Transportation of Pediatric Femur Fractures to a Tertiary Care Center: a Retrospective Review

Abstract

Introduction

Pediatric femur fractures are common injuries presenting to tertiary care trauma centers. Transportation of these patients occurs most commonly via ambulance or flight. The purpose of this study is to evaluate whether mode of transportation affects time to surgery or hospital stay for pediatric patients with femur fractures.

Methods

Utilizing a trauma registry we queried pediatric femur fractures between January 2001 and December 2009. Patient age, gender, mechanism of injury, month of injury, type of fracture, transportation, county of origin, time to operating room (TTOR), hospital length of stay (HLOS), and treatment received were identified and compared.

Results

In total, 519 femur fractures were identified, 257 (49.5%) of which were isolated injuries. Flight transportation was utilized in 13.6 % (35 of 257) of these isolated fractures. Mean TTOR for flight patients was 29 hours, HLOS 3.2 days. For ambulance transportation mean TTOR was 41 hours, HLOS 3.2 days. Neither variable was statistically different between transportation groups (TTOR p > 0.50; HLOS p > 0.95). No statistical difference was seen in HLOS (p > 0.47) and TTOR (p > 0.71) for patients originating further distances from the hospital.

Conclusion

Transportation method and distance from the hospital did not affect the TTOR and HLOS for isolated pediatric femur fractures. The use of air transportation for this group of patients, many of whom are injured by relatively low energy mechanisms, may be excessively costly and does not accelerate treatment.

Introduction

A traumatic femur fracture is the most common pediatric orthopedic injury requiring hospitalization, accounting for 21.7% of all pediatric orthopedic trauma, with an incidence of 27.2 per 10,000 children¹. Femur fractures are often a result of trauma and are complicated by the evolving nature of a pediatric skeletal system. Understanding the nature of these fractures is important part of providing care. Epidemiological data on pediatric femur fractures does exist^1-9 but no studies to date have specifically looked at pediatric femur fractures at a United States tertiary care center.

It has been shown that distance to hospital is associated with an increased mortality in those with life threatening injuries¹⁰, and treatment at a trauma center may have survival benefit¹¹. Studies comparing the effect of helicopter versus ambulance transport on survival have shown both improved survival and no effect¹². However these studies have all focused on generic trauma, not specifically orthopedic injuries, and the differences in outcomes appears to be confined to the more severe injuries. Hip fractures have been examined, showing that distance traveled to the hospital did not affect hospital length of stay, time to operating room, or mortality¹³. The purpose of this study is to evaluate whether mode of transportation affects time to surgery or hospital stay for pediatric patients with femur fractures.

Methods

The University of Kentucky Chandler Medical Center maintains a registry on all patients admitted through the trauma service. The International Classification of Disease, ninth version (ICD-9), codes 820 and 821 were utilized to identify all femur fractures in patients under 17 years of age between January 2001 and December 2009. Five hundred and nineteen patient records were found. Patient age, gender, mechanism of injury, month of injury, type of fracture, mechanism of transportation to hospital, county of origin, treatment type received, time to operating room (TTOR), and hospital length of stay (HLOS) were recorded for each patient. Treatment received included spica casting, internal fixation, percutaneous fixation, closed reduction, intramedullary (IM) nailing, and no surgical or casting treatment. Due to database limitations, both flexible and rigid nails are included in the IM nail category. The type of fracture was determined by the assigned ICD-9 code. The county of the hospital plus those counties immediately adjacent to the hospital are termed a “proximal” site of origin. All other counties are termed “outlying.” Transportation mechanisms include ambulance referred from an outside hospital, ambulance from the scene, helicopter referred, helicopter from the scene, and patient's personal transportation. A two tailed T-Test was utilized to identify significant differences (P < 0.05). This study was reviewed and approved by the University of Kentucky Institutional Review Board.

