Pediatric Hand Fracture Referring Practices: A Scoping Review

Abstract

Background:

Pediatric hand fractures are common, but few require surgery; therefore, these fractures are often perceived to be overreferred. Our objective is to systematically identify and describe pediatric hand fracture referring practices.

Method:

A scoping review was performed, searching electronic databases and grey literature up to January 2018 to identify referring practices for children (17 years and younger) with hand fractures (defined as radiographically confirmed fractures distal to the carpus) to hand surgeons. All study designs were included, and study selection and data extraction were independently performed in duplicate by 2 reviewers. Outcomes included referring rates, necessity of referral, referring criteria, and management of fractures.

Results:

Twenty (10 cross-sectional, 7 prospective cohorts, and 3 narrative reviews) studies reporting on referring practices or management of 21,624 pediatric hand fractures were included. Proportion of pediatric hand fractures referred to hand surgeons ranged from 6.5% to 100%. Unnecessary referral, defined as those fractures within the scope of primary care management, ranged from 27% to 78.1%. Ten studies reported referring criteria, with 14 unique criteria identified. The most common referring criteria were displacement (36.4%), loss of joint congruity (36.4%), and instability (36.4%). The most common justification for these criteria was increased likelihood of requiring surgery. The most common initial management was immobilization (66%-100%). Final management was provided by orthopedic or plastic hand surgeons with 0% to 32.9% of fractures requiring surgery.

Conclusion:

Referring practices vary widely in the literature. Major gaps in the literature include objective measures and justification for referring criteria and primary care education on hand fracture referring practices.

Introduction

Pediatric hand fractures are common, with reported incidences of up to 624 pediatric hand fractures per 100 000 patients per year.¹ The hand is the most frequently injured body part in children.^2-5 Hand fractures are the second most common fracture, accounting for one-fifth of all pediatric fractures.^2-5 Incidence of hand fractures increases with age, with peak occurrence coinciding with the onset of puberty and an increased participation in contact sports.² The growing skeleton has a unique capacity to heal and remodel and the majority of pediatric hand fractures are managed by immobilization with only approximately 10% requiring surgery.^3,4,6-13

From the perspective of the referring physician, there are 2 common concerns when managing pediatric hand fractures: diminished hand function and disturbance of the growth plate. Injuries to the hand can result in major functional deficits which may affect expression, work, and play. Before skeletal maturity, the growth plate, or physis, is open and vulnerable to injury. The physis represents an area of weakness compared to the surrounding bone because it is unmineralized and the chondrocytes are vulnerable to disruption through shearing or angular forces.¹⁴ While disturbance of bone growth is a major concern for referring physicians, this rarely becomes a real issue, except with Salter Harris type 5 fractures which makeup less than 1% of growth plate fractures.¹⁵ However, type 5 injuries are difficult to diagnosis on radiograph and a high degree of clinical suspicion is necessary.

From the perspective of the hand surgeon, pediatric hand fractures may represent a potential misuse of resources because, while common, few require surgery.^3,4,6-13 In contrast to adults, hand fractures in children have low morbidity,^3-6 as children display quicker recovery from hand injuries and have a low frequency of nonunion.^13,16 For these reasons, pediatric hand fractures are often perceived to be overreferred. The underlying assumption is that similar and more appropriate care can be delivered outside of a surgeon’s office at a lower cost. Unnecessary surgeon visits can be a burden to both the health-care system and the patient as these visits place strain on wait-lists, increase hospital costs, and require patient and care giver time. One study of outpatient upper limb fracture pediatric patients quantified the direct cost of attending clinic in 2016 to be equivalent to US$60.24 and the cost to society due to productivity loss to be US$97.37 per consultation. Moreover, although subjectively patients deemed the appointments necessary, only 12.6% of appointments resulted in a change of management.¹⁷ Another study found for every 100 appointments, 25 working days were lost and 18 patients lost wages. Among 71 patients, 54 days of school were missed.¹⁸

Despite the high incidence of pediatric hand fractures, there is limited evidence for when to refer these fractures and there remains discrepancy between referring and consultant physicians. The American Academy of Pediatric guidelines recommends that complex fractures be referred to a surgeon^19,20; however, the term “complex” is ambiguous and potentially difficult to interpret for a physician without specialized hand training. When looking at the adult literature, referring guidelines are also limited, and moreover, application from adult to children may not be appropriate; the pediatric hand is a growing hand and children have developmental constraints that may change splinting and therapy management.

