Assessment of Splinting Quality: A Prospective Study Comparing Different Practitioners

Abstract

Background

Splinting is routinely performed in the emergency department (ED), and follow-up visits of improperly placed splints are commonplace in orthopaedic clinics. As open reduction and internal fixation (ORIF) of fractures has become the preferred treatment for many injuries, orthopaedic surgeons and emergency physicians have received less instruction on splinting technique. Limited literature exists regarding error/complication rates of splint application. The purpose of this study is to determine: (1) Is there a difference in splinting complication rates between orthopaedic and non-orthopaedic services, and low versus high volume emergency room and urgent care centers? (2) What are the most common technical errors and complications in splint application?

Methods

Patients presenting to orthopaedic clinic with any extremity splint were enrolled in this IRB approved prospective study. Splint characteristics collected included: type of provider placing the splint, duration of wear, type of splint, and material used (i.e. plaster or fiberglass). Errors included inappropriate length, circumferential placement, and direct contact between the ACE bandage and the skin; while complications included swelling, blistering, ulceration, heat injury, and other issues on a case-by-case basis.

Results

203 patients were enrolled in this study. 98 (48%) were splinted by the Orthopaedics service, 69 (34%) were splinted in the trauma hospital ED, and 36 (18%) were treated at an outside hospital. 123/203 (61%) had an error/ complication related to the splint. Error/complication rates for orthopaedics, the trauma hospital ED, and outside hospitals were 46% (45/98), 65% (45/69), and 92% (33/36) respectively. The most common errors were inappropriate length, present in 58/203 (29%) patients, and direct contact between the ACE bandage and skin, present in 50/203 (25%) patients.

Conclusion

The appropriateness and complication rates of splints applied in the ED differ based on the type of provider and the institution. Outside hospitals were found to have the highest complication rates, while the lowest rates were associated with splints placed by Orthopaedics. These findings support the importance of education of proper splinting technique in non-trauma hospitals.

Level of Evidence: III

Introduction

Splints are lightweight, noncircumferential immobilizers that accommodate extremity swelling, making them ideal for the management of acute musculoskeletal conditions.^1-3 Common indications for splinting include sprains, initial fracture stabilization, and postoperatively. In these situations, splinting serves to immobilize and protect the extremity, aid in healing, and lessen pain.

Splinting has potential complications including skin breakdown, ischemia, heat injury, infection, pressure sores, neurological injury, compartment syndrome, and fracture malreduction or non-union.^1,2,4-8 Further, errors in splint placement can place unnecessary restrictions on patient movement, limiting participation in work, exercise, and other activities. These complications can affect patients regardless of patient age or the duration of treatment.⁶ Patients at increased risk of complications include those who are obtunded or comatose, anesthetized, young, developmentally delayed, spastic, and/ or with pre-existing neuropathy.^2,3

Plaster splints have been used to treat fractures for greater than 150 years.⁹ However, as open reduction and internal fixation (ORIF) of fractures has become the preferred treatment for many orthopaedic injuries, orthopaedic surgeons and emergency physicians have received less instruction on proper splint application. In a recent study examining what procedures were considered necessary for graduating medical school students, only 31% of faculty and 26% of resident physicians rated splinting as a “must know” skill.¹⁰ Decreased instruction on proper technique due to this perception may increase the likelihood of erroneous application, thereby increasing the risk of complications. For example, a recent article assessing splinting in the emergency department found that prior to an educational intervention on proper technique the rate of correctly applied splints for hand injuries was 49%.¹¹

Improving splinting practice could lead to improved quality of care, decreased healthcare costs, and decreased societal costs (e.g. lost work hours, lost employment, and excess medical and drug costs). The current study aims to ascertain (1) is there a difference in splinting complication rates between orthopaedic and non-orthopaedic services, and low versus high volume emergency room and urgent care centers; and (2) what are the most common technical errors and complications in splint application.

