ORIGINAL ARTICLE
Circumferential Casting of Distal Radius Fractures Justin Drager, MD, HBSc,* Alberto Carli, MD, MSc,* Bogdan A. Matache, BSc,† Gregory K. Berry, MD, FRCSC,* Rudy Reindl, MD, FRCSC,* and Edward J. Harvey, MD, MSc, FRCSC*
Objectives: To determine the prevalence and predictive factors for the early cast alteration (splitting, trimming, and complete replacement) in patients with distal radius fractures (DRFs) treated in circumferential cast. To determine whether performing early cast alterations affects the fracture alignment.
Conclusions: Cast alteration is commonplace after casting of DRFs but is not associated with the loss of alignment. Patients with polytrauma may benefit from immediate cast splitting. Key Words: distal radius fracture, circumferential cast, cast alteration, cast splitting
Design: Retrospective Cohort Study.
Level of Evidence: Prognostic Level II. See Instructions for Authors for a complete description of levels of evidence.
Setting: Level 1 Trauma Center.
(J Orthop Trauma 2014;28:e186–e190)
Patients: All adult patients who presented with a DRF to a tertiary care hospital over a 3-year period. Intervention: All DRFs without immediate surgical indications are initially treated with circumferential casts at this center.
Outcome Measurements: The following variables were analyzed: patient demographics, polytrauma at the time of injury, physician subspecialty performing reduction, and type of cast alteration. Radiographs were used to assess initial fracture characteristics and secondary displacement of reduction over time. Analysis was performed primarily to identify predictive variables for the early cast alteration and secondarily to determine the effect of these alterations on fracture alignment. Results: 296 patients were included in the study. One of every 4–5 patients had their cast altered within the first 10 days of treatment. One of 3 polytrauma patients had their cast altered. No type of cast alteration was found to be significantly predictive of loss of fracture alignment at 2 or 6 weeks.
Accepted for publication November 21, 2013. From the *Division of Orthopedic Surgery, McGill University Health Center, Montreal General Hospital, Montreal, Canada; and †Faculty of Medicine, McGill University, Montreal, Canada. This work should be attributed to the Division of Orthopedic Surgery, McGill University Health Center, Montreal, Canada. Presented at the American Academy of Orthopedic Surgeons Annual Meeting, 2012, San Francisco, CA and the Canadian Orthopedic Association Annual Meeting, 2011, St. Johns, NL. J. Drager received a research bursary from the study institution. It did not play a role in the investigation. The authors report no conflict of interest. Reprints: Justin Drager, MD, HBSc, Division of Orthopedic Surgery, McGill University Health Center, Montreal General Hospital, 1650 Cedar Avenue, Montreal, QC H3G 1A4, Canada (e-mail: justin.drager@mail. mcgill.ca). Copyright © 2013 by Lippincott Williams & Wilkins
e186
| www.jorthotrauma.com
INTRODUCTION There is substantial variation in the type of cast or splint immobilization applied after the manipulation of fractures of the distal radius. Currently, no method has proven superior in reduction stability or functional outcome such that practitioner preferences continue to dictate therapy.1–4 At this level 1 trauma center, all closed distal radius fractures (DRFs) initially considered for nonoperative management are immobilized in a below-elbow circumferential plaster cast. This protocol is based on the assumption that circumferential casting provides more support to the fracture than splinting.5 Initial full casting also reduces the workload on the outpatient orthopaedic clinic by avoiding the need to change noncircumferential splints to a full cast at followup.6,7 Despite these benefits, casts have limited ability to accommodate ongoing tissue swelling that may result in patient discomfort or potentially serious complications, including pressure sores, neurovascular compromise, such as acute carpel tunnel syndrome or possibly forearm compartment syndrome.4,5 Furthermore, because soft tissue swelling decreases over time, loosening of the cast may lead to a loss of reduction. After operative intervention has been ruled out, a cast can be altered by trimming, splitting, or complete replacement in an attempt to reduce symptoms and/or maintain immobilization.8 Despite the longstanding use of circumferential casting for DRFs, there is a paucity of literature describing the prevalence of cast alterations or predictive factors to suggest which fractures treated by casting are most likely to result in a secondary cast alteration. Finally, although cast alterations may theoretically compromise fracture stability, no study to date has determined if such an assertion is true. The primary objective of this study was to identify the prevalence and predictive factors for early cast alteration. The secondary objective was to determine whether cast alterations performed within the first 10 days after injury affected early (2 weeks) or late (6 weeks) fracture alignment. J Orthop Trauma Volume 28, Number 8, August 2014
J Orthop Trauma Volume 28, Number 8, August 2014
PATIENTS AND METHODS Hospital and university institutional review board approval was obtained before commencing the study. Patient records and radiologic examinations were retrospectively reviewed for all adult patients who presented to a tertiary care hospital with a DRF over a 3-year period (2006–2009). A logbook system was initiated in the orthopaedic clinic that detailed all cast visits and cast alterations. Criteria for inclusion in the study were all patients who presented with a closed DRF, regardless of the initial displacement, and treated nonoperatively with a below-elbow circumferential plaster cast until, at minimum, the first outpatient follow-up visit. Indications for initial operative management of DRFs at our institution are not standardized, and staff surgeons use different criteria for attempting a trial of nonoperative management. This study excluded patients who were initially offered operative management at the time of injury and underwent an acute or subacute surgical intervention before the first clinical follow-up. These patients were excluded, given that their time in cast was insufficient to assess the occurrence of a cast alteration. Patients who did not have clinical or radiographic follow-up records up to 6 weeks after injury were also excluded from this study (Fig. 1). Standard treatment of all closed DRFs in our center involves closed reduction under hematoma block or conscious sedation in the emergency department and immobilization with a molded below-elbow plaster cast in slight ulnar deviation (approximately 15 degrees) and wrist flexion (20– 30 degrees). Fractures are reduced either by an Emergency Medicine physician or an Orthopaedic Surgery resident. Patients with an unsatisfactory postreduction alignment or an unstable fracture pattern, as determined by the surgeon on call, are offered an acute or subacute surgical intervention,
FIGURE 1. Flow diagraph of patient cohort. Ó 2013 Lippincott Williams & Wilkins
Circumferential Casting of DRFs
provided the benefits outweigh the operative risks. Patients with fractures in an acceptable radiologic alignment after reduction are discharged and instructed to return to the orthopaedic clinic or emergency room if they experience increased pain, pressure, or paresthesia in cast. Those returning with such symptoms are examined urgently by an orthopaedic resident in consultation with a staff member, and casts are sometimes altered by splitting, trimming, or complete replacement to relieve symptoms. Any significant neurologic symptoms are closely followed for outcomes requiring timely intervention. All patients treated nonoperatively are followed up in an outpatient orthopaedic clinic within 1–2 weeks of the injury, and if maintaining nonoperative reduction, are seen at least once more by 6 weeks after injury at which point casts are removed to determine clinical healing. Radiographs are taken at each clinic visit, and patients sustaining secondary displacement of the fracture fragments outside of acceptable limits are also offered surgical intervention.
