited series of patients. We reviewed a large number of consecutive patients with scaphoid delayed unions and nonunions treated by screw fixation without bone grafting. Our hospital is a safety net institution for a large urban catchment area and had complete radiographic and clinical data for 19 of our 23 patients. One patient was contacted by telephone and he reported no pain and no revision surgical interventions.

The limitations of this study include not only its retrospective design but also its limited secondary outcome measures. However, our primary outcomes of union, pain, and complications are of utmost importance to clinicians and patients alike. Similar to other authors, we used radiographs to confirm unions. Although bridging trabeculation in radiographs has been demonstrated as soon as 1 month after the injury, there may be problems with interobserver reliability.^{4,13,15,20,21}

Patients being lost to follow-up is not uncommon in the orthopedic trauma literature and can influence results.^22,23 It is speculative to infer that the 3 patients who did not complete a follow-up course did not return because their pain had mitigated.

CONCLUSION 

Like several fractures, the lack of stability and the absence of micro-motion are believed to contribute to fibrous nonunions in scaphoid fractures.¹³ This study provides a large consecutive cohort of patients with minimally displaced scaphoid delayed unions and nonunions that were successfully treated by rigid internal fixation without bone grafting. These results confirm previous reports that bone grafting is not required to provide predictable unions for the majority of scaphoid nonunions.

This paper will be judged for the Resident Writer’s Award.

ABSTRACT

Diaphyseal fractures of the distal humerus have a high rate of union when treated with a functional brace or an above-elbow cast (AEC). This study compares alignment of the humerus and motion of the elbow after functional brace or AEC treatment.

One-hundred and five consecutive patients with a closed, extra-articular fracture of the distal humeral diaphysis were identified in the orthopedic trauma databases of 3 hospitals between 2003 and 2012. Seventy-five patients with a follow-up of at least 6 months or with radiographic and clinical evidence of fracture union were included (51 treated with functional bracing and 24 treated with an AEC).

All of the fractures healed. The average arc of elbow flexion was 130° ± 9° in braced patients vs 127° ± 12° in casted patients. Four patients (8%) in the bracing group and 4 (17%) in the casting group lost >20° of elbow motion. The average varus angulation on radiographs was 17° ± 8° in braced and 13° ± 8° in casted patients, while the average posterior angulation was 9° ± 6° vs 7° ± 7°, respectively.

Closed extra-articular distal diaphyseal humerus fractures heal with both bracing and casting and there are no differences in average elbow motion or radiographic alignment.

Nonoperative treatment of closed fractures of the humeral shaft (AO/OTA [Arbeitsgemeinschaft für Osteosynthesefragen/Orthopaedic Trauma Association] type 12) with a functional brace or above-elbow cast (AEC) is associated with a high union rate, good motion, and good function. Advocates of casting believe that a brace cannot control fracture alignment as well as a cast that allows for immobilization and molding. Advocates of brace treatment are concerned that immobilization in a cast will cause elbow stiffness.^1-11

Continue to: In our differing institutions...

MATERIALS AND METHODS

Between 2003 and 2012, consecutive adult patients treated for a nonpathological fracture of the diaphysis of the distal humerus at the orthopedic trauma service of 3 level 1 academic trauma centers were identified from prospectively collected trauma injury databases. Patients with vascular injury, ipsilateral upper extremity fracture, and periprosthetic fractures were excluded. The attending orthopedic surgeon chose the treatment method and evaluated the range of motion (ROM) of the elbow and radiographic union at the final ambulatory visit. We included patients followed to clinical and radiographic union with a minimum of 6 months of follow-up. We also included patients with <6 months’ follow-up who demonstrated union and had elbow ROM within 10° of the uninjured arm.

We identified 105 consecutive adult patients with a closed nonpathological extra-articular distal humeral shaft fracture (fracture of the distal humeral shaft with an AO/OTA type-12.A, 12.B, or 12.C pattern) treated with an AEC or a brace in our databases.¹² Two patients in the brace group chose surgery to improve alignment within 3 weeks of injury and were excluded from the analysis. Twenty-eight patients had inadequate follow-up.