Materials and Methods

Our Institutional Review Board approved this prospective study. An a-priori power analysis suggested that a total sample number of 155 patients would be necessary to detect a difference between groups based on the assumption of a 30% complication rate, a power of 80% (beta), and a significance level of 0.05 (alpha).

A single researcher evaluated all patients entering the orthopaedic clinic from May-June 2015 and November-December 2015. Patients were selected using a purposive sampling method based on the presence of any extremity splint. The researcher was not involved in patient care, and prospectively documented and observed patients with splints who presented to the general orthopaedic, pediatric orthopaedic, and hand clinics. Splint characteristics and the underlying skin and soft tissues were assessed for technical errors and complications, as described below. Injury type, splint characteristics (e.g. type and material, placing provider, dates of placement and removal, and days before evaluation), patient demographics (e.g. age and gender), and errors/ complications were recorded.

Definitions of proper splinting technique and errors utilized in this study were drawn from the text “Casts, Splints, and Support Bandages – Nonoperative Treatment and Perioperative Protection” for each injury and splint type.¹² Technical errors addressed included appropriateness of splint type for injury, splint length relative to anatomic landmarks, range of motion limitations, circumferential nature, and neutral positioning of the extremity. Complications assessed included numbness/ tingling, swelling secondary to the splint compression of an extremity, blistering, ulceration, heat injury, and compartment syndrome. Additionally, several errors and complications not discussed in the referenced text were addressed in an “Other” category, which included: direct application of the ACE bandage to the skin, excessive splint loosening, tissue maceration, and splint compression compromising positioning of the fingers or toes. Although ACE contact with the skin may be considered acceptable in some circumstances, it is not recommended on a general basis and has been utilized as an error criterion in other studies.^1,2,13

At the time of data collection, the trauma hospital was the only Level I trauma center in the region. In order to minimize bias, all data regarding date of placement, fracture or injury type, and service was obtained from the patient’s electronic medical record (EMR) after assessment. Data was divided into three groups for final analysis based on provider training and practice environment: “Orthopaedics” (orthopaedic surgery residents or attending physicians in the operating room or on consultation in the emergency department), “Trauma Hospital ED” (emergency physicians, emergency medicine resident physicians, or cast technicians), and “Outside Hospitals” (any practitioner performing the procedure at another facility).

This study was granted a waiver of written consent by the IRB. All patients eligible to participate in this study received a verbal description of the study methods, research question, and their involvement prior to their voluntary enrollment. Patients excluded from this study included those who declined to participate, those who had been previously enrolled in the study for another injury, and those who had altered the splint before its final documented removal.

Statistical analysis was performed using SPSS statistics version 23.0 (IBM, Armonk, New York). A two-tailed unpaired t-test was used to compare the splints placed by Orthopaedics in the operating room versus the emergency room. A chi-squared test with a post-hoc Bonferroni correction was used to assess if the error/ complication rate differed between services. With the Bonferroni correction a P value of less than 0.017 was deemed significant.

Results

A total of 203 patients were enrolled in the study. 98/203 (48%) were splinted by Orthopaedics, 69/203 (34%) were splinted by the Trauma Hospital ED, and 36/203 (18%) were treated at an Outside Hospital. There were 85 females (42%) and 118 males. The average age and standard deviation of patients enrolled in the study was 34 ± 22 years. Splint characteristics are listed in Table 1.

Table 1.

Patient Demographic and Splint Characteristic Data

Splint Characteristics	Number (%)
Service placing splint
Orthopaedics	98 (48)
Trauma Hospital ED	69 (34)
Other ED	36 (18)
Splint type
Long arm	34 (17)
Sugar tong	21 (10)
Short arm	22 (11)
Short arm thumb spica	10 (5)
Short arm to fingers	9 (4)
Radial gutter	2 (1)
Ulnar gutter	15 (7)
Long leg	11 (5)
Short leg	78 (38)
Removable finger splint	1 (0.5)
Splint material
Plaster	171 (84)
Fiberglass quickset	32 (16)
Average number of days from placement to removal	9

