Abstract
Introduction
Amputations and avulsion injuries due to horse-associated activity are rare, yet they can result in significant impairment. The purpose of this study was to further investigate such injuries using a national emergency department database.
Methods
The US National Electronic Injury Surveillance System (NEISS) was used to identify horse-associated amputation and avulsion injuries occurring between 2000–2023. Demographic data of age, sex, and injury details were collected.
Results
There were 34,091 emergency department visits for equine-associated injuries, with 120 (0.35%) due to amputations/avulsions; 53 (44%) patients sustained amputations, and 67 patients (56%) sustained avulsions. The average age was 37 (SD = 21 years). There were 78 female and 42 male patients. The most common mechanism of injury was riding the horse, with further details not specified (31%), followed by equipment issues (19%), bucked/thrown/kicked off the horse (15%), falling off the horse (11%), and others (6%). A rope/chain was involved in 29 patients (24%). There were 55 amputations involving the finger (40), thumb (13), and others (2). Rope-related injuries were more commonly involved in those sustaining amputations versus avulsions (42% vs 10%, P < .001). Males had more rope-associated injuries (36% vs 18%, P = .043).
Conclusions
This is the largest study to date of amputations and/or avulsions due to horse-associated injuries. There were multiple mechanisms of injury, with ropes involved in one-quarter. This baseline data can be useful for evaluating the effectiveness of future prevention programs.
Introduction
Injuries from horses are well described. 1 Such injuries range from simple bruises and strains that likely need no significant medical care to fractures and internal organ injuries (tracheal, esophageal, pulmonary, hepatic or splenic, renal, bladder), spinal injuries (with or without paralysis), and head trauma ranging from concussions to massive intracranial bleeds. Many such injuries may require hospital admission with the potential for death due to either the serious nature of the injury itself (head, internal organ) or complications from such injuries (eg, a fatal pulmonary embolism secondary to a pelvic fracture). There is very little literature on amputations2–6 and avulsions 7 sustained from horse interactions, yet these injuries can also result in significant functional and/or cosmetic impairment. It was the purpose of this study to further investigate the types of amputations and avulsions (tearing away of a body part) from equestrian activity using a national emergency department (ED) database.
Materials and Methods
The data for this study were obtained from the US National Electronic Injury Surveillance System (NEISS) database coordinated by the US Consumer Product Safety Commission. The general materials and methods and statistical analyses were the same as previously described, 8 except that search criteria included only amputations and avulsions associated with horse-related activity. From the NIESS data, multiple demographic variables were collected (sex, race, age, diagnosis of injury, incident locale of injury, anatomic location of the injury, disposition from the ED, and hospital size [strata]). Also, there is a column entitled “narrative comments” that gives a vignette of the injury, in essence a mini-history, which was also collected. The NEISS collects injury data daily from ∼100 hospitals that have emergency departments in the United States and its territories. The NEISS data is publicly available (https://www.cpsc.gov/cgibin/NEISSQuery/home.aspx), and acquisition/guidelines are available at www.cpsc.gov/library/neiss.html. This study was considered exempt by our Institutional Review Board.
For this study specifically, the NEISS data for the consumer product code 1239 (horseback riding—activity, apparel, or equipment) were downloaded for the years 2000–2023. From this data, those with a diagnosis code of 50 (amputation) or 72 (avulsion) were identified. Because the NEISS only codes for the most severe diagnosis, the narrative comments were searched for the terms (amp) and (avul), as a person with a more severe injury (eg, pneumothorax, splenic laceration) could also have sustained an amputation or avulsion but would not have been coded as such since the other injury was more severe. Each case discovered by this search was reviewed and, when confirmed to be an amputation or avulsion, was then added to those found using the NEISS codes of 50 and 72, resulting in the final data set. Disposition from the ED was classified as discharged or not discharged; patients transferred from the initial NEISS hospital to another facility were defined as not discharged. Race was classified as White, Black, and other as per the NEISS.
The type and anatomic location of the injury was tabulated. The mechanism of injury was grouped into 6 different categories using the narrative comments: the patient 1) was bucked, thrown, or kicked off of the horse; 2) was stepped on by the horse; 3) fell off the horse; 4) simply rode the horse with no other details provided, 5) had discrete equipment issues (tack), and 6) all others. These others were being bitten by the horse, kicked by the horse, and striking an object. We also ascertained if equipment issues and/or a rope was specifically involved or thought to cause the injury. Finally, a sampling of interesting cases using the narrative comments was chosen to provide context and examples of how these injuries occur.
