Do All-terrain or Recreational Off-highway Vehicle Crashes Have a Higher Odds of Amputation?

Abstract

Background

Riding off-road vehicles is associated with the risk of injury to the extremities. There are two main types of four-wheel recreational off-road vehicles: quads or all-terrain vehicles (ATVs), which are essentially four-wheel off-road motorcycles, and recreational off-highway vehicles (ROVs), also colloquially referred to as utility terrain vehicles, which have side-by-side seating, higher maximum speeds, and a roll cage. There are multiple orthopaedic society position statements on ATVs, but none on ROVs. Perhaps this is because the injury patterns and differences between the two vehicles have not been elucidated.

Questions/purposes

(1) What patient, vehicle (ROVs versus ATVs), and injury factors are associated with amputation? (2) What are the anatomic location distributions of fractures and amputations by vehicle type?

Methods

Records of all patients in our hospital’s billing system who had both a diagnostic code indicating an accident related to an off-road vehicle and one indicating an extremity or pelvic fracture between February 2014 and January 2020 were screened; this resulted in the identification of 328 patients with fractures resulting from off-road vehicle collisions. A total of 16% (51 of 328) of patients were excluded from the analysis because their injury did not involve either an ATV or an ROV; 277 patients were included in the final analysis. The following variables were collected: age at time of the injury, gender, BMI, vehicle type, Gustilo-Anderson type if applicable, amputation level if applicable, anatomic locations of injuries, ethanol level, and drug screen. ATV crashes accounted for 52% (145 of 277) of patients, and ROV crashes accounted for 48% (132 of 277). Patients from ATV crashes did not differ from those in ROV crashes in terms of mean age (24 ± 16 years versus 24 ± 13 years; p = 0.82), BMI (25 ± 7 kg/m² versus 26 ± 6 kg/m²; p = 0.18), or gender (79% [114 of 145] men/boys versus 77% [102 of 132]; p = 0.79). Among patients who had a drug or ethanol screen, there was a higher percent of ATV riders who used marijuana (39% [19 of 49] versus 17% [7 of 42]; p = 0.04), but there were no differences in abnormal blood alcohol screen or abnormal nonmarijuana drug screen; however, these results were available in only about one-third of patients (99 of 277 for ethanol and 91 of 277 for drug screen). Statistical analysis was performed using logistic regression analysis for factors associated with amputation, with p values < 0.05 considered significant.

Results

After controlling for differences in demographic factors, the stepwise increase in Gustilo-Anderson grade of open fracture (OR 9.8 [95% CI 3.6 to 27.0]; p < 0.001) and ROV vehicle type (OR 15.7 [95% CI 3.6 to 68.5]; p < 0.001) were both associated with amputation. There was no increase in the odds of amputation associated with age (OR 1.0 [95% CI 0.9 to 1.1]; p = 0.81), gender (OR 1.4 [95% CI 0.3 to 5.8]; p = 0.68), or BMI (OR 1.1 [95% CI 0.9 to 1.2]; p = 0.37). The most frequent ATV fractures occurred in the forearm and wrist (22% [45 of 203]), whereas most ROV injuries occurred through the metacarpals (41% [107 of 262] of fractures and 58% [18 of 31] of amputations).

Conclusion

ROV crashes are associated with a higher odds of amputation when compared with ATV crashes. Because most ROV injuries were in the forearm and below, this likely occurs when upper extremities are crushed and mangled under the roll cage in rollover ROV crashes. Because of this danger, we urge our orthopaedic societies to either update current ATV position statements to include ROVs or release separate statements on ROVs.

Level of Evidence

Level III, prognostic study.

Introduction

Riding off-road vehicles is an activity that is associated with injury risk. Each year, more than 135,000 people in the United States are injured while riding off-road vehicles; more than 800 of those injuries are fatal [20]. There are two main types of four-wheel, recreational, off-road vehicles. A quad, also known as an all-terrain vehicle (ATV), is a vehicle the rider sits on and controls using handlebars, similar to a motorcycle (Fig. 1). In contrast to ATVs, recreational off-highway vehicles (ROVs), also referred to as utility terrain vehicles (UTVs) or side-by-sides, have side-by-side seating with seatbelts, a steering wheel, and a roll cage (Fig. 2), and they can travel at highway speeds [12, 19]. Both vehicles can be operated without a license when driven off public roadways. Although the terms UTV and ROV are used interchangeably in colloquial references, the accurate term for what recreational off-roaders use is an ROV [12]. The high center of gravity and high-speed capabilities of both kinds of these vehicles may contribute to the high frequency of injury associated with their use. The most common injuries seen in off-road vehicle collisions are fractures, contusions, or abrasions, with the highest risk of fatality during rollovers and collisions [5].