The overall error/complication rate was 61% (123/203), with specific issues listed in Table 2. The combined error/complication rates for Orthopaedics, the Trauma Hospital ED, and Outside Hospitals were 46% (45/98), 65% (45/69), and 92% (33/36) respectively. Splints placed by Orthopaedics in the operating room versus the emergency room did not differ significantly (P=0.31), and were combined into a single group. There was a significant difference in incorrectly placed splints and/ or complications between the three groups, with Orthopaedics having the fewest errors or complications (P<0.001). The average ages for patients splinted were 38.8 ± 19.2, 30.8 ± 24.7, and 23.8 ± 17.7 years for Orthopaedics, the Trauma Hospital ED, and Outside Hospitals, respectively. A chi-squared test with post-hoc Bonferroni correction showed that the Trauma Hospital ED was not significantly different from either group with regard to age (P(Orthopaedics) = 0.03 and P(OSH) = 0.1); however, the patients splinted by Orthopaedics were on average older than those patients seen at Outside Hospitals (P<0.001). Nevertheless, the calculated odds ratio indicated that patient age was not a significant predictor of splinting complications or errors (P=0.86, 95% confidence interval 0.99-1.01).

Table 2.

Errors and Complications Associated with Splints

Errors and Complications	All Services N = 203 (%)	Orthopaedics N = 98 (%)	Trauma Hospital ED N = 69 (%)	Outside Hospitals N = 36 (%)	P-value
Any error or complication	123 (61)	45a (46)	45b (65)	33c (92)		<0.0001a
Inappropriate splint type	14 (7)	0a (0)	6b (9)	8b (22)	<0.0001a
Inappropriate splint length	58 (29)	22 (22)	20 (29)	16 (44)	0.164
Crossing joint/Limiting ROM	49 (24)	21 (21)	16 (23)	12 (33)
Too Short	9 (4)	1 (1)	4 (6)	4 (11)
Circumferential (>2/3)	10 (5)	5 (5)	5 (7)	0 (0)	0.264
Swelling	7 (3.4)	2 (2)	2 (3)	3 (8)	0.199
Blistering	6 (3)	2 (2)	2 (4)	2 (3)	0.685
One	3 (1)	1 (1)	1 (1)	1 (3)
Multiple at one end	3 (1)	1 (1)	1 (1)	1 (3)
Multiple scattered blisters	0 (0)	0 (0)	0 (0)	0 (1)
Ulceration(s)	17 (8)	3a (3)	8a,b (12)	6b (17)	0.021a
Stage I: Non-blanchable erythema	14 (7)	3 (3)	7 (10)	4 (11)
Stage II: Partial thickness	1 (0.5)	0 (0)	1 (1)	0 (0)
Stage III: Full thickness skin loss	2 (1)	0 (0)	0 (0)	2 (6)
Stage IV: Full thickness tissue loss	0 (0)	0 (0)	0 (0)	0 (0)
Heat injury	0 (0)	0 (0)	0 (0)	0 (0)	-
Other	74 (37)	25a (26)	20a (29)	29b (81)	<0.0001a
Bandage directly contacting skin	50 (25)	18a (18)	10a (15)	22b (61)	<0.0001a
Inappropriate flexion/extension	10 (5)	3 (3)	5 (7)	2 (6)	0.460

Subscripts denote subsets of categories that do not differ significantly from one another at the 0.05 level. For “Any error or complication” Orthopaedics (a) is significantly different from both Trauma Hospital ED (b) and Outside Hospitals (c). For “Inappropriate splint type” Orthopaedics (a) is significantly different from Trauma Hospital ED (b) and Outside Hospitals (b), which do not differ from one another. For “Ulcerations” Orthopaedics (a) is significantly different from Outside Hospitals (b), but Trauma Hospital ED (a,b) does not differ significantly from either group. For “Other” Orthopaedics (a) and Trauma Hospital ED (a) do not differ from one another, but both are significantly different from Outside Hospital ED (b). For “Bandage directly contacting skin” Orthopaedics (a) and Trauma Hospital ED (a) do not differ from one another, but both are significantly different from Outside Hospital ED (b).

