SCIENTIFIC ARTICLE
A Stepwise Algorithm for Surgical Treatment of Type II Displaced Pediatric Phalangeal Neck Fractures Jonas L. Matzon, MD, Roger Cornwall, MD
Purpose To evaluate a stepwise reduction algorithm including closed, percutaneous, and open reduction techniques followed by percutaneous pin fixation for displaced pediatric phalangeal neck fractures. Methods Sixty-one consecutive children (mean age, 9.4 y; range, 2e18 y) presenting with closed, type II displaced phalangeal neck fractures were treated using the following algorithm. If satisfactory reduction was achieved with closed reduction, percutaneous pinning (CRPP) was performed. If not achieved, then percutaneous reduction and pinning (PRPP) was performed using a temporary intrafocal joystick for reduction and for osteoclasis as needed. If percutaneous reduction failed, open reduction and percutaneous pinning (ORPP) was performed. Using the Al-Qattan system, radiographic and clinical outcomes were retrospectively graded for union, deformity, range of motion, and function. Results Satisfactory alignment was achieved in all 61 fractures, by CRPP in 49 and PRPP in 12. No fracture required open reduction. Mean number of days from injury to surgery was 8 days for CRPP and 17 days for PRPP. All fractures treated after 13 days required percutaneous pinning. Fifty-three patients were followed for at least 1 year or until full functional recovery was achieved, with 45 excellent, 4 good, 1 fair, and 3 poor results. Four complications accounted for the fair and poor results, including 1 flexion contracture, 1 nonunion following pin track infection, and 1 case of avascular necrosis following a severe crush injury in the CRPP group and 1 flexion contracture following PRPP. Conclusions Our stepwise algorithm for surgical treatment of closed, type II displaced pediatric phalangeal neck fractures produced 92% good to excellent results while minimizing the need for open reduction even in late-presenting fractures. (J Hand Surg Am. 2014;39(3):467e473. Copyright Ó 2014 by the American Society for Surgery of the Hand. All rights reserved.) Type of study/level of evidence Therapeutic IV. Key words Pediatric, percutaneous fixation, percutaneous reduction, phalangeal neck fractures.
From the Division of Orthopaedic Surgery, Children’s Hospital of Philadelphia, Philadelphia, Pennsylvania; and the Division of Orthopaedic Surgery, Cincinnati Children’s Hospital Medical Center, Cincinnati, Ohio. Received for publication July 23, 2013; accepted in revised form December 12, 2013. No benefits in any form have been received or will be received related directly or indirectly to the subject of this article. Corresponding author: Roger Cornwall, MD, Cincinnati’s Children’s Hospital Medical Center, Division of Orthopaedic Surgery, 3333 Burnet Ave., MLC 2017, Cincinnati, OH 45229; e-mail: [email protected]. 0363-5023/14/3903-0009$36.00/0 http://dx.doi.org/10.1016/j.jhsa.2013.12.014
P
account for 13% of finger fractures in children.1 Based on the AlQattan classification, phalangeal neck fractures can be nondisplaced (type I), displaced with boneto-bone contact at the fracture site (type II), and displaced with no bone-to-bone contact (type III).2 The majority of these fractures are displaced, and displaced fractures can cause considerable functional problems if not reduced anatomically.3e5 Similarly, reduced fractures have a strong tendency to redisplace unless stabilized by pin fixation.2,4 Therefore, HALANGEAL NECK FRACTURES
Ó 2014 ASSH
r
Published by Elsevier, Inc. All rights reserved.
r
467
468
DISPLACED PEDIATRIC PHALANGEAL NECK FRACTURES
current recommendations for the treatment of such fractures dictate timely reduction and pin fixation to stabilize the fracture.2,6e9 Although there is little question that pin fixation is required to stabilize the fracture once reduced, the best method for reduction is debatable. Some authors recommend open reduction,2,8,10 and others recommend closed reduction.6,7,9 Proponents of open reduction cite the inability to adequately correct the deformity by closed means in all cases, whereas proponents of closed reduction cite the risk of avascular necrosis of the phalangeal condyles following open reduction.9 A complicating factor is that many children present to the hand surgeon 2 to 3 weeks after injury. By then, closed reduction is often unsuccessful, but open reduction may carry an even greater risk of injuring the blood supply to the healing fracture fragments.9 For this reason, a percutaneous reduction technique has been described as an alternative to open reduction when closed reduction is unsuccessful, with a theoretically lower risk of avascular necrosis compared with open reduction.9,11 However, the roles and efficacies of the various reduction techniques have not been elucidated, as previous retrospective studies reporting the results of treatment of phalangeal neck fractures have included a variety of treatment strategies and/or a small number of patients,2e4,8e11 which has led to a lack of consensus regarding treatment strategies.12 The primary purpose of the study was to evaluate the effectiveness of a stepwise algorithm to obtain reduction incorporating closed, percutaneous, and open reduction techniques in a consecutive series of patients with closed, type II displaced phalangeal neck fractures. The secondary purpose was to report the clinical results of those patients treated with this algorithm.