Data Collection The patients’ charts were reviewed, and the following variables were analyzed: patient age, patient gender, presence or absence of polytrauma at the time of injury (2 or more significant injuries sustained requiring inpatient management), subspecialty of physician performing initial reduction (emergentologist vs. orthopaedic resident), and type of cast alteration (if any) in the first 10 days after fracture. Cast alterations were stratified into 3 different types: cast trimming (removing sharp edges or loosening focal pressure points), cast splitting (longitudinal dorsal/volar splitting), and complete cast replacement. Any patients who returned to seek medical advice with an abnormal neurologic examination of the radial, ulnar, and/or median nerve or underwent an operative decompression procedure, postcast application were documented. Three observers separately reviewed prefracture and postfracture radiographs. All prereduction fracture radiographs were classified according to the AO/OTA fracture classification for the distal radius.9 Posteroanterior and lateral radiographs from each outpatient visit until 6-week after injury were scrutinized for several well-defined radiographic parameters (radial inclination, ulnar variance, and dorsal/volar tilt). Displacement of the fracture from the initial reduction was defined as the presence of any of the following changes in radiographic parameters: 1. .10 degrees change in radial inclination. 2. .5 degrees change in volar tilt. 3. .3 mm change in ulnar variance. Fracture alignment was assessed at 2 time points, early assessment at approximately 2 weeks after injury, and then again at the termination of casting at approximately 6 weeks. Patients with a fracture that displaced into the range of surgical management at the first follow-up were taken to the operating room. These patients were therefore not evaluated at the 6-week time point. Alternatively, patients who did not have radiographic follow-up, until after 2 weeks, were omitted from the early time point but included at the 6-week mark. www.jorthotrauma.com |
e187
J Orthop Trauma Volume 28, Number 8, August 2014
Drager et al
FIGURE 2. Age distribution of population.
Statistical Analysis Statistical analyses were performed using SPSS software (version 19.0; IBM Crop, Armonk, NY). Two independent logistic regressions were performed after data collection. The first regression examined the predictive nature of each separate patient-dependent variable on the presence or absence of a cast alteration. These variables include the following: linear (age), nominal (AO fracture classification), and binomial variables (gender, subspecialty of physician performing reduction, presence or absence of polytrauma). The second regression examined the predictive nature of the previously mentioned variables and the binomial variable, presence or absence of a cast alteration, on fracture alignment (maintenance/displacement) at both 2 and 6 weeks after injury.
RESULTS Over a 3-year period, 489 patients presented to our tertiary institution with DRFs (Fig. 1). Twenty patients had incomplete clinical or radiographic data and were not retained
TABLE 1. Distribution of Cast Alterations Variable Gender Male Female AO classification A B C Treating physician Orthopaedics Emergency room Trauma Isolated injury Polytrauma
Cast Altered
No Alteration
20 (24) 46 (22)
65 (76) 165 (78)
45 (23) 11 (22) 10 (21)
154 (77) 39 (78) 37 (79)
59 (22) 7 (21)
204 (78) 26 (79)
56 (21) 10 (36)
212 (79) 18 (64)
Values are given as the number of patients with percentage in parentheses.