To our knowledge, there is no review that systematically summarizes the evidence regarding pediatric hand fracture referring practices. We used a scoping review methodology to understand the range and types of referring practices. We included both referring and consultant perspectives to inform the future design of pediatric referring recommendations. Specifically, this study’s purpose is to identify and describe referring practices, from primary care physicians to hand surgeons, of radiographically confirmed hand fractures in children 17 years and younger.

Methods and Materials

A scoping review was conducted to capture a broad range of studies and to summarize knowledge from a variety of sources and types of evidence. Using the Joanna Briggs Institute-framework for scoping reviews, we conducted the following: (1) identified the research question, (2) identified relevant studies, (3) completed study selection, (4) charted the data, and (5) collated, summarized, and reported the results.²¹ In the spirit of a scoping review, this was an iterative process to ensure comprehensive inclusion of literature. This work involved identifying, reviewing, and categorizing data from primary articles and did not involve human participants and was thus exempt from ethics approval.

A broad search of the medical literature to identify all studies and journals that focused on pediatric hand fracture referring practices was conducted including the following online databases: EMBASE, MEDLINE (OVID), PubMed, Cochrane Central Register of Controlled Trials, and the Cochrane Database of Systematic Reviews with no limitations on date (inception to January 3, 2018) or language of publication. No search filters were used. The search strategy was repeated on Google and Google Scholar searching through the first 5 pages of results for grey literature. The grey literature search strategy was completed within a single session for each search strategy due to the dynamic nature of the Internet. Bibliographies of included studies were searched by hand for other relevant articles. Expert consultation (F.F. and A.H.) was sought to identify any potential ongoing or unpublished studies.

Searches included 2 themes: (1) hand fracture and (2) triage or referring practices. These 2 themes were combined using the Boolean operator “and.” The search strategy was approved by a Health Sciences Librarian (See Online Appendix 1 for Search Strategy).

Level 1 screening was performed for title and abstract. Articles were included if they reported on referring practices or management of hand fractures. Management was purposely included at the level 1 screening because depending on the article perspective, for example, from a primary care or emergency department perspective, referral can be considered management. Nonhuman studies were excluded. The search was not limited by age at this level. All types of articles, including observational, trials, and reviews were considered.

Articles passing the level 1 screening underwent a level 2 screening for full-text review. Articles identified through grey literature and by local specialists were included at this stage. Eligible articles were included for data extraction if they reported on referring practices of hand fractures (defined as distal to the carpus) in patients 17 years or younger. Referrals to hand surgeons included both orthopedic and plastic surgeons. Hand surgery represents an overlap in scope of practice that is dependent of geography; in Canada, hand fractures are primarily managed by plastic surgeons whereas, for example, in the United States, hand fractures are primarily managed by orthopedic surgeons. Articles were excluded if they reported exclusively on adult patients without any pediatric patients or if they reported on hand injuries without fractures.

If a study presented unclear data, the corresponding author was e-mailed and a follow-up e-mail was sent after 2 weeks for nonresponders. Specific reasons for exclusion and inter-rater reliability (calculated using Cohen κ statistic) were determined at the full-text screening phase.

The number of studies included and excluded was summarized using a Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) flowchart. A data collection form based on the Cochrane Effective Practice and Organization of Care (EPOC) was created to ensure that we were comprehensive in extracting relevant study characteristics and included: general information, eligibility, population and setting, methods, participants, outcomes, results, and applicability.²² Our primary outcome was referring practices, that is: which hand fractures were referred, who referred them, and where they were referred to. Screening and data extraction were performed in duplicate by 2 independent reviewers (R.H. and A.T.). Disagreements were resolved by discussion and obtaining consensus between the 2 reviewers. Data were categorized and reported descriptively. Once collected, data were discussed among reviewers and categorized by themes. As this was a scoping review, no formal quality assessment was required. This article does not contain any studies with human or animal subjects.

Results

The search strategy identified 675 potential studies and after duplicates were removed, 420 were eligible for level 1 screening (Figure 1). Grey literature search and expert consultation produced 3 additional studies. Twenty seven studies passed title and abstract screening and underwent full-text review. Twenty original studies on pediatric hand fracture referring practices were included in final data extraction and the level 2 screening Cohen κ statistic was 0.823.