A total of 75 patients were included in the study. At the first and second institutions, 51 patients were treated with functional bracing with an average follow-up of 7 months. At the third institution, 24 patients were treated with an AEC with an average follow-up of 4 months. Seventeen out of 24 patients in the long arm casting group and 19 out of 51 patients in the bracing group, who were included since they had <6 months of follow-up, demonstrated union and had elbow ROM within 10° of the uninjured arm. Differing methods of closed immobilization were the result of differing treatment algorithms at each institution.

The patients who were treated with a functional brace averaged 34 years of age (range, 18-90 years) and included 27 men and 24 women. The brace was removed at an average of 11.5 weeks (range, 8-18 weeks) after initial injury. Six patients had an injury-associated radial nerve palsy, all of which fully recovered within an average of 4 months (range, 0.5-7 months). Sixteen patients were injured due to a fall from standing height, 2 due to a fall from a greater height than standing, 16 in a motor-vehicle accident, 15 during a sport activity, and 2 were not specifically documented.

Continue to: Four patients had concomitant...

The patients who were treated with an AEC had an average age of 32 years (range,18-82 years) and included 14 men and 10 women. The cast was removed at an average of 4.2 weeks (range, 3-7 weeks) after the initial injury. Two patients had an injury-associated radial nerve palsy, both of which fully recovered. Five patients were injured due to a fall from standing height, 1 due to a fall from a height greater than standing, 7 during a motor-vehicle accident, 5 during a sport activity, and 6 were not documented. Two patients sustained concomitant injuries: one patient sustained a tibia-fibula fracture, and another patient sustained facial trauma.

The 2 groups were comparable in age and gender, as well as the injury mechanism (Table).

Table. Patient Demographics and Outcome Data

FUNCTIONAL BRACING TECHNIQUE

Upon presentation after injury, patients were immobilized in a coaptation splint (Figure 1A). Within 10 days, the arm was placed in a pre-manufactured polyethylene functional brace (Corflex) and the arm was supported with a simple sling. Patients were allowed to use the hand for light tasks and move the elbow, but most patients were not capable of active elbow flexion exercises until early healing was established 4 to 6 weeks after injury. Shoulder motion was discouraged until radiographic union. Patients started active, self-assisted elbow and shoulder stretching exercises, and weaned from the brace once radiographic union was confirmed between 6 and 10 weeks after injury (Figures 1B, 1C).

ABOVE-ELBOW CASE

Patients were also initially immobilized in a coaptation splint upon initial presentation. Within 7 days, an above-elbow fiberglass cast with neutral forearm rotation and 90° of elbow flexion was applied with a supracondylar mold, followed by radiographic imaging (Figure 2A). With the fractured arm dependent, a valgus mold was applied as the material hardened in order to align the fracture site and limit varus angulation.

Continue to: There were no shoulder...

STATISTICAL ANALYSIS

Alignment of the humerus (including varus-valgus alignment and apex anterior-posterior alignment) was measured on anteroposterior and lateral radiographs as the angle between lines bisecting the humeral diaphysis proximal and distal to the fracture. The normality of the data was tested using the Kolmogorov-Smirnov test. To statistically compare continuous variables with a normal distribution, t-tests were used; otherwise the Wilcoxon t-test was applied. The Pearson’s Chi-Square test was used to statistically compare dichotomous variables, except when expected cell frequency was <5, in which case the Fisher exact test was used. The level of significance was set at P < .05.

RESULTS

RANGE OF MOTION AND RADIOGRAPHIC ALIGNMENT

The average range of elbow motion was 130° ± 9° after brace treatment and 127° ± 12° after cast treatment (P = .26). Four patients (8%) treated with a brace and 3 (12%) treated with a cast lost >20° of elbow motion.

All the fractures healed. The average varus angulation on the anteroposterior radiograph was 17° (range, 2°-26°) in braced patients and 13 (range, 5°-31°) in casted patients (P = .11). The average posterior angulation on the lateral radiograph was 9° (range, 0°-28°) in braced patients vs 7° (range, 2°-33°) in casted patients (P = .54).