in the operating room. If satisfactory reduction was achieved, percutaneous interfragmentary pins were placed to stabilize the fracture (Fig. 2). The size of the stabilizing K-wires was dependent on the child’s age and ranged from 0.7 to 1.1 mm. One or 2 K-wires were passed obliquely from distal to proximal to stabilize the fracture. A crossed pin configuration was used for transverse fractures, and a parallel or divergent configuration was used for oblique fractures. A single K-wire was used in very distal fractures or small fragments in which 2 pins could not be placed. The number of pins used for each patient was not recorded. Regardless of pin number or configuration, the pins were placed with the adjacent interphalangeal joint in as little flexion as possible to minimize the risk of a flexion contracture. If, however, closed reduction was unsuccessful at achieving satisfactory alignment, percutaneous reduction was performed using a temporary intrafocal K-wire for reduction and percutaneous osteoclasis as needed and as described by Waters et al.11 Specifically, a K-wire was placed by hand under fluoroscopic guidance into the fracture site at the concavity of the deformity. Depending on the direction of displacement (dorsal, dorsoradial, or dorsoulnar), the K-wire was placed either through or around the central extensor apparatus so that the K-wire was inserted into the concave side of the fracture in the plane of maximum deformity. A sweeping motion was used to disrupt fracture callus. Then the K-wire was driven by hand through the fracture site until the tip was just distal and palmar to the distal end of the far cortex of the proximal fragment, such that the palmar cortex of the proximal fragment could be used as a fulcrum for levering the distal fragment into position. The adjacent interphalangeal joint was simultaneously flexed as necessary to assist in the reduction. The size of the K-wire was dependent on the child’s age and ranged from 0.9 to 1.4 mm.11 If percutaneous reduction was successful, percutaneous interfragmentary pins were placed as described previously. Open reduction was reserved for fractures that could be reduced by neither closed nor percutaneous methods and when the criteria for remodeling were not met, as described by Cornwall and Waters13 (Fig. 1). Regardless of the method of reduction, the patients were immobilized in a cast following pinning. Patients were seen in the office at 4 to 5 weeks after surgery. If clinical and radiographic union was apparent, both the cast and the K-wires were removed, and range of motion exercises were started at home. Patients returned to the office at 6 to 7 weeks
MATERIALS AND METHODS Treatment algorithm Since 2004, the senior author (R.C.) has used a stepwise treatment algorithm for closed, type II displaced phalangeal neck fractures in children (Fig. 1). Non-displaced or minimally displaced fractures with normal adjacent interphalangeal joint motion and no coronal or rotational deformity were treated conservatively with close follow-up to ensure maintenance of satisfactory alignment until healing. Fractures with initial displacement sufficient to affect adjacent joint motion or produce coronal or rotational deformity were treated with an attempt at closed reduction J Hand Surg Am.
r
Vol. 39, March 2014
DISPLACED PEDIATRIC PHALANGEAL NECK FRACTURES
469
FIGURE 1: Minimally invasive stepwise treatment algorithm for displaced phalangeal neck fractures in children. *Fractures were considered healed if they were nontender and with no radiolucent fracture line visible, regardless of time from injury and amount of callus. **Success is based on no coronal or rotational deformity and normal adjacent interphalangeal joint motion. †See text for remodeling criteria used. zSalvage procedures include osteotomy and subcondylar fossa reconstruction.
after surgery for evaluation of range of motion. Patients with at least 90 of flexion and less than 10 of flexion contracture were discharged. If range of motion was limited, the patient was referred to occupational therapy and seen back in the office 4 weeks later.
treatment (surgical or nonsurgical) by another doctor, types I and III phalangeal neck fractures, open fractures, and failure to complete follow-up. Sample characteristics Outpatient and inpatient medical records were retrieved and reviewed for age, sex, location of fracture, mechanism of injury, date of injury, definitive treatment, follow-up, K-wire removal, type of definitive treatment, postoperative care, and referral to occupational therapy. There were 43 boys and 18 girls, with an average age of 9.4 years (range, 2e18 y). The most common injury mechanism was sports (n ¼ 31), followed by a fall (n ¼ 11), and crush injury in a door (n ¼ 8). The most frequently fractured finger was the little finger (n ¼ 40), followed by the ring finger (n ¼ 12), the middle
Study patients This retrospective case series of patients treated with a prospectively applied treatment algorithm was approved by our institutional review board. Sixty-one consecutive children treated surgically from 2004 to 2008 for closed, type II displaced phalangeal neck fractures using the stepwise algorithm were identified using Current Procedural Terminology billing codes. Patients were excluded from participation for age older than 18 years, a previous attempt at definitive J Hand Surg Am.
r
Vol. 39, March 2014
470
DISPLACED PEDIATRIC PHALANGEAL NECK FRACTURES
FIGURE 2: Anteroposterior A and lateral B radiographs of a 22-day-old incipient malunion. A K-wire is placed into the dorsal callus to perform osteoclasis C; the K-wire is pointed distally and introduced into the fracture site D; the K-wire is advanced until the tip rests at the dorsal end of the volar cortex of the proximal fragment, and the pin is used to lever the distal fragment into flexion E. F Following reduction, full interphalangeal joint flexion and disrupted dorsal callus are visible. The fracture is then secured with crossed K-wired G, H, which are removed 4 weeks later after radiographs demonstrate satisfactory healing I, J.
finger (n ¼ 9), the index finger (n ¼ 7), and the thumb (n ¼ 3). Forty-four fractures involved the proximal phalanx, and 17 involved the middle phalanx.
patient/parent satisfaction, and complaints of pain. For patients who achieved functional range of motion shortly after K-wire removal, long-term follow-up was not required. For patients who did not achieve functional range of motion during clinic follow-up, long-term follow-up was collected over the telephone and through examination in clinic. Data were collected regarding range of motion, deformity, pain, and dysfunction. Radiographic and clinical outcomes were retrospectively graded by the Al-Qattan system,2 including union, deformity, range of motion, and function (Table 1).
Radiographic and clinical outcomes Pre-, intra-, and postoperative radiographs were reviewed for direction and degree of displacement, location of fracture, quality of reduction, avascular necrosis, and ultimate deformity, as defined by AlQattan (Table 1). Preoperative sagittal angulation was not recorded owing to concomitant coronal plane angulation and difficulty reliably identifying the axis of the condyles seen in parallax. Postoperative sagittal alignment was measured and considered acceptable if the adjacent proximal interphalangeal joint has normal motion. Medical records were reviewed for functional outcomes data, including range of motion at early follow-up after K-wire removal, progress in occupational therapy, final range of motion, deformity, J Hand Surg Am.
Statistical analysis Nominal variables were compared using the Fisher exact test given the expectedly small sample sizes in the more invasive groups (percutaneous and open reduction). Continuous variables (assuming parametric distributions) were compared using 2-tailed t tests. Statistical significance was set at P less than .05. r
Vol. 39, March 2014
471
DISPLACED PEDIATRIC PHALANGEAL NECK FRACTURES
TABLE 1. Comparison of Al-Qattan (5) Outcomes Classification Between Closed Reduction and Percutaneous Reduction of Pediatric Displaced Phalangeal Neck Fractures Excellent
Good
Fair
Poor
Classification Union
Yes
Yes
Yes
Nonunion or avascular necrosis
50 at IP/DIP 50 e90 at PIP
< 50 at PIP
No
Mild
Severe
Good
Useful
Lost
38 (86%)
3 (7%)
1 (2%)
2 (5%)
7 (78%)
1 (11%)
0 (0%)
1 (11%)
Joint range of motion
Full
> 50 at IP/DIP > 90 at PIP
Residual deformity
No
Digit function
Normal
CRPP PRPP
Results
DIP, distal interphalangeal joint of fingers; IP, interphalangeal joint of thumb; PIP, proximal interphalangeal joint of fingers.