As the NEISS data set comprises only ∼100 hospitals, it is stratified by hospital size, which is determined by the number of ED visits for each hospital with assigned weights for each hospital. For stratified and weighted data, statistical analysis typically produces national estimates and 95% confidence intervals, customary for such data set designs. 9 However, the number of patients in this study was small, and when the actual number of patients is small (<20), the estimates become unstable and should be interpreted with caution. 10 Rather than using potentially unstable national estimates, we simply studied this data as a cohort of patients in a traditional, retrospective manner of prospectively collected data. Continuous data are expressed as the mean and standard deviation, and categorical data as frequencies and percentages. Differences between continuous variables were determined using the Student t-test. Differences between categorical variables were determined with Fisher's exact test for 2 × 2 tables and the χ2 test for tables larger than 2 × 2. A P < .05 was considered statistically significant.
Results
Over the 24-year period from 2000 through 2023, there were 34,091 ED visits for equine-associated injuries, with 120 (0.35%) describing an amputation or avulsion (Table 1); 53 (44%) were amputations and 67 (56%) avulsions. The average age was 37 (SD 21 years) (range 2 to 86 years); the second decade of life was the most frequent age of injury (Figure 1). There were 78 (65%) female and 42 (35%) male patients; when the race was known (83 patients), 77 (91%) were white. The patient was discharged from the ED in 73% of cases. There were no reported deaths. The patient was mounted on the horse 64% of the time and not mounted 33%, with 3% of the injuries occurring during mounting or dismounting. The most common mechanism of injury was simply riding the horse (31%), followed by discrete equipment (tack) issues (19%), stepped on by the horse (18%), bucked/thrown/kicked off the horse (15%), falling off the horse (11%), and all others 6%. A rope/chain was involved in 29 patients (24%), with a rope in 28 and a chain in 1. When the incident locale was known, most of the injuries occurred either at the patient's home/farm or a place of sport/recreation.
Age histogram.
Demographic data of amputation and avulsion injuries from equines.
Numbers in parentheses are percentages. All are column percentages except for the top row.
p-Value for only the mounted and unmounted groups.
p-Value excluding unknown locations.
There were 55 amputations in 53 patients, which included the finger in 40, the thumb in 13, and in 1 case each an ear and a toe. There was 1 patient with bilateral thumb amputations and 1 patient with amputations of at least 2 digits on one hand. Of the 55 amputations, 15 involved the fingertips, and the remaining 40 involved more proximal levels. There were 73 avulsions in 67 patients and involved the finger in 24; toe in 14; forearm in 6; hand, lip, or thumb in 4 cases; elbow, foot, thigh, and upper trunk each in 2 cases; and 1 avulsion injury involving either the breast, ear, hip, knee, lower leg, mouth, scalp, or shoulder. Of the avulsions involving the fingers/toes, the level was known in 23; 18 involved the finger/toenail and 5 were more proximal than the nail.
Thirteen other associated injuries were noted in 11 patients (9%). These were a closed head injury in 2 patients, a humerus fracture in 2 patients, and then in 1 patient each a fracture of the sixth cervical vertebrae, first lumbar vertebra, femoral shaft, ankle, rib, and radial head. One patient sustained 3 additional injuries, which were a rib, nasal, and lumbar transverse process fracture.
Differences by Mechanism of Injury
Those who sustained an amputation were most frequently simply riding the horse (53%) compared to those with an avulsion (14%) (Table 1); while those with an avulsion more commonly fell off the horse (18% vs 2%), were bucked/thrown/kicked off the horse (22% vs 8%), or stepped on by the horse (28% vs 18%, P < .001) (Figure 2A). A rope was more commonly involved in those sustaining amputations compared to avulsions (42% vs 10%, P = <.001) (Figure 2B). Those with avulsions were more commonly discharged from the ED (84% vs 58%, P = .004) (Figure 2C). Females had more injuries when not mounted on the horse (42% vs 18%, P = 0.03) while males had a higher percentage of injuries involving a rope (36% vs 18%, P = .043). When a rope was involved, an amputation was more common compared to an avulsion (76% vs 24%, P < .001) (Table 2). There was no difference in age between the rope involvement groups.
Comparison between the amputation and avulsion groups. A: By injury mechanism (P < .001). B: By rope involvement (P < .005). C: By disposition from the ED (P < .005).
Demographics by rope involvement.
Numbers in parentheses are column percentages.
Illustrative Case Examples From the Narrative Comments
A 58-year-old woman was loading her horse with a lunge (long line 25 to 35 feet) line when the horse backed up and the line wrapped around her hands, resulting in amputation of both right and left thumbs.
A 27-year-old man got his right finger caught between the saddle horn and a rope, amputating his distal finger.