Continue to: Two weeks after initiating brace...

COMPLICATIONS

Two weeks after initiating brace treatment, an obese patient suffered a rash with desquamation that necessitated discontinuation of the brace. However, the skin and fracture ultimately healed with a coaptation splint and sling support without additional complications. In the casting cohort, 2 patients returned to the emergency department after AEC placement because of swelling of the hand and pain in the cast. Both casts were removed and reapplied.

DISCUSSION

Fractures of the distal third of the humeral diaphysis heal without surgery. Fracture angulation and elbow stiffness are the concerns that lead to variations in nonoperative treatment.^1-3 Advocates of casting believe they can get better alignment without losing elbow motion, and advocates of bracing feel that the brace is less cumbersome.^1-3,5-8 We compared these treatments retrospectively and found them comparable.

This study should be considered in light of its limitations. Many patients were lost to follow-up in our urban trauma centers. We do not know if these patients did better, worse, or the same as the patients we were able to evaluate, but our opinion is that patients having problems were more likely to return. The evaluation time was relatively short, but motion can only improve in the longer-term. Two patients that were initially braced chose surgery, probably because either they or their surgeon were nervous about the radiographic appearance of the fracture. In our opinion, continued nonoperative treatment of these patients would not affect the findings.

Cast treatment of distal diaphyseal humerus fractures does not cause permanent elbow stiffness. This is confirmed by our results; as casted patients did not lose final ROM compared to the bracing cohort. These injuries are extra-articular and casted patients are transitioned to bracing once humeri have significant union demonstrated by the arm moving as a unit. To our knowledge, there is no other study that has evaluated casting for these fractures, but it may be that evidence of permanent stiffness with nonoperative treatment of distal metaphyseal fractures of the humerus [AO/OTA type 13] is misapplied to distal humeral shaft fractures [AO/OTA type 12].^3,9,10,12 For brace treatment, Sarmiento and colleagues⁹ showed no significant elbow stiffness in a consecutive cohort of 69 patients, while Jawa and colleagues⁵ showed no increased elbow stiffness compared to plate fixation. Given the accumulated data,^3,5,6,8,13 advocates of operative treatment for distal third diaphyseal humerus fractures¹² can no longer site elbow stiffness as a disadvantage of nonoperative treatment, whether with cast or brace.

As shown in this study, patients that choose nonoperative treatment can expect their fracture to heal with an average of approximately 15° of varus angulation, as well as 2 others evaluating brace treatment.^5,9 Some will heal with as much as 30° of varus angulation.^5,9 The arm may look a little different, particularly in thin patients, but there is no evidence that this angulation affects function. The risks, discomforts, and inconveniences of surgery can be balanced with the ability of surgery to improve alignment and allow elbow motion a few weeks earlier. The aesthetics of the scar after surgery may not be better than the deformity after nonoperative treatment. Patients should be involved in these decisions.

Continue to: No cost comparison...

There is a passive implication that operative treatment of distal third diaphyseal humerus fractures affords better results and union for patients in the discussed literature. Our results demonstrate that the distal diaphyseal humerus has a natural anatomic and biologic propensity to heal with closed immobilization. Patients should be made aware that while operative treatments exist for this fracture pattern, nonoperative treatment modalities have proven to be efficacious using a variety of immobilization methods. Thus, patients that prefer nonoperative treatment of a distal third diaphyseal humerus fracture can choose between a cast or a brace with confidence of the efficacy of the nonoperative treatment.

ABSTRACT

Diaphyseal fractures of the distal humerus have a high rate of union when treated with a functional brace or an above-elbow cast (AEC). This study compares alignment of the humerus and motion of the elbow after functional brace or AEC treatment.

One-hundred and five consecutive patients with a closed, extra-articular fracture of the distal humeral diaphysis were identified in the orthopedic trauma databases of 3 hospitals between 2003 and 2012. Seventy-five patients with a follow-up of at least 6 months or with radiographic and clinical evidence of fracture union were included (51 treated with functional bracing and 24 treated with an AEC).