e188
| www.jorthotrauma.com
for analysis. 173 patients were initially offered surgical management and were therefore also excluded from the study. 296 patients were initially treated nonoperatively and retained for analysis. Of these, 211 (71%) patients were women, and mean patient age was 69 years with a range of 18–96 years (Fig. 2). Twenty-eight patients (9%) sustained polytraumatic injuries. Fracture characteristics were distributed as follows: 199 (67%) type A, 50 (17%) type B, and 47 (16%) type C. The majority (263, 89%) of closed reductions were performed by orthopaedic residents, with the remainder (33, 11%) being performed by emergency room physicians. Of the 296 patients initially treated in cast, 66 (22.2%) patients underwent a cast alteration during the early treatment. Splitting was the most common alteration performed, occurring 33 times (11.1% of the study population), followed by application of a new cast (24; 8.1%) and cast trimming (9; 3.0%). When stratified according to the fracture classification, patients with type A fractures had an alteration frequency of 23%, type B 22%, and type C 21%. Twenty-two percent of patients treated by orthopaedic residents had their cast altered as opposed to 21% of those treated by emergency room physicians. Thirty-six percent of patients with multiple injuries had their cast altered compared with 21% of patients with isolated DRFs. These results are summarized in Table 1. Using a logistic regression model, we determined that the presence of polytrauma was predictive of performing a cast alteration (P = 0.02). Increasing patient age (P = 0.161), increasing articular involvement of fracture (P = 0.621), and type of physician performing reduction (P = 0.668) had no predictive value for cast alterations. No patients in our series returned after initial cast application with symptoms consistent with acute carpel tunnel syndrome or forearm compartment syndrome that required operative decompression. Six patients returned with neuropathic symptoms in the affected hand after cast placement. Of these, 2 patients experienced sensory paresthesias over the radial nerve distribution. The onset of symptoms occurred at the time of injury before cast application. At the first followup, the distal portion of the patient’s casts was trimmed above the thumb metacarpophalangeal joint, and the sensory symptoms resolved over time. One patient presented at the first Ó 2013 Lippincott Williams & Wilkins
J Orthop Trauma Volume 28, Number 8, August 2014
Circumferential Casting of DRFs
follow-up with decreased sensation over the ulnar nerve distribution. The onset of symptoms occurred after cast application in the emergency room and resolved with time after the cast was removed and replaced at the first follow-up. Three patients presented with median nerve dysfunctions. One patient was known to have existing carpel tunnel syndrome in the affected hand before injury. The symptoms were exacerbated at the time of injury and did not resolve after reduction and cast application. There was no clinical evidence of acute carpel tunnel syndrome requiring urgent decompression. The cast was split prophylactically. This neuropathy progressed over time, and at 3 weeks, the patient showed the signs of motor dysfunction and underwent median nerve decompression. Two additional patients were presented at the first follow-up with decreased sensation over the median nerve distribution. Both patients’ symptoms were present at the time of injury before cast application and completely resolved over time. Overall, 39% of patients at 2 weeks and 43% at 6 weeks demonstrated radiographic evidence of some fracture displacement as per the above criteria. About 18% of all patients underwent surgical fixation after a trial of nonoperative management. Table 2 demonstrates the distribution of significant fracture displacement, as defined above, stratified by the 3 types of cast alternations. Logistic regression demonstrated that performing a cast alteration was not significantly predictive for fracture displacement at 2 (P = 0.798) or 6 weeks (P = 0.987). No specific alteration was found to be predictive of fracture displacement at either 2 or 6 weeks: cast splitting (P = 0.807; P = 0.710), cast trimming (P . 0.9; P . 0.9), or cast replacement (P = 0.505; P = 0.869). Of note, increasing patient age (P , 0.01) and fracture classification (P = 0.049) were found to be predictive of fracture displacement at 6 weeks.
DISCUSSION Circumferential casting may provide superior initial structural support for DRFs compared with splinting and spares the patient from undergoing a secondary casting procedure.6 Casting in the acute setting has been previously cautioned because of concerns of soft tissue constriction potentially resulting in increased pain, focal pressure necrosis, nerve injury, or even forearm compartment syndrome.5 Once
events requiring urgent operative intervention have been ruled out, casts can be altered by splitting, trimming, or replacement in an attempt to improve any patient symptoms. The prevalence and predictive factors for this occurrence and the effect of altering the cast on fracture reduction are currently unknown. Therefore, the objectives of our study were to identify the prevalence and predictive factors of the early cast alteration and to determine whether cast alteration compromised fracture alignment. This series demonstrates that 1 of every 4–5 patients treated in cast had their cast altered within the first 10 days of treatment. In our opinion, most of these alterations are performed for patients returning with pain and numbness from a tight cast, focal pressure points, or cast loosening. Importantly, no patient treated in cast in this study returned with the symptoms of forearm compartment syndrome or acute carpel tunnel syndrome resulting in an acute operative decompression. Split casts have been shown to reduce compartment pressures and better accommodate tissue swelling as compared with a single-sided splint.10,11 Although some patients return with a tight cast, timely treatment with splitting, trimming, or replacement can reduce patient symptoms and requires less outpatient adjustment time compared with the removal of a splint and subsequent casting for every patient with a DRF. The presence of a polytraumatic injury was the single significant predictive factor of a patient undergoing a cast alteration in our series. Patients with polytrauma usually suffer high-energy injuries, such as motor vehicle accidents, and are likely to have more severe bony and soft tissue injuries. Furthermore, the systemic inflammatory response and aggressive fluid resuscitation contribute to abundant soft tissue swelling in these patients.5 Moreover, these patients might be obtunded, comatose, or have other distracting injuries, impairing their ability to verbalize the symptoms of compartment syndrome or nerve injury.12 Because of this finding, we advocate that when treating patients with a DRF as part of a polytraumatic injury, circumferential casts should be immediately split as a preventative measure. In addition, practitioners should perform serial clinical examinations of the injured extremity. Little has been reported about the stability of DRFs after early cast alterations. There exists a theoretical consensus that splitting casts may compromise fracture alignment.4,10 A study conducted by Crickard et al13 demonstrated that splitting
TABLE 2. Effect of Cast Alteration on Reduction Stability at 2 and 6 Weeks*† 2 Weeks Alteration None Cast replaced Cast split Cast trimmed
No Data‡ 13 (6) 2 (8) 2 (6) 0
Stable 131 13 19 6
(57) (54) (58) (67)
6 Weeks Unstable 86 9 12 3
(37) (38) (36) (33)
Previous Surgery§ 40 (17) 3 (12.5) 4 (12) 0
Stable 101 15 17 8
Unstable
(44) (62.5) (52) (89)
89 6 12 1
(39) (25) (36) (11)
*The values are given as the number of patients with percentage in parentheses. †Reduction stability defined as described in method. ‡Indicates patients whose first follow-up radiographs taken after 2 weeks post injury. §Indicates patients who failed conservative management and went on to early operative treatment.