Fig. 1.

An ATV coasts over Imperial Sand Dunes, California. This image is in the public domain and can be accessed at: https://www.flickr.com/photos/blmcalifornia/29802635476/in/photostream/.

Fig. 2.

Recreational off-road vehicles (UTVs) race up sand dunes. This image is in the public domain and can be accessed at: https://www.flickr.com/photos/blmcalifornia/29211973923/in/photostream/.

With a variety of terrain and warm weather nearly year-round, Southern California is host to a high volume of off-road riding and, therefore, a high volume of off-road crashes throughout the year. Many patients involved in ATV and ROV collisions are brought to our institution annually. Despite having a roll cage, ROVs can still result in serious injuries, especially during rollover crashes. Intuitively, it might seem that having a roll cage and an ROV cockpit would be safer than riding in an exposed position on top of an ATV. However, whether this is in fact the case has not, to our knowledge, been evaluated. With the recent emergence of the ROV as the preferred off-road vehicle over the ATV [12], more information is needed on its safety profile, particularly as it compares to the ATV.

Therefore, we asked: (1) What patient, vehicle (ROVs versus ATVs), and injury factors are associated with amputation? (2) What are the anatomic location distributions of fractures and amputations by vehicle type?

Patients and Methods

Study Design and Setting

This was a retrospective, comparative study based on a chart review performed at a Level 1 trauma center, Loma Linda University Health, in Loma Linda, CA, USA. The hospital is in a suburban area and is a referral center for a large catchment basin composed of the surrounding desert area in all directions, primarily north, south, and east. The patients were either managed by the orthopaedic surgery service or the hand surgery service, which includes members of both orthopaedic surgery and plastic surgery services.

Patients

The chart review used a search that combined the ICD V86 prefix (accident related to an off-road vehicle) with all ICD and Current Procedural Terminology (CPT) codes for extremity and pelvic fractures between February 2014 and January 2020. Spine and skull fractures were not included. The query yielded 328 patients; on further review, 16% (51 of 328) of patients were excluded because their collision did not involve either an ATV or ROV. The final analysis included 277 patients.

Participants’ Baseline Data

In all, 52% (145 of 277) of patients were involved in ATV crashes, and 48% (132 of 277) were in ROV crashes. Patients from ATV crashes did not differ from those in ROV crashes in terms of mean age (24 ± 16 years versus 24 ± 13 years; p = 0.82), BMI (25 ± 7 kg/m² versus 26 ± 6 kg/m²; p = 0.18), or gender (79% [114 of 145]) men/boys versus 77% [102 of 132]; p = 0.79) (Table 1). Among patients who had a drug or ethanol screen, there was a higher percentage of ATV riders who used marijuana (39% [19 of 49] versus 17% [7 of 42]; p = 0.04), but there were no differences in abnormal blood alcohol screen (23% [12 of 53] versus 41% [19 of 46]; p = 0.08) or abnormal nonmarijuana drug screen (22% [11 of 49] versus 33% [14 of 42]; p = 0.36); however, these results were available in only one-third of patients (99 of 277 for ethanol and 91 of 277 for drug screen) (Table 1).

Table 1.

Demographic comparison between ATV riders and ROV riders

Parameter	ATV (n = 145)	ROV (n = 132)	p value
Age in years	24 ± 16	24 ± 13	0.82
BMI in kg/m2	25 ± 7	26 ± 6	0.18
Gender, men and boys	79 (114)	77 (102)	0.79
Positive blood alcohol level	23 (12 of 53)	41 (19 of 46)	0.08
Positive drug screen	22 (11 of 49)	33 (14 of 42)	0.36
Positive marijuana screen	39 (19 of 49)	17 (7 of 42)	0.04

Data presented as mean ± SD or % (n).

For blood alcohol, drug, and marijuana screening, only a fraction of patients were tested at the discretion of the Emergency Department; denominators are included.