Results

In total, 519 patients were identified; 371 (71.48%) were male and 148 (28.51%) female, a 2.5:1 ratio. Motor vehicle accidents (MVAs) (27.75%) and falls (25.82%) were the most common mechanism of injury, followed by all terrain vehicle (ATV) accidents (10.02%), unspecified causes (8.67%), and motorcycle accidents (7.90%). Eight of the nine (88.89%) assault injuries occurred in children under one year of age, and was the second most common mechanism of injury in this age group. Females had higher incidence rates of MVAs and fall fractures, and fewer sports, ATV, motorcycle, and gunshot wounds than males. Table 1 details the various mechanisms of injury. The average age of female patients, 7.92 years, was significantly different (P = 0.015) than that of males, 9.14 years.

Table 1.

Mechanism of injury for all fractures recorded.

	Female	Female %	Male	Male %	Total	Total %
Assault	3	2.03	6	162	9	1.73
ATV	8	5.41	44	11.86	52	10.02
Bike	4	2.70	21	5.66	25	4.82
Crush	4	2.70	4	1.08	8	1.54
Fall	47	31.76	87	23.45	134	25.82
Motor Vehicle Accident	62	41.89	82	22.10	144	27.75
Other	9	6.08	36	9.70	45	8.67
Other Vehical	2	1.35	5	1.35	7	1.35
Pedestrian Accident	8	5.41	18	4.85	26	5.01
Sport	1	0.68	25	6.74	26	5.01
Gun Shot Wound	0	0.00	2	0.54	2	0.39
Motorcyle	0	0.00	41	11.05	41	7.90
TOTAL	148	28.52	371	71.48	519	100.00

Of all fractures, 93.42% were closed, and 6.58% open. The most common type of fracture was an unspecified closed shaft fracture, which accounted for 65.04% of all fractures, followed by closed distal fractures at 12.59%. Figure 1 details the percentage of each type of fracture observed. Of the 519 total fractures, 257 (49.5%) were isolated injuries.

Figure 1. Percentage of each fracture type based on ICD-9 coding.

For treatment received, an IM Nail (44.70%) and spica casting (16.96%) were the most common, followed by no surgical or casting treatment (15.22%). No significant differences were observed between males and females. Treatment correlated with age. Patients under 12 months of age received no surgical or casting treatment 84.62% of time. Patients from 1-5 years of age most commonly received spica casting, and patients over 5 years of age an IM nail (Figure 2). No significant difference was seen in HLOS for spica casting versus IM nail (P = 0.74).

Figure 2. The percent of total treatments for a given age group.

The average HLOS was 4.97±6.30 days. The average TTOR was 40.33±99.31 hours. There was no difference between male and female HLOS (P = 0.47). For TTOR, males averaged 46.02 hours versus 24.94 hours for female patients, which was close to being significantly different (P = 0.059). No significant differences were seen in TTOR and HLOS for proximal patients versus outlying patients. No significant differences were seen in HLOS and TTOR for patients when divided into the age groups of <2 years old, 2-5 years old, 6-12 years old, and 12-16 years old.

Among patients with isolated femur fractures, 13.62% utilized flight transportation. Mean TTOR for flight patients was 29±84 hours, HLOS was 3.2±4.3 days. For ambulance transportation mean TTOR was 41±90 hours, HLOS was 3.2±3.6 days. For personal transportation mean TTOR was 40.24±95.37 hours, HLOS was 2.54±3.49 days. No significant difference was seen in TTOR or HLOS for any of the methods of transportation for isolated femur fractures.

Discussion

When isolated fractures are examined there were no differences in HLOS or TTOR for any of the transportation mechanisms. Isolated fractures were examined in this instance to control for polytrauma. Patients flown in via helicopter did not reach the operating room any quicker than those brought in by ambulance or via personal transportation. This data suggests that in cases of isolated fractures, the utilization of helicopter transportation does not increase the speed of treatment.

Based on the costs of helicopter use, an ambulance is likely a more cost efficient alternative¹². This is particularly applicable when patients are being transferred from an outlying hospital, where it has been shown helicopter use may not actually increase transport times²⁵. Additionally Crandall et al. has shown that provider determined transfer times for trauma cases that exceed two hours have no adverse effect on patient outcome²⁶, supporting the idea that urgent use of helicopters for faster transfer times is not necessary. Matsushima et al. also showed that surgical team workload, ISS, and caseload had no correlation to TTOR, minimizing confounding variables for this statistic²⁷. We hypothesize that this lack of significant difference in TTOR for isolated femur fractures would hold true for other isolated pediatric orthopedic injuries.