The most common error was inappropriate length, which occurred in 58/203 (29%) patients (Figure 1), but did not occur at different rates between the groups. In the category of “Other” the most common error was direct contact between the ACE bandage and patient’s skin, which occurred in 50/203 (25%) of patients overall. Other common complications were ulcerations and/ or blistering (Figure 2), which occurred in a total of 23/203 (11%) cases. The second most common issue in this category was failure to immobilize the limb in an appropriate position based on the injury, instead positioning it in inappropriate flexion, extension, or with torsion. This occurred in 10/203 (5%) of patients overall. The most common issue was excessive plantarflexion at the ankle when the patient was placed in a short or long leg splint, which accounted for 7/10 (70%) cases (Figures 3). Circumferential coverage of the extremity was observed in 10/203 (5%) patients (Figures 1B and 4). Additional problems included splint loosening requiring the patient to use additional supportive tape (1), need for a patient to pad the proximal end of splint due to abrasion (1), tissue maceration (1), and splint compression causing phalanges to be crossed (1).

Figure 1a-b:

Examples of inadequate splints. (a) Short arm thumb spica splint covering elbow proximally and MCP joints of 2nd through 5th digits distally. Additionally, the ACE bandage is in direct contact with the skin both proximally and distally. (b) Short arm thumb spica splint placing wrist is in volar flexion rather than a mild amount of dorsiflexion. Fiberglass quickset material overlapping from the mid-forearm distally over the thumb, creating a fully circumferential splint. Inappropriate proximal and distal length of splint again demonstrated.

Figure 3.

Short leg fiberglass splint with ankle placed in inappropriate equinus.

Figure 4.

Short leg splint. Plaster of the proximal portion surrounds greater than two-thirds of leg and is nearly touching at the anterior tibia.

Figure 2a-b:

Examples of ulceration after splinting. (a) Ulceration at second MCP joint secondary to an inappropriately tight splint. (b) Ulceration and ecchymosis at fifth MCP joint due to splint tightness and insufficient padding.

Although application of the ACE bandage directly to the skin has previously been cited as an error, ACE bandages are often directly applied to the skin in the absence of a splint or cotton padding, causing the authors to question this criterion as a true “error.” With this criterion removed, the overall complication rate decreased to 51% (105/203). Thus, 64% (32/50) of splints with the ACE directly contacting the skin had some other error. Group-specific error and complication rates with this criterion removed for Orthopaedics, the Trauma Hospital ED, and Outside Hospitals were 34% (33/98), 61% (42/69), and 83% (30/36) respectively. A repeated chi-squared test confirms that the significant difference in incorrectly placed splints and/ or complications between the three groups is maintained with the elimination of this criterion, with Orthopaedics having the fewest issues (P<0.001).

Discussion

The mantra, “There are no hypochondriacs in casts” is essential to remember for both casted and splinted patients.² Incorrectly applied splints can cause complications ranging from mild issues such as blisters or partial thickness ulceration to limb- or life-threatening complications such as ischemia or infection.^1,2,4-8

Our findings of an overall complication rate of 61% (123/203) and direct application of bandage to the skin in 25% (50/203) are comparable with prior studies and show that the rates of technical errors and complications of splint placement are alarmingly high. A recent study of pediatric extremity injuries noted complications in 93% of the patients evaluated, with the most prevalent complication being direct application of bandage to the skin, seen in 77% of patients.⁴ Another study examining the outcomes of treatment of mallet finger injuries identified skin complications in 38/84 (45%) patients, with 3 cases of skin necrosis secondary to direct application of the strap fastener to the skin.⁷ A study comparing outcomes of pediatric distal radius fracture immobilization found that 24% of patients had sores, and 14% experienced pain from the splint.¹³

Similar to previous studies, this study utilized blanket criteria for technical errors in splint placement including contact of ACE wrap with the skin and malpositioning of joints. Although deviations may be appropriate in specific situations, these actions are not advisable on a global basis, and for the purposes of rigorous study they were all considered errors. Generally, one would expect purposeful deviations from guidelines to be employed more often by the Orthopaedics group, which would artificially inflate the error rate for this group – nonetheless Orthopaedics had the lowest error/ complication rate.