Ó 2013 Lippincott Williams & Wilkins
www.jorthotrauma.com |
e189
Drager et al
significantly decreases the bending forces and load to failure of a fiberglass cast. Neilson et al8 showed that a split plaster cast has the ability to accommodate volar/dorsal load similar to that of an unaltered cast when split on the dorsal aspect but not on any other axis. However, to the best of our knowledge, no studies have examined radiographic fracture stability after splitting or other early cast alterations. Findings from this study demonstrate that early cast alteration, whether splitting, replacing, or trimming a cast, had no deleterious effect on the early or late alignment of DRFs. In this series, patient age and fracture classification were found to be correlated with fracture displacement at 6 weeks. These results are in line with the previous reports of increasing age and intraarticular involvement of a fracture being risk factors for secondary displacement of DRFs.14–16 Although not a primary objective of this study, this finding may act as a validation in our study population. This study does possess limitations. Because of the retrospective design of the study, indications for immediate operative treatment were not standardized and were surgeonspecific. Therefore, it is impossible to know if the sizeable surgically treated group could have benefited from nonoperative management and been included in the study. However, this does not change the observed alteration rate. Furthermore, the follow-up was not performed after cast removal, eliminating the ability to assess return to work/ activities of daily living or long-term subjective and functional outcome scores in those patients with cast alterations. However, all patients had cast removal by 6 weeks with final radiographs out of cast, and this study purports to look at only cast alterations. The current retrospective study demonstrates that cast alteration is commonplace after fracture of the distal radius. However, more than 75% of nonoperatively managed fractures in our series did not need a cast alteration, leading to accelerated outpatient assessment that would otherwise be spent casting 100% of all splinted injuries. Most of these alterations occur in our opinion because of pain and numbness or focal pressure from a tight cast and casts should be assessed immediately after application and at an initial follow-up within the first week. Polytrauma patients may
e190
| www.jorthotrauma.com
J Orthop Trauma Volume 28, Number 8, August 2014
benefit from immediate splitting of a cast given the higher frequency of cast alterations performed in this population and the often inability of these patients to voice complaints. We have demonstrated that in our study population, early cast alteration is a safe procedure that can reduce patient symptoms and did not compromise fracture alignment. REFERENCES 1. Handoll HH, Madhok R. Conservative interventions for treating distal radial fractures in adults. Cochrane Database Syst Rev. 2003:CD000314. 2. O’Connor D, Mullett H, Doyle M, et al. Minimally displaced Colles’ fractures: a prospective randomized trial of treatment with a wrist splint or a plaster cast. J Hand Surg Br. 2003;28:50–53. 3. Wik TS, Aurstad AT, Finsen V. Colles’ fracture: dorsal splint or complete cast during the first 10 days? Injury. 2009;40:400–404. 4. Grafstein E, Stenstrom R, Christenson J, et al. A prospective randomized controlled trial comparing circumferential casting and splinting in displaced Colles fractures. CJEM. 2010;12:192–200. 5. Turner RG, Faber KJ, Athwal GS. Complications of distal radius fractures. Hand Clin. 2010;26:85–96. 6. Wulf CA, Ackerman DB, Rizzo M. Contemporary evaluation and treatment of distal radius fractures. Hand Clin. 2007;23:209–226, vi. 7. Kopylov P. Fractures of the Distal Radius. A Practical Approach to Management. 2nd ed. Fernandez DL, Jupiter JB, eds: Springer-Verlag. ISBN: 0 387 95195 4; Price: £125.50, [euro] 179.00. J Hand Surg Br. 2003;28:280–280. 8. Nielsen DM, Ricketts DM. Where to split plaster casts. Injury. 2005;36: 588–589. 9. Marsh JL, Slongo TF, Agel J, et al. Fracture and dislocation classification compendium-2007-Orthopaedic Trauma Association classification, database and outcomes committee. J Orthop Trauma. 2007;21:S1–S133. 10. Younger AS, Curran P, McQueen MM. Backslabs and plaster casts: which will best accommodate increasing intracompartmental pressures? Injury. 1990;21:179–181. 11. Dresing K, Peterson T, Schmit-Neuerburg KP. Compartment pressure in the carpal tunnel in distal fractures of the radius. A prospective study. Arch Orthop Trauma Surg. 1994;113:285–289. 12. Halanski M, Noonan KJ. Cast and splint immobilization: complications. J Am Acad Orthop Surg. 2008;16:30–40. 13. Crickard CV, Riccio AI, Carney JR, et al. Analysis and comparison of the biomechanical properties of univalved and bivalved cast models. J Pediatr Orthop. 2011;31:39–43. 14. Mackenney PJ, McQueen MM, Elton R. Prediction of instability in distal radial fractures. J Bone Joint Surg Am. 2006;88:1944–1951. 15. Cooney WP. Management of Colles’ fractures. J Hand Surg Br. 1989;14: 137–139. 16. Lafontaine M, Hardy D, Delince P. Stability assessment of distal radius fractures. Injury. 1989;20:208–210.