According to our findings, significantly more fractures occurred in males than females. This has been previously described^2-9 and is likely due to males' increased disposition for riskier behaviors, as seen by their higher incidence of motorcycle, gun shot, and ATV injuries. These differences in injury mechanism may also account for the higher average age for males. Falls and MVAs represented over half of all fractures, and were the first and second most common causes of injury in those under 12, respectively. This is largely consistent with prior research that has shown either falls or MVAs being first and second throughout this age range^2,5-9. One exception is a study by Heideken et al. that showed sports accidents to be the most frequent cause in the 4 - 12 age range⁴. The high rate of ATV injuries is representative of the rural population base of the care center.

In children under one year of age, 30.77% of fractures were a result of assault, which was most common mechanism of injury in this age group after MVAs. This is especially disturbing considering assault is often only reported when the physician has concrete evidence abuse has taken place. Many suspected cases of assault or non-accidental trauma are not reported¹⁴, and this percentage may in fact be much higher. The American Academy of Orthopedic Surgeons recommends that all diaphyseal fractures in children under 36 months of age be evaluated for child abuse¹⁵. The existing data relating non-accidental trauma to femur fractures is highly variable. For children less than a year old, femur fractures due to abuse range from 8.5% - 90%^8,16,17. Loder et al. reported 15% of fractures in those less than two being due to abuse², Beals and Tufts report 30% in those less than four¹⁸, and Beuss and Kaelin report 7% in those less than four¹⁹. One common theme is that age is the biggest factor in determining non accidental trauma^17,20. In a study by Rex et al. on femoral fractures owing to definite abuse, 92.8% of their cases occurred in children younger than one²⁰, similar to our study's 88.89%.

Treatment received correlated with age. The majority of infants received no further surgical or casting treatment after the fracture was reduced and set. Young children between one and five years of age most often received spica casting, and those over five years most often an intramedullary nail. Due to database limitations we are unable to report on the breakdown of flexible versus rigid nails, though we presume based on standard faculty practices that the use of titanium elastic nails predominated in this group. Management of femur fractures in children ages 6-10 has evolved away from spica casting and towards intramedullary flexible nails^1,21, and our data parallels this trend. Titanium elastic nails have been shown to hasten fracture union, reduce the rate of malunion and shortening, and allow earlier rehabilitation and return to school for ages those aged 5-15^22,23. However evidence for use of methods other than titanium elastic nails for unstable fractures has recently emerged²⁴.

The average HLOS of 4.97 days was less than that of two older studies^1,2 and consistent with that found by Heideken et al. in 20054. This may be due to a more current data set, indicating that evolving treatment methods and techniques are having a beneficial effect. When examining these studies in chronological order the HLOS does decrease, something that Heideken et al. also demonstrated⁴. In contrast to Loder et al, a longer HLOS was not seen in the older age groups². The average TTOR for females was very close to being significantly shorter than that for males (P = 0.059). This is counter-intuitive as males are more likely to be involved in gunshot, motorcycle, and ATV injuries which would presumably require more urgent care. However a proportionally higher percentage of females sustain femur fractures in MVAs, which are also a priority at trauma centers. There was no significant difference in HLOS or TTOR for patients brought in from outlying counties versus those from proximal counties. Distance from the hospital does not appear to play a role in the timely delivery of treatment for this patient population. This is consistent with a prior study on isolated hip fractures¹³, though is in contrast to studies on all cause trauma¹⁰. However even when examining all cause trauma, the difference in outcomes appears to be confined to the most severe injuries. This would further support the notion that pediatric femur fracture outcomes are largely unaffected by distance travelled.

In conclusion, transportation method and distance from the hospital did not affect the TTOR and HLOS for isolated pediatric femur fractures. The use of air transportation for this group of patients, many of whom are injured by relatively low energy mechanisms, may be excessively costly and does not accelerate treatment.