Figure 1.

Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) flow diagram.

A summary of the study characteristics included in this scoping review is provided in Table 1. The most common study designs were: cross-sectional (50.0%), prospective cohort (35.0%), and narrative review (15.0%). The studies were conducted in 6 countries including: United Kingdom (35.0%), United States (30.0%), and Canada (15.0%). Fourteen (70.0%) studies were published after 2000. Extracted data were organized into 3 main identifiable themes: (1) referring rates and necessity of referrals, (2) referring criteria, and (3) primary care education. One study reported on both referring rates and necessity of referrals and referring criteria; therefore, 21 study themes are reported.²³

Table 1.

Characteristics of the Studies Included in Scoping Review.

Characteristic	Studies (N = 20)
Study design
Cross-sectional study	10
Prospective cohort	7
Narrative review	3
Origin of study
United Kingdom	8
United States	6
Canada	3
Netherlands	1
Saudi Arabia	1
Switzerland	1
Year of publication
2009-2017	8
2000-2008	6
Prior	6
Themes	N = 21
Referring rates and necessity of referral	13
Referring criteria	5
Primary care education on pediatric hand fractures	3

Fourteen of 20 studies reported exclusively on pediatric patients, while 6 studies reported on adult and pediatric patients. The mean age reported across the pediatric only studies was 7.8 to 11.5 years old. The 20 included studies represented 2 624 hand fractures. One study reported from both the emergency department and a plastic surgery clinic; therefore, 21 locations are reported.²⁴ The most common study location was a surgeon’s clinic (13/21 studies), with 7 studies reported from an orthopedic clinic and 6 studies reported from a plastic surgery clinic. Seven studies reported from the emergency department and one from a community primary care clinic. Patient and fracture demographics across studies were comparable. Single, isolated fractures affecting the fifth digit in male participants from 10 to 15 years of age were most common.^{3,11,12,25-27}

Theme 1: Referring Rates and Necessity of Referrals

Table 2 summarizes the referring rates, necessity of referral, and management. Depending on authorship, referring rates varied: Primary care physicians reported referring rates to hand surgeons of 6.5% to 82.6%^8,28-31 and hand surgeons reported referring rates from primary care physicians of 35% to 100%.^{3,11,23,24,27,32} Five studies commented on necessity of referrals. Of those referred, 27% to 78% were deemed unnecessary.^25,26,33,34 Unnecessary referrals were hand fractures within the scope of a primary care physician that did not require management by a hand surgeon; the definition of within the scope of a primary care physician was unique to each study. Ramasubbu et al found that 41% of hand fractures and 38% of finger fractures referred to their clinic were unnecessary as there was no fracture present on radiograph.²⁶ Hsu et al found 22% of phalangeal fractures and 70% of metacarpal fractures referred were unnecessary.³⁴ Hsu et al categorized necessity of referrals as per the American Board of Pediatrics and American Academy of Pediatric Surgery criteria, where nondisplaced phalangeal and metacarpal fractures fall under the realm of primary care.^19,20 The most common management by hand surgeons was immobilization, ranging from 66% to 100%.^{3,10,11,12,24,26} Surgical management ranged from 0% to 32.9%.^{3,10,11,12,24-26,34} Intra-articular fractures were commonly reported as fractures that required surgery. Davis et al found 44% of intra-articular fractures required secondary operative fixation or splint adjustment,³² and in the study by Stanton et al, intra-articular fractures accounted for 17% of all fractures and 26% of fractures that required surgery.⁶ Among the intra-articular fractures, unstable proximal interphalangeal joint fractures were noted by studies at increased risk of surgery.^{3,10,12,24,28,33} Associated soft-tissue injuries, such as amputations⁸, subluxation or dislocation,¹⁰ and neurovascular injuries¹¹ were also reported along with fractures that required surgery.

Table 2.

Referral and Management.