All of the fractures healed. The average arc of elbow flexion was 130° ± 9° in braced patients vs 127° ± 12° in casted patients. Four patients (8%) in the bracing group and 4 (17%) in the casting group lost >20° of elbow motion. The average varus angulation on radiographs was 17° ± 8° in braced and 13° ± 8° in casted patients, while the average posterior angulation was 9° ± 6° vs 7° ± 7°, respectively.

Closed extra-articular distal diaphyseal humerus fractures heal with both bracing and casting and there are no differences in average elbow motion or radiographic alignment.

Nonoperative treatment of closed fractures of the humeral shaft (AO/OTA [Arbeitsgemeinschaft für Osteosynthesefragen/Orthopaedic Trauma Association] type 12) with a functional brace or above-elbow cast (AEC) is associated with a high union rate, good motion, and good function. Advocates of casting believe that a brace cannot control fracture alignment as well as a cast that allows for immobilization and molding. Advocates of brace treatment are concerned that immobilization in a cast will cause elbow stiffness.^1-11

Continue to: In our differing institutions...

MATERIALS AND METHODS

Between 2003 and 2012, consecutive adult patients treated for a nonpathological fracture of the diaphysis of the distal humerus at the orthopedic trauma service of 3 level 1 academic trauma centers were identified from prospectively collected trauma injury databases. Patients with vascular injury, ipsilateral upper extremity fracture, and periprosthetic fractures were excluded. The attending orthopedic surgeon chose the treatment method and evaluated the range of motion (ROM) of the elbow and radiographic union at the final ambulatory visit. We included patients followed to clinical and radiographic union with a minimum of 6 months of follow-up. We also included patients with <6 months’ follow-up who demonstrated union and had elbow ROM within 10° of the uninjured arm.

We identified 105 consecutive adult patients with a closed nonpathological extra-articular distal humeral shaft fracture (fracture of the distal humeral shaft with an AO/OTA type-12.A, 12.B, or 12.C pattern) treated with an AEC or a brace in our databases.¹² Two patients in the brace group chose surgery to improve alignment within 3 weeks of injury and were excluded from the analysis. Twenty-eight patients had inadequate follow-up.

A total of 75 patients were included in the study. At the first and second institutions, 51 patients were treated with functional bracing with an average follow-up of 7 months. At the third institution, 24 patients were treated with an AEC with an average follow-up of 4 months. Seventeen out of 24 patients in the long arm casting group and 19 out of 51 patients in the bracing group, who were included since they had <6 months of follow-up, demonstrated union and had elbow ROM within 10° of the uninjured arm. Differing methods of closed immobilization were the result of differing treatment algorithms at each institution.

The patients who were treated with a functional brace averaged 34 years of age (range, 18-90 years) and included 27 men and 24 women. The brace was removed at an average of 11.5 weeks (range, 8-18 weeks) after initial injury. Six patients had an injury-associated radial nerve palsy, all of which fully recovered within an average of 4 months (range, 0.5-7 months). Sixteen patients were injured due to a fall from standing height, 2 due to a fall from a greater height than standing, 16 in a motor-vehicle accident, 15 during a sport activity, and 2 were not specifically documented.

Continue to: Four patients had concomitant...

The patients who were treated with an AEC had an average age of 32 years (range,18-82 years) and included 14 men and 10 women. The cast was removed at an average of 4.2 weeks (range, 3-7 weeks) after the initial injury. Two patients had an injury-associated radial nerve palsy, both of which fully recovered. Five patients were injured due to a fall from standing height, 1 due to a fall from a height greater than standing, 7 during a motor-vehicle accident, 5 during a sport activity, and 6 were not documented. Two patients sustained concomitant injuries: one patient sustained a tibia-fibula fracture, and another patient sustained facial trauma.

The 2 groups were comparable in age and gender, as well as the injury mechanism (Table).