Ó 2013 Lippincott Williams & Wilkins
Circumferential Casting of Distal Radius Fractures Justin Drager, MD, HBSc,* Alberto Carli, MD, MSc,* Bogdan A. Matache, BSc,† Gregory K. Berry, MD, FRCSC,* Rudy Reindl, MD, FRCSC,* and Edward J. Harvey, MD, MSc, FRCSC*
Objectives: To determine the prevalence and predictive factors for the early cast alteration (splitting, trimming, and complete replacement) in patients with distal radius fractures (DRFs) treated in circumferential cast. To determine whether performing early cast alterations affects the fracture alignment.
Conclusions: Cast alteration is commonplace after casting of DRFs but is not associated with the loss of alignment. Patients with polytrauma may benefit from immediate cast splitting. Key Words: distal radius fracture, circumferential cast, cast alteration, cast splitting
Design: Retrospective Cohort Study.
Level of Evidence: Prognostic Level II. See Instructions for Authors for a complete description of levels of evidence.
Setting: Level 1 Trauma Center.
(J Orthop Trauma 2014;28:e186–e190)
Patients: All adult patients who presented with a DRF to a tertiary care hospital over a 3-year period. Intervention: All DRFs without immediate surgical indications are initially treated with circumferential casts at this center.
Outcome Measurements: The following variables were analyzed: patient demographics, polytrauma at the time of injury, physician subspecialty performing reduction, and type of cast alteration. Radiographs were used to assess initial fracture characteristics and secondary displacement of reduction over time. Analysis was performed primarily to identify predictive variables for the early cast alteration and secondarily to determine the effect of these alterations on fracture alignment. Results: 296 patients were included in the study. One of every 4–5 patients had their cast altered within the first 10 days of treatment. One of 3 polytrauma patients had their cast altered. No type of cast alteration was found to be significantly predictive of loss of fracture alignment at 2 or 6 weeks.
Accepted for publication November 21, 2013. From the *Division of Orthopedic Surgery, McGill University Health Center, Montreal General Hospital, Montreal, Canada; and †Faculty of Medicine, McGill University, Montreal, Canada. This work should be attributed to the Division of Orthopedic Surgery, McGill University Health Center, Montreal, Canada. Presented at the American Academy of Orthopedic Surgeons Annual Meeting, 2012, San Francisco, CA and the Canadian Orthopedic Association Annual Meeting, 2011, St. Johns, NL. J. Drager received a research bursary from the study institution. It did not play a role in the investigation. The authors report no conflict of interest. Reprints: Justin Drager, MD, HBSc, Division of Orthopedic Surgery, McGill University Health Center, Montreal General Hospital, 1650 Cedar Avenue, Montreal, QC H3G 1A4, Canada (e-mail: justin.drager@mail. mcgill.ca). Copyright © 2013 by Lippincott Williams & Wilkins
e186
| www.jorthotrauma.com
INTRODUCTION There is substantial variation in the type of cast or splint immobilization applied after the manipulation of fractures of the distal radius. Currently, no method has proven superior in reduction stability or functional outcome such that practitioner preferences continue to dictate therapy.1–4 At this level 1 trauma center, all closed distal radius fractures (DRFs) initially considered for nonoperative management are immobilized in a below-elbow circumferential plaster cast. This protocol is based on the assumption that circumferential casting provides more support to the fracture than splinting.5 Initial full casting also reduces the workload on the outpatient orthopaedic clinic by avoiding the need to change noncircumferential splints to a full cast at followup.6,7 Despite these benefits, casts have limited ability to accommodate ongoing tissue swelling that may result in patient discomfort or potentially serious complications, including pressure sores, neurovascular compromise, such as acute carpel tunnel syndrome or possibly forearm compartment syndrome.4,5 Furthermore, because soft tissue swelling decreases over time, loosening of the cast may lead to a loss of reduction. After operative intervention has been ruled out, a cast can be altered by trimming, splitting, or complete replacement in an attempt to reduce symptoms and/or maintain immobilization.8 Despite the longstanding use of circumferential casting for DRFs, there is a paucity of literature describing the prevalence of cast alterations or predictive factors to suggest which fractures treated by casting are most likely to result in a secondary cast alteration. Finally, although cast alterations may theoretically compromise fracture stability, no study to date has determined if such an assertion is true. The primary objective of this study was to identify the prevalence and predictive factors for early cast alteration. The secondary objective was to determine whether cast alterations performed within the first 10 days after injury affected early (2 weeks) or late (6 weeks) fracture alignment. J Orthop Trauma Volume 28, Number 8, August 2014
J Orthop Trauma Volume 28, Number 8, August 2014
PATIENTS AND METHODS Hospital and university institutional review board approval was obtained before commencing the study. Patient records and radiologic examinations were retrospectively reviewed for all adult patients who presented to a tertiary care hospital with a DRF over a 3-year period (2006–2009). A logbook system was initiated in the orthopaedic clinic that detailed all cast visits and cast alterations. Criteria for inclusion in the study were all patients who presented with a closed DRF, regardless of the initial displacement, and treated nonoperatively with a below-elbow circumferential plaster cast until, at minimum, the first outpatient follow-up visit. Indications for initial operative management of DRFs at our institution are not standardized, and staff surgeons use different criteria for attempting a trial of nonoperative management. This study excluded patients who were initially offered operative management at the time of injury and underwent an acute or subacute surgical intervention before the first clinical follow-up. These patients were excluded, given that their time in cast was insufficient to assess the occurrence of a cast alteration. Patients who did not have clinical or radiographic follow-up records up to 6 weeks after injury were also excluded from this study (Fig. 1). Standard treatment of all closed DRFs in our center involves closed reduction under hematoma block or conscious sedation in the emergency department and immobilization with a molded below-elbow plaster cast in slight ulnar deviation (approximately 15 degrees) and wrist flexion (20– 30 degrees). Fractures are reduced either by an Emergency Medicine physician or an Orthopaedic Surgery resident. Patients with an unsatisfactory postreduction alignment or an unstable fracture pattern, as determined by the surgeon on call, are offered an acute or subacute surgical intervention,
FIGURE 1. Flow diagraph of patient cohort. Ó 2013 Lippincott Williams & Wilkins
Circumferential Casting of DRFs
provided the benefits outweigh the operative risks. Patients with fractures in an acceptable radiologic alignment after reduction are discharged and instructed to return to the orthopaedic clinic or emergency room if they experience increased pain, pressure, or paresthesia in cast. Those returning with such symptoms are examined urgently by an orthopaedic resident in consultation with a staff member, and casts are sometimes altered by splitting, trimming, or complete replacement to relieve symptoms. Any significant neurologic symptoms are closely followed for outcomes requiring timely intervention. All patients treated nonoperatively are followed up in an outpatient orthopaedic clinic within 1–2 weeks of the injury, and if maintaining nonoperative reduction, are seen at least once more by 6 weeks after injury at which point casts are removed to determine clinical healing. Radiographs are taken at each clinic visit, and patients sustaining secondary displacement of the fracture fragments outside of acceptable limits are also offered surgical intervention.