TABLE 2. Comparison of Patient and Injury Characteristics Between Closed Reduction and Percutaneous Reduction of Pediatric Displaced Phalangeal Neck Fractures CRPP Mean SD (Range)
PRPP Mean SD (Range)
Independent t Tests P Value
Age (y)
9.2 3.4 (2e17)
9.8 4.2 (3e18)
.61
Days to presentation
5.1 2.9 (0e12)
14.3 9.0 (3e35)
< .001*
Days to surgery
7.6 3.2 (0e13)
16.6 8.4 (5e36)
< .001*
*P < .05 indicates statistical significance.
RESULTS Sample characteristics Forty-nine patients were treated using closed reduction percutaneous pinning (CRPP), and 12 required percutaneous reduction percutaneous pinning (PRPP). Using our treatment algorithm, no patients required open reduction percutaneous pinning (ORPP). In all patients, a satisfactory reduction was achieved with no coronal or rotational deformity and with sagittal alignment allowing full adjacent interphalangeal joint range of motion (assessed after reduction but before fixation pin placement). Patients treated with the CRPP presented significantly earlier (P < .001) and had surgery sooner (P < .001) than patients treated with PRPP (Table 2), but both treatment groups were similar in age (P ¼ .61).
had not been discharged from clinic returned to clinic for a follow-up examination an average of 2.5 1.0 years (range, 1.0e4.2 y) after treatment. Eight patients (5 CRPP, 3 PRPP) were lost to follow-up and were excluded from radiographic analyses. Radiographs at final follow-up demonstrated anatomical alignment in 45 patients, acceptable alignment in 7 patients (no coronal or rotational deformity, with sagittal angulation ranging from 30 flexion to 5 extension but with normal adjacent interphalangeal joint motion), and an anatomical nonunion in 1 patient. Occupational therapy was prescribed for 15 patients, 8 of whom actually attended therapy, which was too small a number for powered statistical analysis. Of the 53 patients with complete follow-up, 49 (92%) had good to excellent results (Table 1). One patient (CRPP) had a fair result secondary to a proximal interphalangeal joint flexion contracture with final range of motion of 30 to 90 . The remaining 3 results were poor secondary to a proximal interphalangeal joint flexion contracture (PRPP) with final range of motion of 45 to 80 , a nonunion following a pin track infection (CRPP), and avascular
Radiographic and clinical outcomes Patients were followed until discharged from clinic or by phone after a minimum of 1 year. Thirty-four patients achieved full range of motion by an average of 3 3 months (range, 1e17 mo) and were discharged from clinic. An additional 19 patients who J Hand Surg Am.
r
Vol. 39, March 2014
472
DISPLACED PEDIATRIC PHALANGEAL NECK FRACTURES
necrosis (CRPP) after a crush injury with multiple associated fractures. The latter patient was symptomatic, and avascular necrosis was confirmed on serial radiographs. No difference in outcomes could be detected between the CRPP and the PRPP groups, but this study did not have adequate power to detect a difference.
or more weeks after injury should be familiar with techniques of percutaneous and open reduction. Similarly, although the current study cannot determine the maximum delay from injury to treatment after which percutaneous reduction will be futile, this technique was successful beyond 5 weeks in our series (36 days). Therefore, we recommend considering percutaneous reduction for any fracture that is not yet fully healed, as determined by the presence of tenderness and a radiolucent fracture line, regardless of the amount of callus or the time from injury. Proposed risk factors for concurrent avascular necrosis include having a crush injury; presenting with joint stiffness, ischemia, or congestion; fragmenting the head during closed reduction; and passing the Kwire multiple times through the bone.16,17 The one such risk factor present in our case of avascular necrosis was crush injury. No case of avascular necrosis followed PRPP. The low complication rate among the 12 patients in the PRPP group, especially the absence of avascular necrosis despite being treated beyond 13 days after injury, is in contrast to the high complication rate reported following open reduction of 13 similar late-presenting fractures by Topouchian et al,9 in which none achieved good or excellent results and 4 developed avascular necrosis. The safety of the percutaneous reduction technique has also been reported by others.11,15 Thus, our data support attempting percutaneous reduction prior to proceeding to open reduction in late-presenting fractures. Two patients developed flexion contractures with anatomical fracture alignment, likely due to soft tissue contracture. Because of the potential shortening of the volar plate and the tethering of the extensor hood with retrograde crossed K-wires, we prefer to stabilize fractures with the adjacent interphalangeal joint maximally extended, although some flexion is required in order to achieve reduction and to penetrate the distal fragment. Alternative strategies exist for the treatment of late-presenting phalangeal neck fractures. Simmons and Peters18 described subcondylar fossa reconstruction for sagittal malunions in hyperextension, although this technique only improves adjacent interphalangeal joint flexion without correcting coronal or rotational malalignment. Nonoperative treatment to allow for fracture remodeling was initially discredited,2,3 but multiple authors have published case reports of children completely remodeling phalangeal neck fractures.19e22 However, remodeling cannot restore coronal or rotational alignment and should be considered a viable treatment choice only if there is good rotational and coronal alignment, a
DISCUSSION Phalangeal neck fractures are relatively common in children1 and can lead to poor functional results if not reduced and stabilized.2e5 Thus, these fractures require prompt reduction and fixation.2,6e9 However, the best method for reduction of these fractures has remained controversial. Whereas several authors recommend closed reduction,6,7,9 proponents of open reduction cite the difficulty in achieving satisfactory reduction by closed means in all cases,2,4,14 especially because such fractures often present after fracture healing has begun. Al-Qattan2 reported good to excellent results in 21 of 22 patients treated with ORPP for type II phalangeal neck fractures; however, he reported 1 case of avascular necrosis. Another study reported 82% good to excellent results following open reduction and internal fixation in 17 type II phalangeal neck fractures.8 In a series by Topouchian et al,9 however, all 13 patients treated with open reduction and internal fixation for a phalangeal neck fracture had fair to poor results, and 4 of the 13 showed evidence of avascular necrosis. Thus, open reduction is not without risk, especially in the late-presenting fractures for which closed reduction is impossible. For this reason, Waters et al11 described percutaneous reduction as an alternative to open reduction when closed reduction was not successful, with a theoretically lower risk of avascular necrosis compared with open reduction. In their study, excellent results including the absence of avascular necrosis were achieved in all 8 patients who presented an average of 16 days after their injury.11 Londner et al15 reported good to excellent results in all 13 patients treated with PRPP, with no avascular necrosis. The findings of the current study support the use of a stepwise algorithm for the treatment of pediatric phalangeal neck fractures rather than arguing in favor of one reduction technique over another. Although the current study cannot differentiate specific fractures that should be treated with a particular technique of reduction, no fracture treated beyond 13 days after injury was successfully reduced by closed means alone. Therefore, surgeons treating fractures 3 J Hand Surg Am.