Table. Patient Demographics and Outcome Data

FUNCTIONAL BRACING TECHNIQUE

Upon presentation after injury, patients were immobilized in a coaptation splint (Figure 1A). Within 10 days, the arm was placed in a pre-manufactured polyethylene functional brace (Corflex) and the arm was supported with a simple sling. Patients were allowed to use the hand for light tasks and move the elbow, but most patients were not capable of active elbow flexion exercises until early healing was established 4 to 6 weeks after injury. Shoulder motion was discouraged until radiographic union. Patients started active, self-assisted elbow and shoulder stretching exercises, and weaned from the brace once radiographic union was confirmed between 6 and 10 weeks after injury (Figures 1B, 1C).

ABOVE-ELBOW CASE

Patients were also initially immobilized in a coaptation splint upon initial presentation. Within 7 days, an above-elbow fiberglass cast with neutral forearm rotation and 90° of elbow flexion was applied with a supracondylar mold, followed by radiographic imaging (Figure 2A). With the fractured arm dependent, a valgus mold was applied as the material hardened in order to align the fracture site and limit varus angulation.

Continue to: There were no shoulder...

STATISTICAL ANALYSIS

Alignment of the humerus (including varus-valgus alignment and apex anterior-posterior alignment) was measured on anteroposterior and lateral radiographs as the angle between lines bisecting the humeral diaphysis proximal and distal to the fracture. The normality of the data was tested using the Kolmogorov-Smirnov test. To statistically compare continuous variables with a normal distribution, t-tests were used; otherwise the Wilcoxon t-test was applied. The Pearson’s Chi-Square test was used to statistically compare dichotomous variables, except when expected cell frequency was <5, in which case the Fisher exact test was used. The level of significance was set at P < .05.

RESULTS

RANGE OF MOTION AND RADIOGRAPHIC ALIGNMENT

The average range of elbow motion was 130° ± 9° after brace treatment and 127° ± 12° after cast treatment (P = .26). Four patients (8%) treated with a brace and 3 (12%) treated with a cast lost >20° of elbow motion.

All the fractures healed. The average varus angulation on the anteroposterior radiograph was 17° (range, 2°-26°) in braced patients and 13 (range, 5°-31°) in casted patients (P = .11). The average posterior angulation on the lateral radiograph was 9° (range, 0°-28°) in braced patients vs 7° (range, 2°-33°) in casted patients (P = .54).

Continue to: Two weeks after initiating brace...

COMPLICATIONS

Two weeks after initiating brace treatment, an obese patient suffered a rash with desquamation that necessitated discontinuation of the brace. However, the skin and fracture ultimately healed with a coaptation splint and sling support without additional complications. In the casting cohort, 2 patients returned to the emergency department after AEC placement because of swelling of the hand and pain in the cast. Both casts were removed and reapplied.

DISCUSSION

Fractures of the distal third of the humeral diaphysis heal without surgery. Fracture angulation and elbow stiffness are the concerns that lead to variations in nonoperative treatment.^1-3 Advocates of casting believe they can get better alignment without losing elbow motion, and advocates of bracing feel that the brace is less cumbersome.^1-3,5-8 We compared these treatments retrospectively and found them comparable.

This study should be considered in light of its limitations. Many patients were lost to follow-up in our urban trauma centers. We do not know if these patients did better, worse, or the same as the patients we were able to evaluate, but our opinion is that patients having problems were more likely to return. The evaluation time was relatively short, but motion can only improve in the longer-term. Two patients that were initially braced chose surgery, probably because either they or their surgeon were nervous about the radiographic appearance of the fracture. In our opinion, continued nonoperative treatment of these patients would not affect the findings.