Data Collection The patients’ charts were reviewed, and the following variables were analyzed: patient age, patient gender, presence or absence of polytrauma at the time of injury (2 or more significant injuries sustained requiring inpatient management), subspecialty of physician performing initial reduction (emergentologist vs. orthopaedic resident), and type of cast alteration (if any) in the first 10 days after fracture. Cast alterations were stratified into 3 different types: cast trimming (removing sharp edges or loosening focal pressure points), cast splitting (longitudinal dorsal/volar splitting), and complete cast replacement. Any patients who returned to seek medical advice with an abnormal neurologic examination of the radial, ulnar, and/or median nerve or underwent an operative decompression procedure, postcast application were documented. Three observers separately reviewed prefracture and postfracture radiographs. All prereduction fracture radiographs were classified according to the AO/OTA fracture classification for the distal radius.9 Posteroanterior and lateral radiographs from each outpatient visit until 6-week after injury were scrutinized for several well-defined radiographic parameters (radial inclination, ulnar variance, and dorsal/volar tilt). Displacement of the fracture from the initial reduction was defined as the presence of any of the following changes in radiographic parameters: 1. .10 degrees change in radial inclination. 2. .5 degrees change in volar tilt. 3. .3 mm change in ulnar variance. Fracture alignment was assessed at 2 time points, early assessment at approximately 2 weeks after injury, and then again at the termination of casting at approximately 6 weeks. Patients with a fracture that displaced into the range of surgical management at the first follow-up were taken to the operating room. These patients were therefore not evaluated at the 6-week time point. Alternatively, patients who did not have radiographic follow-up, until after 2 weeks, were omitted from the early time point but included at the 6-week mark. www.jorthotrauma.com |
e187
J Orthop Trauma Volume 28, Number 8, August 2014
Drager et al
FIGURE 2. Age distribution of population.
Statistical Analysis Statistical analyses were performed using SPSS software (version 19.0; IBM Crop, Armonk, NY). Two independent logistic regressions were performed after data collection. The first regression examined the predictive nature of each separate patient-dependent variable on the presence or absence of a cast alteration. These variables include the following: linear (age), nominal (AO fracture classification), and binomial variables (gender, subspecialty of physician performing reduction, presence or absence of polytrauma). The second regression examined the predictive nature of the previously mentioned variables and the binomial variable, presence or absence of a cast alteration, on fracture alignment (maintenance/displacement) at both 2 and 6 weeks after injury.
RESULTS Over a 3-year period, 489 patients presented to our tertiary institution with DRFs (Fig. 1). Twenty patients had incomplete clinical or radiographic data and were not retained
TABLE 1. Distribution of Cast Alterations Variable Gender Male Female AO classification A B C Treating physician Orthopaedics Emergency room Trauma Isolated injury Polytrauma
Cast Altered
No Alteration
20 (24) 46 (22)
65 (76) 165 (78)
45 (23) 11 (22) 10 (21)
154 (77) 39 (78) 37 (79)
59 (22) 7 (21)
204 (78) 26 (79)
56 (21) 10 (36)
212 (79) 18 (64)
Values are given as the number of patients with percentage in parentheses.