r
Vol. 39, March 2014
DISPLACED PEDIATRIC PHALANGEAL NECK FRACTURES
congruent, adjacent interphalangeal joint, achieved/ achievable union of fracture, substantial growth potential, tolerance of patient and family to limitation of adjacent joint motion during remodeling, and willingness of patient and family to be to tolerate a period of observation.13 We, therefore, recommend including nonoperative treatment in only selected cases as outlined in the algorithm in Figure 1. Our study has several limitations. Given that a single surgeon treated all fractures, the results may not be widely applicable. The technique of percutaneous reduction has been well-described11,15,23 and is not dissimilar to the technique of Kapandji for reducing distal radius fractures.24 However, a learning curve exists, and surgeons unfamiliar with the technique may need to proceed with open reduction. Second, as a retrospective study, there is the potential for bias. However, randomizing one third of patients to receive open reduction would have exposed a large number of patients to unnecessary operative risk. Although the uneven treatment arms may appear as a limitation, they convey the adherence and utility of the algorithm, as most patients were treated successfully with CRPP. Third, 8 of 61 patients were lost to follow-up. In addition, the relatively short follow-up for those that achieved early excellent results may preclude the later detection of subclinical avascular necrosis. Fourth, we hesitate to extrapolate our findings to adults, given the greater predisposition in adults to stiffness, often requiring rigid internal fixation and, thus, open reduction. Finally, the lack of need for open reduction in this study does not imply that all such fractures can be so managed.
6. Graham TJ, Hastings H. Fractures and dislocations of the hand and carpus in children. In: Gupta AK, Scheker LR, eds. The Growing Hand: Diagnosis and Management of the Upper Extremity in Children. London: Harcourt; 2000:591e607. 7. Graham TJW. Fractures and dislocations of the hand and carpus in children. In: Beaty JHK, ed. Rockwood and Wilkins’ Fractures in Children. Philadelphia: Lippincott Williams & Wilkins; 2001: 271e379. 8. Kang HJ, Sung SY, Ha JW, Yoon HK, Hahn SB. Operative treatment for proximal phalangeal neck fractures of the finger in children. Yonsei Med J. 2005;46(4):491e495. 9. Topouchian V, Fitoussi F, Jehanno P, Frajman JM, Mazda K, Pennecot GF. Treatment of phalangeal neck fractures in children: technical suggestion [in French]. Chir Main. 2003;22(6):299e304. 10. Dixon GL Jr, Moon NF. Rotational supracondylar fractures of the proximal phalanx in children. Clin Orthop Relat Res. 1972;83: 151e156. 11. Waters PM, Taylor BA, Kuo AY. Percutaneous reduction of incipient malunion of phalangeal neck fractures in children. J Hand Surg Am. 2004;29(4):707e711. 12. Tomaino MM. Percutaneous reduction of incipient malunion of phalangeal neck fractures. J Hand Surg Am. 2005;30(2):409, author reply 409e410. 13. Cornwall R, Waters PM. Remodeling of phalangeal neck fracture malunions in children: case report. J Hand Surg Am. 2004;29(3): 458e461. 14. Leonard MH. Open reduction of fractures of the neck of the proximal phalanx in children. Clin Orthop Relat Res. 1976;116:176e179. 15. Londner J, Salazard B, Gay A, Samson P, Legré R. A new technique of intrafocal pinning for phalangeal neck fractures in children [in French]. Chir Main. 2008;27(1):20e25. 16. Al-Qattan MM. Phalangeal neck fractures with concurrent vascular injury. J Hand Surg Eur Vol. 2009;34(1):104e109. 17. Al-Qattan MM. Nonunion and avascular necrosis following phalangeal neck fractures in children. J Hand Surg Am. 2010;35(8): 1269e1274. 18. Simmons BP, Peters TT. Subcondylar fossa reconstruction for malunion of fractures of the proximal phalanx in children. J Hand Surg Am. 1987;12(6):1079e1082. 19. Al-Qattan MM, Rasool MN, El Shayeb A. Remodelling in a malunited phalangeal neck fracture. Injury. 2004;35(11):1207e1210. 20. Hennrikus WL, Cohen MR. Complete remodelling of displaced fractures of the neck of the phalanx. J Bone Joint Surg Br. 2003;85(2):273e274. 21. Mintzer CM, Waters PM, Brown DJ. Remodelling of a displaced phalangeal neck fracture. J Hand Surg Br. 1994;19(5):594e596. 22. Puckett BN, Gaston RG, Peljovich AE, Lourie GM, Floyd WE III. Remodeling potential of phalangeal distal condylar malunions in children. J Hand Surg Am. 2012;37(1):34e41. 23. Waters PM, Bae DS. Problematic hand fractures. In: Waters PM, Bae DS, eds. Pediatric Hand and Upper Limb Surgery: A Practical Guide. Philadelphia: Lippincott Williams & Wilkins; 2012: 439e452. 24. Greatting MD, Bishop AT. Intrafocal (Kapandji) pinning of unstable fractures of the distal radius. Orthop Clin North Am. 1993;24(2): 301e307.
REFERENCES 1. Mahabir RC, Kazemi AR, Cannon WG, Courtemanche DJ. Pediatric hand fractures: a review. Pediatr Emerg Care. 2001;17(3):153e156. 2. Al-Qattan MM. Phalangeal neck fractures in children: classification and outcome in 66 cases. J Hand Surg Br. 2001;26(2):112e121. 3. Barton NJ. Fractures of the phalanges of the hand in children. Hand. 1979;11(2):134e143. 4. Leonard MH, Dubravcik P. Management of fractured fingers in the child. Clin Orthop Relat Res. 1970;73:160e168. 5. Hastings H II, Simmons BP. Hand fractures in children. A statistical analysis. Clin Orthop Relat Res. 1984;188:120e130.
J Hand Surg Am.