Cast treatment of distal diaphyseal humerus fractures does not cause permanent elbow stiffness. This is confirmed by our results; as casted patients did not lose final ROM compared to the bracing cohort. These injuries are extra-articular and casted patients are transitioned to bracing once humeri have significant union demonstrated by the arm moving as a unit. To our knowledge, there is no other study that has evaluated casting for these fractures, but it may be that evidence of permanent stiffness with nonoperative treatment of distal metaphyseal fractures of the humerus [AO/OTA type 13] is misapplied to distal humeral shaft fractures [AO/OTA type 12].^3,9,10,12 For brace treatment, Sarmiento and colleagues⁹ showed no significant elbow stiffness in a consecutive cohort of 69 patients, while Jawa and colleagues⁵ showed no increased elbow stiffness compared to plate fixation. Given the accumulated data,^3,5,6,8,13 advocates of operative treatment for distal third diaphyseal humerus fractures¹² can no longer site elbow stiffness as a disadvantage of nonoperative treatment, whether with cast or brace.

As shown in this study, patients that choose nonoperative treatment can expect their fracture to heal with an average of approximately 15° of varus angulation, as well as 2 others evaluating brace treatment.^5,9 Some will heal with as much as 30° of varus angulation.^5,9 The arm may look a little different, particularly in thin patients, but there is no evidence that this angulation affects function. The risks, discomforts, and inconveniences of surgery can be balanced with the ability of surgery to improve alignment and allow elbow motion a few weeks earlier. The aesthetics of the scar after surgery may not be better than the deformity after nonoperative treatment. Patients should be involved in these decisions.

Continue to: No cost comparison...

There is a passive implication that operative treatment of distal third diaphyseal humerus fractures affords better results and union for patients in the discussed literature. Our results demonstrate that the distal diaphyseal humerus has a natural anatomic and biologic propensity to heal with closed immobilization. Patients should be made aware that while operative treatments exist for this fracture pattern, nonoperative treatment modalities have proven to be efficacious using a variety of immobilization methods. Thus, patients that prefer nonoperative treatment of a distal third diaphyseal humerus fracture can choose between a cast or a brace with confidence of the efficacy of the nonoperative treatment.

ABSTRACT

Diaphyseal fractures of the distal humerus have a high rate of union when treated with a functional brace or an above-elbow cast (AEC). This study compares alignment of the humerus and motion of the elbow after functional brace or AEC treatment.

One-hundred and five consecutive patients with a closed, extra-articular fracture of the distal humeral diaphysis were identified in the orthopedic trauma databases of 3 hospitals between 2003 and 2012. Seventy-five patients with a follow-up of at least 6 months or with radiographic and clinical evidence of fracture union were included (51 treated with functional bracing and 24 treated with an AEC).

All of the fractures healed. The average arc of elbow flexion was 130° ± 9° in braced patients vs 127° ± 12° in casted patients. Four patients (8%) in the bracing group and 4 (17%) in the casting group lost >20° of elbow motion. The average varus angulation on radiographs was 17° ± 8° in braced and 13° ± 8° in casted patients, while the average posterior angulation was 9° ± 6° vs 7° ± 7°, respectively.

Closed extra-articular distal diaphyseal humerus fractures heal with both bracing and casting and there are no differences in average elbow motion or radiographic alignment.

Nonoperative treatment of closed fractures of the humeral shaft (AO/OTA [Arbeitsgemeinschaft für Osteosynthesefragen/Orthopaedic Trauma Association] type 12) with a functional brace or above-elbow cast (AEC) is associated with a high union rate, good motion, and good function. Advocates of casting believe that a brace cannot control fracture alignment as well as a cast that allows for immobilization and molding. Advocates of brace treatment are concerned that immobilization in a cast will cause elbow stiffness.^1-11

Continue to: In our differing institutions...

MATERIALS AND METHODS

Between 2003 and 2012, consecutive adult patients treated for a nonpathological fracture of the diaphysis of the distal humerus at the orthopedic trauma service of 3 level 1 academic trauma centers were identified from prospectively collected trauma injury databases. Patients with vascular injury, ipsilateral upper extremity fracture, and periprosthetic fractures were excluded. The attending orthopedic surgeon chose the treatment method and evaluated the range of motion (ROM) of the elbow and radiographic union at the final ambulatory visit. We included patients followed to clinical and radiographic union with a minimum of 6 months of follow-up. We also included patients with <6 months’ follow-up who demonstrated union and had elbow ROM within 10° of the uninjured arm.