e188
| www.jorthotrauma.com
for analysis. 173 patients were initially offered surgical management and were therefore also excluded from the study. 296 patients were initially treated nonoperatively and retained for analysis. Of these, 211 (71%) patients were women, and mean patient age was 69 years with a range of 18–96 years (Fig. 2). Twenty-eight patients (9%) sustained polytraumatic injuries. Fracture characteristics were distributed as follows: 199 (67%) type A, 50 (17%) type B, and 47 (16%) type C. The majority (263, 89%) of closed reductions were performed by orthopaedic residents, with the remainder (33, 11%) being performed by emergency room physicians. Of the 296 patients initially treated in cast, 66 (22.2%) patients underwent a cast alteration during the early treatment. Splitting was the most common alteration performed, occurring 33 times (11.1% of the study population), followed by application of a new cast (24; 8.1%) and cast trimming (9; 3.0%). When stratified according to the fracture classification, patients with type A fractures had an alteration frequency of 23%, type B 22%, and type C 21%. Twenty-two percent of patients treated by orthopaedic residents had their cast altered as opposed to 21% of those treated by emergency room physicians. Thirty-six percent of patients with multiple injuries had their cast altered compared with 21% of patients with isolated DRFs. These results are summarized in Table 1. Using a logistic regression model, we determined that the presence of polytrauma was predictive of performing a cast alteration (P = 0.02). Increasing patient age (P = 0.161), increasing articular involvement of fracture (P = 0.621), and type of physician performing reduction (P = 0.668) had no predictive value for cast alterations. No patients in our series returned after initial cast application with symptoms consistent with acute carpel tunnel syndrome or forearm compartment syndrome that required operative decompression. Six patients returned with neuropathic symptoms in the affected hand after cast placement. Of these, 2 patients experienced sensory paresthesias over the radial nerve distribution. The onset of symptoms occurred at the time of injury before cast application. At the first followup, the distal portion of the patient’s casts was trimmed above the thumb metacarpophalangeal joint, and the sensory symptoms resolved over time. One patient presented at the first Ó 2013 Lippincott Williams & Wilkins
J Orthop Trauma Volume 28, Number 8, August 2014
Circumferential Casting of DRFs
follow-up with decreased sensation over the ulnar nerve distribution. The onset of symptoms occurred after cast application in the emergency room and resolved with time after the cast was removed and replaced at the first follow-up. Three patients presented with median nerve dysfunctions. One patient was known to have existing carpel tunnel syndrome in the affected hand before injury. The symptoms were exacerbated at the time of injury and did not resolve after reduction and cast application. There was no clinical evidence of acute carpel tunnel syndrome requiring urgent decompression. The cast was split prophylactically. This neuropathy progressed over time, and at 3 weeks, the patient showed the signs of motor dysfunction and underwent median nerve decompression. Two additional patients were presented at the first follow-up with decreased sensation over the median nerve distribution. Both patients’ symptoms were present at the time of injury before cast application and completely resolved over time. Overall, 39% of patients at 2 weeks and 43% at 6 weeks demonstrated radiographic evidence of some fracture displacement as per the above criteria. About 18% of all patients underwent surgical fixation after a trial of nonoperative management. Table 2 demonstrates the distribution of significant fracture displacement, as defined above, stratified by the 3 types of cast alternations. Logistic regression demonstrated that performing a cast alteration was not significantly predictive for fracture displacement at 2 (P = 0.798) or 6 weeks (P = 0.987). No specific alteration was found to be predictive of fracture displacement at either 2 or 6 weeks: cast splitting (P = 0.807; P = 0.710), cast trimming (P . 0.9; P . 0.9), or cast replacement (P = 0.505; P = 0.869). Of note, increasing patient age (P , 0.01) and fracture classification (P = 0.049) were found to be predictive of fracture displacement at 6 weeks.
DISCUSSION Circumferential casting may provide superior initial structural support for DRFs compared with splinting and spares the patient from undergoing a secondary casting procedure.6 Casting in the acute setting has been previously cautioned because of concerns of soft tissue constriction potentially resulting in increased pain, focal pressure necrosis, nerve injury, or even forearm compartment syndrome.5 Once
events requiring urgent operative intervention have been ruled out, casts can be altered by splitting, trimming, or replacement in an attempt to improve any patient symptoms. The prevalence and predictive factors for this occurrence and the effect of altering the cast on fracture reduction are currently unknown. Therefore, the objectives of our study were to identify the prevalence and predictive factors of the early cast alteration and to determine whether cast alteration compromised fracture alignment. This series demonstrates that 1 of every 4–5 patients treated in cast had their cast altered within the first 10 days of treatment. In our opinion, most of these alterations are performed for patients returning with pain and numbness from a tight cast, focal pressure points, or cast loosening. Importantly, no patient treated in cast in this study returned with the symptoms of forearm compartment syndrome or acute carpel tunnel syndrome resulting in an acute operative decompression. Split casts have been shown to reduce compartment pressures and better accommodate tissue swelling as compared with a single-sided splint.10,11 Although some patients return with a tight cast, timely treatment with splitting, trimming, or replacement can reduce patient symptoms and requires less outpatient adjustment time compared with the removal of a splint and subsequent casting for every patient with a DRF. The presence of a polytraumatic injury was the single significant predictive factor of a patient undergoing a cast alteration in our series. Patients with polytrauma usually suffer high-energy injuries, such as motor vehicle accidents, and are likely to have more severe bony and soft tissue injuries. Furthermore, the systemic inflammatory response and aggressive fluid resuscitation contribute to abundant soft tissue swelling in these patients.5 Moreover, these patients might be obtunded, comatose, or have other distracting injuries, impairing their ability to verbalize the symptoms of compartment syndrome or nerve injury.12 Because of this finding, we advocate that when treating patients with a DRF as part of a polytraumatic injury, circumferential casts should be immediately split as a preventative measure. In addition, practitioners should perform serial clinical examinations of the injured extremity. Little has been reported about the stability of DRFs after early cast alterations. There exists a theoretical consensus that splitting casts may compromise fracture alignment.4,10 A study conducted by Crickard et al13 demonstrated that splitting
TABLE 2. Effect of Cast Alteration on Reduction Stability at 2 and 6 Weeks*† 2 Weeks Alteration None Cast replaced Cast split Cast trimmed
No Data‡ 13 (6) 2 (8) 2 (6) 0
Stable 131 13 19 6
(57) (54) (58) (67)
6 Weeks Unstable 86 9 12 3
(37) (38) (36) (33)
Previous Surgery§ 40 (17) 3 (12.5) 4 (12) 0
Stable 101 15 17 8
Unstable
(44) (62.5) (52) (89)
89 6 12 1
(39) (25) (36) (11)
*The values are given as the number of patients with percentage in parentheses. †Reduction stability defined as described in method. ‡Indicates patients whose first follow-up radiographs taken after 2 weeks post injury. §Indicates patients who failed conservative management and went on to early operative treatment.