473
r
Vol. 39, March 2014
A Stepwise Algorithm for Surgical Treatment of Type II Displaced Pediatric Phalangeal Neck Fractures Jonas L. Matzon, MD, Roger Cornwall, MD
Purpose To evaluate a stepwise reduction algorithm including closed, percutaneous, and open reduction techniques followed by percutaneous pin fixation for displaced pediatric phalangeal neck fractures. Methods Sixty-one consecutive children (mean age, 9.4 y; range, 2e18 y) presenting with closed, type II displaced phalangeal neck fractures were treated using the following algorithm. If satisfactory reduction was achieved with closed reduction, percutaneous pinning (CRPP) was performed. If not achieved, then percutaneous reduction and pinning (PRPP) was performed using a temporary intrafocal joystick for reduction and for osteoclasis as needed. If percutaneous reduction failed, open reduction and percutaneous pinning (ORPP) was performed. Using the Al-Qattan system, radiographic and clinical outcomes were retrospectively graded for union, deformity, range of motion, and function. Results Satisfactory alignment was achieved in all 61 fractures, by CRPP in 49 and PRPP in 12. No fracture required open reduction. Mean number of days from injury to surgery was 8 days for CRPP and 17 days for PRPP. All fractures treated after 13 days required percutaneous pinning. Fifty-three patients were followed for at least 1 year or until full functional recovery was achieved, with 45 excellent, 4 good, 1 fair, and 3 poor results. Four complications accounted for the fair and poor results, including 1 flexion contracture, 1 nonunion following pin track infection, and 1 case of avascular necrosis following a severe crush injury in the CRPP group and 1 flexion contracture following PRPP. Conclusions Our stepwise algorithm for surgical treatment of closed, type II displaced pediatric phalangeal neck fractures produced 92% good to excellent results while minimizing the need for open reduction even in late-presenting fractures. (J Hand Surg Am. 2014;39(3):467e473. Copyright Ó 2014 by the American Society for Surgery of the Hand. All rights reserved.) Type of study/level of evidence Therapeutic IV. Key words Pediatric, percutaneous fixation, percutaneous reduction, phalangeal neck fractures.
From the Division of Orthopaedic Surgery, Children’s Hospital of Philadelphia, Philadelphia, Pennsylvania; and the Division of Orthopaedic Surgery, Cincinnati Children’s Hospital Medical Center, Cincinnati, Ohio. Received for publication July 23, 2013; accepted in revised form December 12, 2013. No benefits in any form have been received or will be received related directly or indirectly to the subject of this article. Corresponding author: Roger Cornwall, MD, Cincinnati’s Children’s Hospital Medical Center, Division of Orthopaedic Surgery, 3333 Burnet Ave., MLC 2017, Cincinnati, OH 45229; e-mail: [email protected]. 0363-5023/14/3903-0009$36.00/0 http://dx.doi.org/10.1016/j.jhsa.2013.12.014
P
account for 13% of finger fractures in children.1 Based on the AlQattan classification, phalangeal neck fractures can be nondisplaced (type I), displaced with boneto-bone contact at the fracture site (type II), and displaced with no bone-to-bone contact (type III).2 The majority of these fractures are displaced, and displaced fractures can cause considerable functional problems if not reduced anatomically.3e5 Similarly, reduced fractures have a strong tendency to redisplace unless stabilized by pin fixation.2,4 Therefore, HALANGEAL NECK FRACTURES
Ó 2014 ASSH
r
Published by Elsevier, Inc. All rights reserved.
r
467
468
DISPLACED PEDIATRIC PHALANGEAL NECK FRACTURES
current recommendations for the treatment of such fractures dictate timely reduction and pin fixation to stabilize the fracture.2,6e9 Although there is little question that pin fixation is required to stabilize the fracture once reduced, the best method for reduction is debatable. Some authors recommend open reduction,2,8,10 and others recommend closed reduction.6,7,9 Proponents of open reduction cite the inability to adequately correct the deformity by closed means in all cases, whereas proponents of closed reduction cite the risk of avascular necrosis of the phalangeal condyles following open reduction.9 A complicating factor is that many children present to the hand surgeon 2 to 3 weeks after injury. By then, closed reduction is often unsuccessful, but open reduction may carry an even greater risk of injuring the blood supply to the healing fracture fragments.9 For this reason, a percutaneous reduction technique has been described as an alternative to open reduction when closed reduction is unsuccessful, with a theoretically lower risk of avascular necrosis compared with open reduction.9,11 However, the roles and efficacies of the various reduction techniques have not been elucidated, as previous retrospective studies reporting the results of treatment of phalangeal neck fractures have included a variety of treatment strategies and/or a small number of patients,2e4,8e11 which has led to a lack of consensus regarding treatment strategies.12 The primary purpose of the study was to evaluate the effectiveness of a stepwise algorithm to obtain reduction incorporating closed, percutaneous, and open reduction techniques in a consecutive series of patients with closed, type II displaced phalangeal neck fractures. The secondary purpose was to report the clinical results of those patients treated with this algorithm.
in the operating room. If satisfactory reduction was achieved, percutaneous interfragmentary pins were placed to stabilize the fracture (Fig. 2). The size of the stabilizing K-wires was dependent on the child’s age and ranged from 0.7 to 1.1 mm. One or 2 K-wires were passed obliquely from distal to proximal to stabilize the fracture. A crossed pin configuration was used for transverse fractures, and a parallel or divergent configuration was used for oblique fractures. A single K-wire was used in very distal fractures or small fragments in which 2 pins could not be placed. The number of pins used for each patient was not recorded. Regardless of pin number or configuration, the pins were placed with the adjacent interphalangeal joint in as little flexion as possible to minimize the risk of a flexion contracture. If, however, closed reduction was unsuccessful at achieving satisfactory alignment, percutaneous reduction was performed using a temporary intrafocal K-wire for reduction and percutaneous osteoclasis as needed and as described by Waters et al.11 Specifically, a K-wire was placed by hand under fluoroscopic guidance into the fracture site at the concavity of the deformity. Depending on the direction of displacement (dorsal, dorsoradial, or dorsoulnar), the K-wire was placed either through or around the central extensor apparatus so that the K-wire was inserted into the concave side of the fracture in the plane of maximum deformity. A sweeping motion was used to disrupt fracture callus. Then the K-wire was driven by hand through the fracture site until the tip was just distal and palmar to the distal end of the far cortex of the proximal fragment, such that the palmar cortex of the proximal fragment could be used as a fulcrum for levering the distal fragment into position. The adjacent interphalangeal joint was simultaneously flexed as necessary to assist in the reduction. The size of the K-wire was dependent on the child’s age and ranged from 0.9 to 1.4 mm.11 If percutaneous reduction was successful, percutaneous interfragmentary pins were placed as described previously. Open reduction was reserved for fractures that could be reduced by neither closed nor percutaneous methods and when the criteria for remodeling were not met, as described by Cornwall and Waters13 (Fig. 1). Regardless of the method of reduction, the patients were immobilized in a cast following pinning. Patients were seen in the office at 4 to 5 weeks after surgery. If clinical and radiographic union was apparent, both the cast and the K-wires were removed, and range of motion exercises were started at home. Patients returned to the office at 6 to 7 weeks
MATERIALS AND METHODS Treatment algorithm Since 2004, the senior author (R.C.) has used a stepwise treatment algorithm for closed, type II displaced phalangeal neck fractures in children (Fig. 1). Non-displaced or minimally displaced fractures with normal adjacent interphalangeal joint motion and no coronal or rotational deformity were treated conservatively with close follow-up to ensure maintenance of satisfactory alignment until healing. Fractures with initial displacement sufficient to affect adjacent joint motion or produce coronal or rotational deformity were treated with an attempt at closed reduction J Hand Surg Am.