We identified 105 consecutive adult patients with a closed nonpathological extra-articular distal humeral shaft fracture (fracture of the distal humeral shaft with an AO/OTA type-12.A, 12.B, or 12.C pattern) treated with an AEC or a brace in our databases.¹² Two patients in the brace group chose surgery to improve alignment within 3 weeks of injury and were excluded from the analysis. Twenty-eight patients had inadequate follow-up.

A total of 75 patients were included in the study. At the first and second institutions, 51 patients were treated with functional bracing with an average follow-up of 7 months. At the third institution, 24 patients were treated with an AEC with an average follow-up of 4 months. Seventeen out of 24 patients in the long arm casting group and 19 out of 51 patients in the bracing group, who were included since they had <6 months of follow-up, demonstrated union and had elbow ROM within 10° of the uninjured arm. Differing methods of closed immobilization were the result of differing treatment algorithms at each institution.

The patients who were treated with a functional brace averaged 34 years of age (range, 18-90 years) and included 27 men and 24 women. The brace was removed at an average of 11.5 weeks (range, 8-18 weeks) after initial injury. Six patients had an injury-associated radial nerve palsy, all of which fully recovered within an average of 4 months (range, 0.5-7 months). Sixteen patients were injured due to a fall from standing height, 2 due to a fall from a greater height than standing, 16 in a motor-vehicle accident, 15 during a sport activity, and 2 were not specifically documented.

Continue to: Four patients had concomitant...

The patients who were treated with an AEC had an average age of 32 years (range,18-82 years) and included 14 men and 10 women. The cast was removed at an average of 4.2 weeks (range, 3-7 weeks) after the initial injury. Two patients had an injury-associated radial nerve palsy, both of which fully recovered. Five patients were injured due to a fall from standing height, 1 due to a fall from a height greater than standing, 7 during a motor-vehicle accident, 5 during a sport activity, and 6 were not documented. Two patients sustained concomitant injuries: one patient sustained a tibia-fibula fracture, and another patient sustained facial trauma.

The 2 groups were comparable in age and gender, as well as the injury mechanism (Table).

Table. Patient Demographics and Outcome Data

FUNCTIONAL BRACING TECHNIQUE

Upon presentation after injury, patients were immobilized in a coaptation splint (Figure 1A). Within 10 days, the arm was placed in a pre-manufactured polyethylene functional brace (Corflex) and the arm was supported with a simple sling. Patients were allowed to use the hand for light tasks and move the elbow, but most patients were not capable of active elbow flexion exercises until early healing was established 4 to 6 weeks after injury. Shoulder motion was discouraged until radiographic union. Patients started active, self-assisted elbow and shoulder stretching exercises, and weaned from the brace once radiographic union was confirmed between 6 and 10 weeks after injury (Figures 1B, 1C).

ABOVE-ELBOW CASE

Patients were also initially immobilized in a coaptation splint upon initial presentation. Within 7 days, an above-elbow fiberglass cast with neutral forearm rotation and 90° of elbow flexion was applied with a supracondylar mold, followed by radiographic imaging (Figure 2A). With the fractured arm dependent, a valgus mold was applied as the material hardened in order to align the fracture site and limit varus angulation.

Continue to: There were no shoulder...

STATISTICAL ANALYSIS

Alignment of the humerus (including varus-valgus alignment and apex anterior-posterior alignment) was measured on anteroposterior and lateral radiographs as the angle between lines bisecting the humeral diaphysis proximal and distal to the fracture. The normality of the data was tested using the Kolmogorov-Smirnov test. To statistically compare continuous variables with a normal distribution, t-tests were used; otherwise the Wilcoxon t-test was applied. The Pearson’s Chi-Square test was used to statistically compare dichotomous variables, except when expected cell frequency was <5, in which case the Fisher exact test was used. The level of significance was set at P < .05.