Data Sources and Variables

Data collected for review included patient age at the time of injury, gender, BMI, vehicle type, open or closed fracture status, Gustilo-Anderson type when applicable, amputation level when applicable, anatomic locations of injuries, ethanol level, and drug screen. Age, gender, and BMI were all gathered from the electronic medical record; Gustilo-Anderson type and amputation level, when applicable, were gathered from the operative note written by the attending surgeon; type of vehicle was gathered from the resident consult note as we have been including this prospectively since 2009 because of a prior study; anatomic locations of injuries were from both consult and operative notes; and ethanol and drug information was from laboratory data corresponding to the date of presentation. Open fractures were described in accordance to the previously described Gustilo-Anderson classification [14].

Of ROV riders evaluated for orthopaedic injuries, 83% (109 of 132) had open fractures, whereas 26% (39 of 149) of ATV riders had open fractures. The odds of amputations paralleled that of open fractures; 21% (28 of 132) of ROV riders versus only 1% (2 of 145) of ATV riders sustained one or more amputations.

Ethical Approval

We obtained ethical review board approval for this study.

Statistical Analysis

We performed a logistic regression analysis to evaluate the association of age, gender, BMI, Gustilo-Anderson fracture type, and vehicle type with the odds of amputation. All data analyses were performed using SPSS software (Version 22, IBM Corp), with p values < 0.05 considered significant.

Results

Factors Associated With Amputation

After controlling for differences in demographic and injury factors related to amputation, the stepwise increase in Gustilo-Anderson type of open injury (OR 9.8 [95% CI 3.6 to 27.0]; p < 0.001) and ROV vehicle type (OR 15.7 [95% CI 3.6 to 68.5]; p < 0.001) were both associated with amputation (Table 2). There was no increase in the odds of amputation associated with age (OR 1.0 [95% CI 0.9 to 1.1]; p = 0.81), BMI (OR 1.1 [95% CI 0.9 to 1.2]; p = 0.37), or gender (OR 1.4 [95% CI 0.3 to 5.8]; p = 0.68) (Table 2).

Table 2.

Association of increasing age, increasing BMI, gender, and increasing open fracture severity on the risk of amputation when comparing ATVs and ROVs

Parameter	OR (95% CI)	p value
Increasing age, per year	1.0 (0.9 to 1.1)	0.81
Increasing BMI, per kg/m2	1.1 (0.9 to 1.2)	0.37
Gender, men compared with women	1.4 (0.3 to 5.8)	0.68
Increasing open fracture severitya	9.8 (3.6 to 27.0)	< 0.001
Vehicle type, ROV	15.7 (3.6 to 68.5)	< 0.001

^aIncreasing fracture grade, with each increased step: I, II, IIIA, IIIB, IIIC.

Anatomic Locations of Fractures and Amputations by Vehicle

ROV riders had a higher proportion of upper extremity injuries compared with ATV riders (90% [119 of 132] versus 45% [65 of 145]) (Supplementary Table 1; http://links.lww.com/CORR/A957). ATV riders had more pelvic/acetabular fractures and lower extremity fractures compared with ROV riders (4% [6 of 145] versus 0% [0 of 132] and 40% (58 of 145) versus 10% [13 of 132], respectively). In ROV riders, midhand and palm locations accounted for more than two-thirds of amputations.

Discussion

There is inherent danger involved in the operation of both ATVs and ROVs, as evidenced by off-road vehicle injury data spanning from the late 1980s through the 2010s showing tens of thousands of deaths and millions of injuries [22]. With unmodified vehicles from the factory capable of reaching speeds up to 100 mph or more, the risks will continue to increase [17, 21]. Despite the danger to life and limb, people will continue to ride these machines for their intended use: recreation. As demonstrated in our study, in addition to the inherent danger of operating off-road vehicles, when fractures are sustained, riders of ROVs have a higher odds of open fracture and amputation than do riders of ATVs. Multiple orthopaedic societies including the American Academy of Orthopaedic Surgeons, the Orthopaedic Trauma Association, and the Pediatric Orthopaedic Society of North America have recognized the dangers of ATVs and have all released position statements on the safe use of these vehicles; however, there have not been statements pertaining to ROVs [1].