	Sample Size (Hand Fractures)	Referral Rate (%)		Referral Criteria for Fractures	Management (%)
Author		Total	Inappropriate		Immobilization	Surgery
Ablove 199835	NR	NR	NR	Open, intra-articular, displaced or associated with amputation	NR	NR
Al-Qattan 2015	NR	NR	NR	Any involvement of phalangeal neck	NR	NR
Anzarut 200723	281	NR	NR	NR	NR	NR
Bhende 19938	92	82.6	NR	Open, displaced, complex closeda or association with amputation or nail bed laceration	95.7	4.3
Davis 199032	678	100	27	All hand fractures should be referred	NR	2.9
Gornitzky 201625	73	NR	57	NR	NR	27.8
Hill 199829	632	30.7-31.3	NR	NR	NR	NR
Hsu 201234	529	NR	30-78.1	American Board of Pediatrics and American Academy of Pediatric Surgery Criteriab	NR	7.2
Larsen 200430	28 231	41	NR	NR	NR	NR
Leonard 197010	263	NR	NR	NR	75	10
Liu 201411	83	NR	NR	Unstable, rotated, displaced, intra-articular, compound or short	84	12.0
Mahabir 200112	232	NR	NR	Open, unstable or significant malrotation deformity	67.1	32.9
Murphy 201033	1875	NR	>50	NR	NR	NR
Nofsinger 200228	NR	NR	NR	Any at “high risk for surgery”c	NR	NR
Ramasubbu 201626	21	NR	38.1	NR	100	0
Stanton 20076	183	NR	NR	NR	NR	15
Usal 199231	26	6.5	NR	NR	NR	NR
Vadivelu 200627	236	35	NR	NR	NR	NR
Weber 200924	37	NR	NR	PIPJ fractures that are unstable on passive testing, associated with dislocations, have fragments >1/3 articular surface or present late (>5 days)	100	0d
Young 2013	303	NR	NR	NR	66	5

Abbreviations: NR, not recorded; PIPJ, proximal interphalangeal joint; UCL, ulnar collateral ligament.

^a Not defined by authors.⁸

^b The American Board of Pediatrics: nondisplaced fractures not requiring reduction or long casts and finger fractures treated with buddy taping or splinting only are within the scope of primary care practice.¹⁹ The American Academy of Pediatric Surgery Advisory Panel Guidelines for Referral recommends any child with complex fractures and dislocations be referred.²⁰

^c High risk: base of thumb fractures, UCL avulsion fractures or UCL injuries with gross instability, phalangeal neck fractures, displaced middle phalangeal fractures, severe distal phalangeal crush injuries, and distal phalangeal injuries with tendon involvement.²⁸

^d Two patients with avulsion fractures of the volar plate with large epiphyseal fragments and subluxation of the PIP had open reduction internal fixations with K-wires and were excluded from the study.²⁴

Theme 2: Referring Criteria

Ten studies reported on referring criteria to a hand surgeon with 14 unique criteria reported (Table 3). The most common criteria included loss of joint congruity (ie, when the fracture involves the joint and fracture fragments are displaced; 36.4%), displacement (when the fracture fragments are not well-aligned in anatomical position; 36.4%), and instability (fractures that continue to be displaced or mal-rotated after a closed reduction; 36.4%). Four studies justified their referring criteria by recommending referral of fractures with a greater likelihood of requiring surgery, the definition of which varied for each study.^10-12,28 Only one study encouraged an increase in referring practices; Davis et al found 25% of emergency department hand fracture management to be unsatisfactory and subsequently concluded that all hand fractures should be referred to a surgeon’s clinic.³² Only one study was authored by a primary care physician.⁸ The most commonly reported criteria—loss of joint congruity and displacement—were each a criterion in only 4 studies. There were 8 criteria that were recommended by only one study each.

Table 3.

Referring criteria, with definition, used by each study.

Author	Displaced	Joint Incongruity	Unstable	Open	Specific Pattern	Amputated	Malrotated	All	Angulated	Compound	Dislocated	Late	Short
	Fragments Not Aligned	Involves Joint and Fragments Displaced	Continued Displacement After CR	Soft Tissue Deficit Over Fracture	As Defined by Authora	Associated With Missing Part of Digit	Rotation on Physical Exam	Any Hand Fracture	Rotation on X-Ray	Fragment Comminution	Associated With Joint Dislocation	Presents After 5+ Days	Decreased Length of Digit
Primary care
Bhende 19938	X			X	X	X
Hand surgeon
Ablove 199835	X	X		X		X
Al-Qattan 2015					X
Davis 199032								X
Hsu 201234	X
Leonard 197010		X	X						X
Liu 201411	X	X	X				X			X			X
Mahabir 200112			X	X			X
Nofsinger 200228					X
Vadivelu 200627
Weber 200924		X	X								X	X
TOTALS	4	4	4	3	3	2	2	1	1	1	1	1	1