r
Vol. 39, March 2014
DISPLACED PEDIATRIC PHALANGEAL NECK FRACTURES
469
FIGURE 1: Minimally invasive stepwise treatment algorithm for displaced phalangeal neck fractures in children. *Fractures were considered healed if they were nontender and with no radiolucent fracture line visible, regardless of time from injury and amount of callus. **Success is based on no coronal or rotational deformity and normal adjacent interphalangeal joint motion. †See text for remodeling criteria used. zSalvage procedures include osteotomy and subcondylar fossa reconstruction.
after surgery for evaluation of range of motion. Patients with at least 90 of flexion and less than 10 of flexion contracture were discharged. If range of motion was limited, the patient was referred to occupational therapy and seen back in the office 4 weeks later.
treatment (surgical or nonsurgical) by another doctor, types I and III phalangeal neck fractures, open fractures, and failure to complete follow-up. Sample characteristics Outpatient and inpatient medical records were retrieved and reviewed for age, sex, location of fracture, mechanism of injury, date of injury, definitive treatment, follow-up, K-wire removal, type of definitive treatment, postoperative care, and referral to occupational therapy. There were 43 boys and 18 girls, with an average age of 9.4 years (range, 2e18 y). The most common injury mechanism was sports (n ¼ 31), followed by a fall (n ¼ 11), and crush injury in a door (n ¼ 8). The most frequently fractured finger was the little finger (n ¼ 40), followed by the ring finger (n ¼ 12), the middle
Study patients This retrospective case series of patients treated with a prospectively applied treatment algorithm was approved by our institutional review board. Sixty-one consecutive children treated surgically from 2004 to 2008 for closed, type II displaced phalangeal neck fractures using the stepwise algorithm were identified using Current Procedural Terminology billing codes. Patients were excluded from participation for age older than 18 years, a previous attempt at definitive J Hand Surg Am.
r
Vol. 39, March 2014
470
DISPLACED PEDIATRIC PHALANGEAL NECK FRACTURES
FIGURE 2: Anteroposterior A and lateral B radiographs of a 22-day-old incipient malunion. A K-wire is placed into the dorsal callus to perform osteoclasis C; the K-wire is pointed distally and introduced into the fracture site D; the K-wire is advanced until the tip rests at the dorsal end of the volar cortex of the proximal fragment, and the pin is used to lever the distal fragment into flexion E. F Following reduction, full interphalangeal joint flexion and disrupted dorsal callus are visible. The fracture is then secured with crossed K-wired G, H, which are removed 4 weeks later after radiographs demonstrate satisfactory healing I, J.
finger (n ¼ 9), the index finger (n ¼ 7), and the thumb (n ¼ 3). Forty-four fractures involved the proximal phalanx, and 17 involved the middle phalanx.
patient/parent satisfaction, and complaints of pain. For patients who achieved functional range of motion shortly after K-wire removal, long-term follow-up was not required. For patients who did not achieve functional range of motion during clinic follow-up, long-term follow-up was collected over the telephone and through examination in clinic. Data were collected regarding range of motion, deformity, pain, and dysfunction. Radiographic and clinical outcomes were retrospectively graded by the Al-Qattan system,2 including union, deformity, range of motion, and function (Table 1).
Radiographic and clinical outcomes Pre-, intra-, and postoperative radiographs were reviewed for direction and degree of displacement, location of fracture, quality of reduction, avascular necrosis, and ultimate deformity, as defined by AlQattan (Table 1). Preoperative sagittal angulation was not recorded owing to concomitant coronal plane angulation and difficulty reliably identifying the axis of the condyles seen in parallax. Postoperative sagittal alignment was measured and considered acceptable if the adjacent proximal interphalangeal joint has normal motion. Medical records were reviewed for functional outcomes data, including range of motion at early follow-up after K-wire removal, progress in occupational therapy, final range of motion, deformity, J Hand Surg Am.
Statistical analysis Nominal variables were compared using the Fisher exact test given the expectedly small sample sizes in the more invasive groups (percutaneous and open reduction). Continuous variables (assuming parametric distributions) were compared using 2-tailed t tests. Statistical significance was set at P less than .05. r
Vol. 39, March 2014
471
DISPLACED PEDIATRIC PHALANGEAL NECK FRACTURES
TABLE 1. Comparison of Al-Qattan (5) Outcomes Classification Between Closed Reduction and Percutaneous Reduction of Pediatric Displaced Phalangeal Neck Fractures Excellent
Good
Fair
Poor
Classification Union
Yes
Yes
Yes
Nonunion or avascular necrosis
50 at IP/DIP 50 e90 at PIP
< 50 at PIP
No
Mild
Severe
Good
Useful
Lost
38 (86%)
3 (7%)
1 (2%)
2 (5%)
7 (78%)
1 (11%)
0 (0%)
1 (11%)
Joint range of motion
Full
> 50 at IP/DIP > 90 at PIP
Residual deformity
No
Digit function
Normal
CRPP PRPP
Results
DIP, distal interphalangeal joint of fingers; IP, interphalangeal joint of thumb; PIP, proximal interphalangeal joint of fingers.