Ó 2013 Lippincott Williams & Wilkins
www.jorthotrauma.com |
e189
Drager et al
significantly decreases the bending forces and load to failure of a fiberglass cast. Neilson et al8 showed that a split plaster cast has the ability to accommodate volar/dorsal load similar to that of an unaltered cast when split on the dorsal aspect but not on any other axis. However, to the best of our knowledge, no studies have examined radiographic fracture stability after splitting or other early cast alterations. Findings from this study demonstrate that early cast alteration, whether splitting, replacing, or trimming a cast, had no deleterious effect on the early or late alignment of DRFs. In this series, patient age and fracture classification were found to be correlated with fracture displacement at 6 weeks. These results are in line with the previous reports of increasing age and intraarticular involvement of a fracture being risk factors for secondary displacement of DRFs.14–16 Although not a primary objective of this study, this finding may act as a validation in our study population. This study does possess limitations. Because of the retrospective design of the study, indications for immediate operative treatment were not standardized and were surgeonspecific. Therefore, it is impossible to know if the sizeable surgically treated group could have benefited from nonoperative management and been included in the study. However, this does not change the observed alteration rate. Furthermore, the follow-up was not performed after cast removal, eliminating the ability to assess return to work/ activities of daily living or long-term subjective and functional outcome scores in those patients with cast alterations. However, all patients had cast removal by 6 weeks with final radiographs out of cast, and this study purports to look at only cast alterations. The current retrospective study demonstrates that cast alteration is commonplace after fracture of the distal radius. However, more than 75% of nonoperatively managed fractures in our series did not need a cast alteration, leading to accelerated outpatient assessment that would otherwise be spent casting 100% of all splinted injuries. Most of these alterations occur in our opinion because of pain and numbness or focal pressure from a tight cast and casts should be assessed immediately after application and at an initial follow-up within the first week. Polytrauma patients may
e190
| www.jorthotrauma.com
J Orthop Trauma Volume 28, Number 8, August 2014
benefit from immediate splitting of a cast given the higher frequency of cast alterations performed in this population and the often inability of these patients to voice complaints. We have demonstrated that in our study population, early cast alteration is a safe procedure that can reduce patient symptoms and did not compromise fracture alignment. REFERENCES 1. Handoll HH, Madhok R. Conservative interventions for treating distal radial fractures in adults. Cochrane Database Syst Rev. 2003:CD000314. 2. O’Connor D, Mullett H, Doyle M, et al. Minimally displaced Colles’ fractures: a prospective randomized trial of treatment with a wrist splint or a plaster cast. J Hand Surg Br. 2003;28:50–53. 3. Wik TS, Aurstad AT, Finsen V. Colles’ fracture: dorsal splint or complete cast during the first 10 days? Injury. 2009;40:400–404. 4. Grafstein E, Stenstrom R, Christenson J, et al. A prospective randomized controlled trial comparing circumferential casting and splinting in displaced Colles fractures. CJEM. 2010;12:192–200. 5. Turner RG, Faber KJ, Athwal GS. Complications of distal radius fractures. Hand Clin. 2010;26:85–96. 6. Wulf CA, Ackerman DB, Rizzo M. Contemporary evaluation and treatment of distal radius fractures. Hand Clin. 2007;23:209–226, vi. 7. Kopylov P. Fractures of the Distal Radius. A Practical Approach to Management. 2nd ed. Fernandez DL, Jupiter JB, eds: Springer-Verlag. ISBN: 0 387 95195 4; Price: £125.50, [euro] 179.00. J Hand Surg Br. 2003;28:280–280. 8. Nielsen DM, Ricketts DM. Where to split plaster casts. Injury. 2005;36: 588–589. 9. Marsh JL, Slongo TF, Agel J, et al. Fracture and dislocation classification compendium-2007-Orthopaedic Trauma Association classification, database and outcomes committee. J Orthop Trauma. 2007;21:S1–S133. 10. Younger AS, Curran P, McQueen MM. Backslabs and plaster casts: which will best accommodate increasing intracompartmental pressures? Injury. 1990;21:179–181. 11. Dresing K, Peterson T, Schmit-Neuerburg KP. Compartment pressure in the carpal tunnel in distal fractures of the radius. A prospective study. Arch Orthop Trauma Surg. 1994;113:285–289. 12. Halanski M, Noonan KJ. Cast and splint immobilization: complications. J Am Acad Orthop Surg. 2008;16:30–40. 13. Crickard CV, Riccio AI, Carney JR, et al. Analysis and comparison of the biomechanical properties of univalved and bivalved cast models. J Pediatr Orthop. 2011;31:39–43. 14. Mackenney PJ, McQueen MM, Elton R. Prediction of instability in distal radial fractures. J Bone Joint Surg Am. 2006;88:1944–1951. 15. Cooney WP. Management of Colles’ fractures. J Hand Surg Br. 1989;14: 137–139. 16. Lafontaine M, Hardy D, Delince P. Stability assessment of distal radius fractures. Injury. 1989;20:208–210.
Ó 2013 Lippincott Williams & Wilkins