Theme 3: Primary Care Education

Three studies reported on primary care education for pediatric hand fractures. Two were narrative reviews^28,32 and one was a cross-sectional study.²³ The narrative reviews were authored by hand surgeons and employed a broad approach to hand fractures. One included how to take a hand history and physical examination.³⁵ Both narratives described hand anatomy, the Salter-Harris classification, and common imaging of the hand. Both described selective management of common hand fractures, such as Boxer fractures and Seymour fractures, including when to refer. One review warned about potential bad outcomes and complications as means for referral justification.²⁸

The cross-sectional study surveyed primary care physicians referring patients to plastic surgery.²³ This study was focused on primary care continuing medical education of plastic surgery topics, and its aim was to assess educational needs. Of all plastic surgery referrals, trauma and upper extremity injuries were most common. Hand fractures were identified in both utilization and perceived need’s assessments as areas that primary care physicians required continuing education.

Discussion

This review found referring practices for pediatric hand fractures were not well described in the literature and had considerable variability. This may be related to lack of standardized guidelines for when to refer pediatric hand fractures and individual institute’s referring culture. Our findings are similar to prior studies reporting that few pediatric hand fractures require surgical management and referring patterns for hand fractures have not been driven by evidence.

Regarding referring criteria, there were 6 criterion that were recommended by only one study each demonstrating a lack of consensus among studies. Some criteria were rater-dependent. For example, displacement, defined as fracture fragments that are not well-aligned in anatomical position, is rater-dependent. Where one physician’s interpretation of malaligned may be 1 mm of displacement or more, another physician’s interpretation may be 3 mm or more. Moreover, radiographic measurement of fracture displacement has been proven to have poor inter-rater reliability.³⁶ There was a paucity of referring criteria authored by primary care physicians and the discrepancy of criteria when categorized by authorship represents a disconnect between referring and consulting doctors. These 2 results may account for the high percentage of referrals deemed unnecessary by hand surgeons.

The most common justification for referring criteria was increased likelihood of requiring surgery. An important consideration is whether “likelihood of requiring surgery” is an adequate outcome measure. While it is important to refer fractures that will require surgery, there may be other nonsurgical fractures that warrant specialty care. For example, a fracture that initially appears anatomically aligned, but due to location and fracture pattern is prone to slipping may be best managed with close follow-up, serial radiographs, and clinical examination. Distinguishing these fractures from stable, nondisplaced fractures can be difficult. These fractures should be referred to a hand surgeon, but initial radiographs can appear innocuous to the nonspecialist. Thus, a more inclusive definition than increased likelihood of requiring surgery should be considered when justifying referring practices of pediatric hand fractures.

There were only 3 studies whose theme focused on education for primary care physicians on pediatric hand fractures. These were authored by hand surgeons and only one study incorporated primary care physicians’ perspectives. Discussion by an integrated management team with input from both referring and consulting physicians as well as patients might result in development of a better approach to “whole person care” of this patient group. This study highlights the need for more research into efficient and effective management of pediatric hand fractures, including educational and resource support for referring physicians.

Thus, pediatric hand fractures referring practices represent a potential area for investigation and research. This study addresses important knowledge gaps in terms of optimal utilization of health-care resources and manpower, which is in alignment with the national Choosing Wisely Canada campaign,³⁷ whereby clinicians and patients are engaged in conversations pertaining to unnecessary diagnostic tests, referrals and treatments, and encourage health-care professionals to make smart and effective care choices.

Strengths of our study include the comprehensive nature of our search and inclusion of all study designs. There are 2 main limitations of this review: first, this review cannot provide recommendations for “best” referring criteria as the individual studies included in this review did not evaluate the necessary outcomes. The second is the inclusion of heterogenous study designs and outcomes. To address wide variance in outcome data, the 2 reviewers used the standardized EPOC tool to independently extract data.²² Quantitative data were also presented as ranges, as meta-analysis would not have been possible.

Conclusion

Overall, there is considerable variation in referring practices for pediatric hand fractures with respect to proportion referred, criteria for referring, and which referrals were unnecessary. Major gaps in the literature include objective parameters and evidence-based justification for referring criteria. Future research should be aimed at incorporating statistical analysis for the identification of objective parameters predictive of fractures that require specialist management.