TABLE 2. Comparison of Patient and Injury Characteristics Between Closed Reduction and Percutaneous Reduction of Pediatric Displaced Phalangeal Neck Fractures CRPP Mean SD (Range)
PRPP Mean SD (Range)
Independent t Tests P Value
Age (y)
9.2 3.4 (2e17)
9.8 4.2 (3e18)
.61
Days to presentation
5.1 2.9 (0e12)
14.3 9.0 (3e35)
< .001*
Days to surgery
7.6 3.2 (0e13)
16.6 8.4 (5e36)
< .001*
*P < .05 indicates statistical significance.
RESULTS Sample characteristics Forty-nine patients were treated using closed reduction percutaneous pinning (CRPP), and 12 required percutaneous reduction percutaneous pinning (PRPP). Using our treatment algorithm, no patients required open reduction percutaneous pinning (ORPP). In all patients, a satisfactory reduction was achieved with no coronal or rotational deformity and with sagittal alignment allowing full adjacent interphalangeal joint range of motion (assessed after reduction but before fixation pin placement). Patients treated with the CRPP presented significantly earlier (P < .001) and had surgery sooner (P < .001) than patients treated with PRPP (Table 2), but both treatment groups were similar in age (P ¼ .61).
had not been discharged from clinic returned to clinic for a follow-up examination an average of 2.5 1.0 years (range, 1.0e4.2 y) after treatment. Eight patients (5 CRPP, 3 PRPP) were lost to follow-up and were excluded from radiographic analyses. Radiographs at final follow-up demonstrated anatomical alignment in 45 patients, acceptable alignment in 7 patients (no coronal or rotational deformity, with sagittal angulation ranging from 30 flexion to 5 extension but with normal adjacent interphalangeal joint motion), and an anatomical nonunion in 1 patient. Occupational therapy was prescribed for 15 patients, 8 of whom actually attended therapy, which was too small a number for powered statistical analysis. Of the 53 patients with complete follow-up, 49 (92%) had good to excellent results (Table 1). One patient (CRPP) had a fair result secondary to a proximal interphalangeal joint flexion contracture with final range of motion of 30 to 90 . The remaining 3 results were poor secondary to a proximal interphalangeal joint flexion contracture (PRPP) with final range of motion of 45 to 80 , a nonunion following a pin track infection (CRPP), and avascular
Radiographic and clinical outcomes Patients were followed until discharged from clinic or by phone after a minimum of 1 year. Thirty-four patients achieved full range of motion by an average of 3 3 months (range, 1e17 mo) and were discharged from clinic. An additional 19 patients who J Hand Surg Am.
r
Vol. 39, March 2014
472
DISPLACED PEDIATRIC PHALANGEAL NECK FRACTURES
necrosis (CRPP) after a crush injury with multiple associated fractures. The latter patient was symptomatic, and avascular necrosis was confirmed on serial radiographs. No difference in outcomes could be detected between the CRPP and the PRPP groups, but this study did not have adequate power to detect a difference.
or more weeks after injury should be familiar with techniques of percutaneous and open reduction. Similarly, although the current study cannot determine the maximum delay from injury to treatment after which percutaneous reduction will be futile, this technique was successful beyond 5 weeks in our series (36 days). Therefore, we recommend considering percutaneous reduction for any fracture that is not yet fully healed, as determined by the presence of tenderness and a radiolucent fracture line, regardless of the amount of callus or the time from injury. Proposed risk factors for concurrent avascular necrosis include having a crush injury; presenting with joint stiffness, ischemia, or congestion; fragmenting the head during closed reduction; and passing the Kwire multiple times through the bone.16,17 The one such risk factor present in our case of avascular necrosis was crush injury. No case of avascular necrosis followed PRPP. The low complication rate among the 12 patients in the PRPP group, especially the absence of avascular necrosis despite being treated beyond 13 days after injury, is in contrast to the high complication rate reported following open reduction of 13 similar late-presenting fractures by Topouchian et al,9 in which none achieved good or excellent results and 4 developed avascular necrosis. The safety of the percutaneous reduction technique has also been reported by others.11,15 Thus, our data support attempting percutaneous reduction prior to proceeding to open reduction in late-presenting fractures. Two patients developed flexion contractures with anatomical fracture alignment, likely due to soft tissue contracture. Because of the potential shortening of the volar plate and the tethering of the extensor hood with retrograde crossed K-wires, we prefer to stabilize fractures with the adjacent interphalangeal joint maximally extended, although some flexion is required in order to achieve reduction and to penetrate the distal fragment. Alternative strategies exist for the treatment of late-presenting phalangeal neck fractures. Simmons and Peters18 described subcondylar fossa reconstruction for sagittal malunions in hyperextension, although this technique only improves adjacent interphalangeal joint flexion without correcting coronal or rotational malalignment. Nonoperative treatment to allow for fracture remodeling was initially discredited,2,3 but multiple authors have published case reports of children completely remodeling phalangeal neck fractures.19e22 However, remodeling cannot restore coronal or rotational alignment and should be considered a viable treatment choice only if there is good rotational and coronal alignment, a
DISCUSSION Phalangeal neck fractures are relatively common in children1 and can lead to poor functional results if not reduced and stabilized.2e5 Thus, these fractures require prompt reduction and fixation.2,6e9 However, the best method for reduction of these fractures has remained controversial. Whereas several authors recommend closed reduction,6,7,9 proponents of open reduction cite the difficulty in achieving satisfactory reduction by closed means in all cases,2,4,14 especially because such fractures often present after fracture healing has begun. Al-Qattan2 reported good to excellent results in 21 of 22 patients treated with ORPP for type II phalangeal neck fractures; however, he reported 1 case of avascular necrosis. Another study reported 82% good to excellent results following open reduction and internal fixation in 17 type II phalangeal neck fractures.8 In a series by Topouchian et al,9 however, all 13 patients treated with open reduction and internal fixation for a phalangeal neck fracture had fair to poor results, and 4 of the 13 showed evidence of avascular necrosis. Thus, open reduction is not without risk, especially in the late-presenting fractures for which closed reduction is impossible. For this reason, Waters et al11 described percutaneous reduction as an alternative to open reduction when closed reduction was not successful, with a theoretically lower risk of avascular necrosis compared with open reduction. In their study, excellent results including the absence of avascular necrosis were achieved in all 8 patients who presented an average of 16 days after their injury.11 Londner et al15 reported good to excellent results in all 13 patients treated with PRPP, with no avascular necrosis. The findings of the current study support the use of a stepwise algorithm for the treatment of pediatric phalangeal neck fractures rather than arguing in favor of one reduction technique over another. Although the current study cannot differentiate specific fractures that should be treated with a particular technique of reduction, no fracture treated beyond 13 days after injury was successfully reduced by closed means alone. Therefore, surgeons treating fractures 3 J Hand Surg Am.