A prior study of non-operative treatment of mallet fracture injuries found that skin complication rates are seven times as likely when prefabricated fiberglass are used instead of custom-made orthoses.⁹ While fiberglass is not inherently problematic, it does have the potential for complications when applied incorrectly. In this study 89% (32/36) of splints placed by an OSH were fiberglass, and a greater percentage of splints placed by an OSH were complicated by ulcers 17% (6/36) than those placed by the Trauma Hospital ED (12%) or Orthopaedics (3%) (P=0.021).

The three study groups were divided based on the training and practice environment. The average age of patients enrolled in the study was 34 ± 22 years. The Orthopaedics cohort was older on average (39 ± 19) than either the Trauma Hospital ED (31 ± 25) or OSH (24 ± 18). This difference likely reflects the inclusion of post-operative patients in the Orthopaedics group, many of whom were elderly. Given the lack of any association between age and complication rate in our study, we felt that the age difference between cohorts was acceptable.

The differences in error and complication rates between ED physicians in our trauma hospital versus OSH may also be secondary to volume of experience versus physician and physician extender instruction by the orthopaedics service in our hospital.

Orthopaedic training programs often utilize surgical skills labs for teaching and practicing techniques and skills on simulators, models, and/ or cadaveric specimens.^14-16 In contrast to many other skills, splinting can be practiced on other trainees and instructors with no ill effects, need for special equipment, or a significant cost to the institution. Further, videos illustrating proper splint placement could be utilized to instruct practitioners at OSH as a review of the previously learned skill. The literature indicates that minor curriculum changes could have significant effects on splinting practice.^17,18 For example, a recent study demonstrated that training paramedic students in a 3-hour course on a cadaveric model increased comfort and proficiency in application of traction splints for femoral shaft fractures.¹¹ In another study evaluating the effectiveness of instructional videos on improving physicians’ procedural skills, pediatric residents were taught to place a volar splint in a small workshop setting and then evaluated on a 5-point scale 2-12 months later by a researcher either without (control) or with (intervention) a review of the technique by means of a 3-minute instructional video prior to performing the procedure. This study found that those who watched the video score 1.87 points higher than those who did not without any increase in the overall time needed to complete the procedure.¹⁷

The present study is best interpreted within the context of its strengths and limitations. All patients were enrolled and evaluated by a single researcher, helping to decrease inconsistency in assessment. Further, all data was collected in a single orthopaedic clinic staffed by a consistent group of attending surgeons, residents, nurses, orthopaedic technicians, and physician assistants who aided in identifying patients and limiting attrition bias. One limitation was the lack of standardization in patient education regarding appropriate splint care. Many patients were excluded from the study because they had removed and replaced the splint prior to evaluation, and it was impossible to tell in many cases if errors such as the bandage directly contacting the skin were secondary to improper technique or poor splint care, as splints were not evaluated at initial application. Another limitation to this study was a lack of control of the type of splinting material utilized. Pre-fabricated devices were used almost exclusively by outside hospitals, and may have negatively impacted measures of technique and patient follow-up as these are more easily removed, and ease of removal has been linked to poor follow-up.¹⁹ Lastly, while the role of all providers in the Trauma Hospital ED and Orthopaedics groups is known, no information is available regarding the training of the providers in the Outside Hospital group, limiting the generalizability of the study.

In summary, this study found significant differences between splint quality when placed by the orthopaedics service, the emergency room staff in our trauma hospital, and outside hospital emergency room staff. The relatively high overall error and complication rates emphasize the importance of meticulous attention to splint quality, and the progressively increased rates in our trauma hospital emergency room versus outside hospital emergency rooms suggests potential benefit of outreach and education of emergency room staff on proper splinting technique.