Limitations

There are some limitations to this study that should be acknowledged. First, because of the lack of black box–like recording data, we were unable to examine the effect that speed at the time of the collision had on these injuries. However, the speed potential of both vehicles has been increasing [6, 17, 21], and we suspect that this increased speed potential has led to higher-speed collisions. Second, the data were collected from patients presenting to a Level 1 trauma center which tends to see a larger proportion of higher-acuity patients and may not represent the general population [4]; however, it likely does represent the subset of people who have experienced injuries from high-energy, off-road vehicle crashes. Third, the study design allowed us to only assess patients who sustained musculoskeletal injuries and did not include patients who had isolated trauma to other systems such as the abdomen or the brain. Additionally, meaningful analysis of intoxication information was limited by the lack of uniform screening on all patients. Finally, the study only included patients who made it to the hospital and did not die either in the field or during transport, and we therefore could not account for true severity of injuries.

Higher Fracture Type Classification Is Associated With an Increased Odds of Amputation

In our study, there was a higher odds of amputation with higher Gustilo-Anderson classification injuries. This is not necessarily surprising; an association between increasing Gustilo-Anderson type and amputation was first noticed during the development of the Gustilo-Anderson classification [10]. An earlier study showed that although patients with Type IIIA fractures have an amputation rate of 0%, the rate increases to 16% with Type IIIB fractures and then to 42% with Type IIIC fractures [10]. This correlates with the extent of bony and soft tissue (especially vascular) injuries, which are major determinants in the decision to amputate [23]. However, there is no consensus on the predictive nature of the classification system. Some studies have cast doubt on the predictive nature of the Gustilo-Anderson classification and favor the Orthopaedic Trauma Association Open Fracture Classification (which combines grades of skin, muscle, and arterial damage, contamination, and bone loss for a composite score), whereas others have favored the Ganga Hospital Open Injury Severity Score (which combines grades of skin, soft tissue, and bony injuries with other comorbidities for a composite score) [7, 9, 11, 15, 18]. Clearly, there is no perfect model.

Anatomic Location of Injuries and Amputations

Nearly all amputations while using an ROV occurred in the upper extremity with most at the metacarpal level. Another quarter of the amputations were at the carpal level or proximal, and only a small proportion were at the phalangeal level. Further, almost all ROV injuries involved the upper extremity. We believe the phenomenon of open fractures and amputations associated with ROV crashes can be explained by the presence of a roll cage, also known as a rollover protection system, which is a tubular structure surrounding the vehicle to protect the riders [3] (Fig. 2). Humans have a subconscious reflex to extend their upper extremities to arrest an imminent fall [8, 13]. Similar to a fall in which the ground is rapidly approaching, in a rollover crash, the ground is also rapidly approaching. In a split second during a rollover crash, the occupants of an ROV can reflexively extend their limbs outside the cockpit of the vehicle, and the limbs can subsequently be crushed between the ground and the bars. With the weights of some ROVs exceeding 2000 lbs, resultant crush injuries can easily cause open fractures and amputations, depending on the weight applied on the limb through the roll cage [21].

In our emergency department, mangled upper extremities in ROV riders are often seen and managed, with injuries predominantly occurring through the palm and wrist. From our experience, when they are able to describe what led to the injury, most patients attribute it to a crush injury by the roll cage. Doors and window netting can keep extremities inside the ROV; however, when ROVs are sold to the consumer, doors and window nets are not required by state or federal law [2, 16]. Doors and window nets are required by many off-road racing organizations to keep the extremities inside the vehicle in case of a rollover [2, 16]. Although racers are protected by the rules of their sport, specific racing organization requirements do not affect most of our patient population. People have the right to make their own choices regarding the pros and cons of using doors and window nets; however, we wonder whether the risks are fully understood when purchasing these vehicles; at the time of writing, although general safety recommendations and the possibility of “severe injury” are listed on all of the top six manufacturer websites (Polaris, Can Am, Honda, Yamaha, Kawasaki, and Textron), specific risks such as mangled limbs or amputations are not mentioned. In response to how devastating the injuries can be and in the interest of public health, we recommend that our orthopaedic societies either update current position statements on ATVs to include ROVs or they should issue separate position statements to render explicit the different (and apparently increased) risks associated with ROVs.

Conclusion

ROV crashes are associated with a higher odds of amputations compared with ATV crashes. Because most ROV injuries were at the forearm level and below, injuries likely occur when upper extremities are crushed and mangled under the roll cage in rollover ROV crashes. Because of the differences in injury patterns, we urge our orthopaedic societies to either update current ATV position statements to include ROVs or release separate statements on ROVs. In particular for ROVs, adding door and window nets may be sufficient to lower the risk of devastating hand, wrist, and forearm injuries.