r
Vol. 39, March 2014
DISPLACED PEDIATRIC PHALANGEAL NECK FRACTURES
congruent, adjacent interphalangeal joint, achieved/ achievable union of fracture, substantial growth potential, tolerance of patient and family to limitation of adjacent joint motion during remodeling, and willingness of patient and family to be to tolerate a period of observation.13 We, therefore, recommend including nonoperative treatment in only selected cases as outlined in the algorithm in Figure 1. Our study has several limitations. Given that a single surgeon treated all fractures, the results may not be widely applicable. The technique of percutaneous reduction has been well-described11,15,23 and is not dissimilar to the technique of Kapandji for reducing distal radius fractures.24 However, a learning curve exists, and surgeons unfamiliar with the technique may need to proceed with open reduction. Second, as a retrospective study, there is the potential for bias. However, randomizing one third of patients to receive open reduction would have exposed a large number of patients to unnecessary operative risk. Although the uneven treatment arms may appear as a limitation, they convey the adherence and utility of the algorithm, as most patients were treated successfully with CRPP. Third, 8 of 61 patients were lost to follow-up. In addition, the relatively short follow-up for those that achieved early excellent results may preclude the later detection of subclinical avascular necrosis. Fourth, we hesitate to extrapolate our findings to adults, given the greater predisposition in adults to stiffness, often requiring rigid internal fixation and, thus, open reduction. Finally, the lack of need for open reduction in this study does not imply that all such fractures can be so managed.
6. Graham TJ, Hastings H. Fractures and dislocations of the hand and carpus in children. In: Gupta AK, Scheker LR, eds. The Growing Hand: Diagnosis and Management of the Upper Extremity in Children. London: Harcourt; 2000:591e607. 7. Graham TJW. Fractures and dislocations of the hand and carpus in children. In: Beaty JHK, ed. Rockwood and Wilkins’ Fractures in Children. Philadelphia: Lippincott Williams & Wilkins; 2001: 271e379. 8. Kang HJ, Sung SY, Ha JW, Yoon HK, Hahn SB. Operative treatment for proximal phalangeal neck fractures of the finger in children. Yonsei Med J. 2005;46(4):491e495. 9. Topouchian V, Fitoussi F, Jehanno P, Frajman JM, Mazda K, Pennecot GF. Treatment of phalangeal neck fractures in children: technical suggestion [in French]. Chir Main. 2003;22(6):299e304. 10. Dixon GL Jr, Moon NF. Rotational supracondylar fractures of the proximal phalanx in children. Clin Orthop Relat Res. 1972;83: 151e156. 11. Waters PM, Taylor BA, Kuo AY. Percutaneous reduction of incipient malunion of phalangeal neck fractures in children. J Hand Surg Am. 2004;29(4):707e711. 12. Tomaino MM. Percutaneous reduction of incipient malunion of phalangeal neck fractures. J Hand Surg Am. 2005;30(2):409, author reply 409e410. 13. Cornwall R, Waters PM. Remodeling of phalangeal neck fracture malunions in children: case report. J Hand Surg Am. 2004;29(3): 458e461. 14. Leonard MH. Open reduction of fractures of the neck of the proximal phalanx in children. Clin Orthop Relat Res. 1976;116:176e179. 15. Londner J, Salazard B, Gay A, Samson P, Legré R. A new technique of intrafocal pinning for phalangeal neck fractures in children [in French]. Chir Main. 2008;27(1):20e25. 16. Al-Qattan MM. Phalangeal neck fractures with concurrent vascular injury. J Hand Surg Eur Vol. 2009;34(1):104e109. 17. Al-Qattan MM. Nonunion and avascular necrosis following phalangeal neck fractures in children. J Hand Surg Am. 2010;35(8): 1269e1274. 18. Simmons BP, Peters TT. Subcondylar fossa reconstruction for malunion of fractures of the proximal phalanx in children. J Hand Surg Am. 1987;12(6):1079e1082. 19. Al-Qattan MM, Rasool MN, El Shayeb A. Remodelling in a malunited phalangeal neck fracture. Injury. 2004;35(11):1207e1210. 20. Hennrikus WL, Cohen MR. Complete remodelling of displaced fractures of the neck of the phalanx. J Bone Joint Surg Br. 2003;85(2):273e274. 21. Mintzer CM, Waters PM, Brown DJ. Remodelling of a displaced phalangeal neck fracture. J Hand Surg Br. 1994;19(5):594e596. 22. Puckett BN, Gaston RG, Peljovich AE, Lourie GM, Floyd WE III. Remodeling potential of phalangeal distal condylar malunions in children. J Hand Surg Am. 2012;37(1):34e41. 23. Waters PM, Bae DS. Problematic hand fractures. In: Waters PM, Bae DS, eds. Pediatric Hand and Upper Limb Surgery: A Practical Guide. Philadelphia: Lippincott Williams & Wilkins; 2012: 439e452. 24. Greatting MD, Bishop AT. Intrafocal (Kapandji) pinning of unstable fractures of the distal radius. Orthop Clin North Am. 1993;24(2): 301e307.
REFERENCES 1. Mahabir RC, Kazemi AR, Cannon WG, Courtemanche DJ. Pediatric hand fractures: a review. Pediatr Emerg Care. 2001;17(3):153e156. 2. Al-Qattan MM. Phalangeal neck fractures in children: classification and outcome in 66 cases. J Hand Surg Br. 2001;26(2):112e121. 3. Barton NJ. Fractures of the phalanges of the hand in children. Hand. 1979;11(2):134e143. 4. Leonard MH, Dubravcik P. Management of fractured fingers in the child. Clin Orthop Relat Res. 1970;73:160e168. 5. Hastings H II, Simmons BP. Hand fractures in children. A statistical analysis. Clin Orthop Relat Res. 1984;188:120e130.
J Hand Surg Am.
473
r
Vol. 39, March 2014