RESULTS

RANGE OF MOTION AND RADIOGRAPHIC ALIGNMENT

The average range of elbow motion was 130° ± 9° after brace treatment and 127° ± 12° after cast treatment (P = .26). Four patients (8%) treated with a brace and 3 (12%) treated with a cast lost >20° of elbow motion.

All the fractures healed. The average varus angulation on the anteroposterior radiograph was 17° (range, 2°-26°) in braced patients and 13 (range, 5°-31°) in casted patients (P = .11). The average posterior angulation on the lateral radiograph was 9° (range, 0°-28°) in braced patients vs 7° (range, 2°-33°) in casted patients (P = .54).

Continue to: Two weeks after initiating brace...

COMPLICATIONS

Two weeks after initiating brace treatment, an obese patient suffered a rash with desquamation that necessitated discontinuation of the brace. However, the skin and fracture ultimately healed with a coaptation splint and sling support without additional complications. In the casting cohort, 2 patients returned to the emergency department after AEC placement because of swelling of the hand and pain in the cast. Both casts were removed and reapplied.

DISCUSSION

Fractures of the distal third of the humeral diaphysis heal without surgery. Fracture angulation and elbow stiffness are the concerns that lead to variations in nonoperative treatment.^1-3 Advocates of casting believe they can get better alignment without losing elbow motion, and advocates of bracing feel that the brace is less cumbersome.^1-3,5-8 We compared these treatments retrospectively and found them comparable.

This study should be considered in light of its limitations. Many patients were lost to follow-up in our urban trauma centers. We do not know if these patients did better, worse, or the same as the patients we were able to evaluate, but our opinion is that patients having problems were more likely to return. The evaluation time was relatively short, but motion can only improve in the longer-term. Two patients that were initially braced chose surgery, probably because either they or their surgeon were nervous about the radiographic appearance of the fracture. In our opinion, continued nonoperative treatment of these patients would not affect the findings.

Cast treatment of distal diaphyseal humerus fractures does not cause permanent elbow stiffness. This is confirmed by our results; as casted patients did not lose final ROM compared to the bracing cohort. These injuries are extra-articular and casted patients are transitioned to bracing once humeri have significant union demonstrated by the arm moving as a unit. To our knowledge, there is no other study that has evaluated casting for these fractures, but it may be that evidence of permanent stiffness with nonoperative treatment of distal metaphyseal fractures of the humerus [AO/OTA type 13] is misapplied to distal humeral shaft fractures [AO/OTA type 12].^3,9,10,12 For brace treatment, Sarmiento and colleagues⁹ showed no significant elbow stiffness in a consecutive cohort of 69 patients, while Jawa and colleagues⁵ showed no increased elbow stiffness compared to plate fixation. Given the accumulated data,^3,5,6,8,13 advocates of operative treatment for distal third diaphyseal humerus fractures¹² can no longer site elbow stiffness as a disadvantage of nonoperative treatment, whether with cast or brace.

As shown in this study, patients that choose nonoperative treatment can expect their fracture to heal with an average of approximately 15° of varus angulation, as well as 2 others evaluating brace treatment.^5,9 Some will heal with as much as 30° of varus angulation.^5,9 The arm may look a little different, particularly in thin patients, but there is no evidence that this angulation affects function. The risks, discomforts, and inconveniences of surgery can be balanced with the ability of surgery to improve alignment and allow elbow motion a few weeks earlier. The aesthetics of the scar after surgery may not be better than the deformity after nonoperative treatment. Patients should be involved in these decisions.

Continue to: No cost comparison...

There is a passive implication that operative treatment of distal third diaphyseal humerus fractures affords better results and union for patients in the discussed literature. Our results demonstrate that the distal diaphyseal humerus has a natural anatomic and biologic propensity to heal with closed immobilization. Patients should be made aware that while operative treatments exist for this fracture pattern, nonoperative treatment modalities have proven to be efficacious using a variety of immobilization methods. Thus, patients that prefer nonoperative treatment of a distal third diaphyseal humerus fracture can choose between a cast or a brace with confidence of the efficacy of the nonoperative treatment.