A 2-year-old girl was feeding a horse when it bit her, resulting in a partial amputation.
A 16-year-old girl was wearing a thumb ring, and while leading the horse, the horse jerked, catching the thumb in the lead strap, amputating the thumb.
A 9-year-old girl fell off a horse and then the horse trampled on her finger, resulting in a near amputation of the finger.
A 60-year-old male sustained an avulsion to his fingertip while wrapping a rope with his horse, catching his finger.
A 20-year-old male was thrown off a horse and then stepped on by the horse, sustaining lacerations over the pectoral major and deltoid muscles with avulsed skin.
A 53-year-old male, while riding his horse and roping a calf, got his left ring finger caught in the rope, resulting in amputation.
A 68-year-old male, while riding a horse, pushed a tree branch out of the way when the branch broke, rebounded, and caught the back of the hand, resulting in skin avulsion with exposed tendons.
A 79-year-old female fell from the horse after it was spooked, resulting in a skin avulsion over the elbow as well as a subarachnoid hematoma.
A 43-year-old female was dragged 100 yards by a horse when it got spooked, resulting in amputation of the right finger, forearm degloving, a first lumbar compression fracture, and facial lacerations.
A 68-year-old female was riding her mule when she hit a branch, sustaining a large avulsion laceration to the left breast.
A 41-year-old female was bitten by her horse when she put her hand in its mouth, resulting in a finger avulsion.
Discussion
There is a paucity of literature regarding acute amputations resulting from equine trauma,2,3 and none that surveys an entire nation from all types of ED access (rural to large city hospitals, and both discharged or admitted from the ED). Morgan et al 3 described 3 men with thumb injuries sustained during rodeo roping competition, resulting in amputations in 2 and a partial avulsion in the third. They noted that in all 3 injuries, the thumbs were entwined between the saddle horn and rope, with crushing and shearing taking place as tension was placed on the rope. A case report 2 described an amputation in a 23-year-old man of his middle finger at the level of the proximal phalanx after being bitten by a horse. The amputated stump had been avulsed, along with the flexor tendons being torn from the musculotendinous junction in the forearm. In our study, 22 of the 120 specifically described the tack (typically leather or nylon equipment used to work with or ride the horse) as being involved in the injury, with at least one-quarter of the amputations involving tack. In a study of 99 horse-related injuries at a level 1 trauma center 11 there were 10 patients who got their fingers tangled in the reins; a finger amputation occurred in 3 of those 10. A study from the United Kingdom 12 published 8 cases where inappropriate grip of the reins or halter rope while leading horses resulted in an avulsion injury to a finger/thumb. Four involved the thumb and the other 4 a finger.
There is more information in the literature regarding rodeo rope injuries. Three men with thumb injuries 3 sustained during rodeo roping competition sustained 2 amputations and 1 partial avulsion. In all 3 injuries the thumbs were entwined between the saddle horn and rope where crushing and shearing occurred as tension was placed on the rope. Digit amputation, especially the thumb, occurs when the digit is caught in a loop of rope against the horn of the horse saddle. This happens during “dallying,” where the rider secures the rope by taking wraps around the saddle horn. Digital entrapment injuries have a poor prognosis regarding replantation. In one study, 5 33% were successful for roping amputations while another study 6 reported a 59% success rate. A third study 13 showed a success rate of 70% but had a 43% rate of complications and a common need for reconstructive measures.
If ED discharge is used as a proxy for injury severity, then most of the injuries would be considered nonsevere, as 73% were discharged. However, many of the injuries would have clearly impacted upper extremity function. Of the 55 amputations, the thumb was involved in 13, and 40 were at levels proximal to the fingertip. While patients with fingertip amputations can live quite normally, a thumb amputation results in considerable decrease in upper extremity function, especially if bilateral, as occurred on one patient in this series. The physical impairment resulting from a thumb amputation at the metacarpophalangeal level, using the American Medical Association permanent physical impairment guidelines,14–16 is 40% hand impairment, 36% upper extremity impairment, or 21.6% for the whole person. A bilateral thumb metacarpophalangeal amputation would then be 43.2% physical impairment for an individual. Similarly, for an amputation of the index or middle finger at the proximal interphalangeal joint level, there is 80% impairment of the finger, 16% of the hand, and 9.5% for the whole person; these numbers for the metacarpophalangeal joint level are 100%, 20%, and 12% respectively.
In this series, the overall female percentage was 65%, lower than in many series.11,17–22 A recent meta-analysis 1 noted a female percentage ranging from 52% to 93.7%. This wide range can be explained by the fact that many recreational and amateur horse-related activities are predominantly undertaken by females. 23 In this study, nonroping injuries were 70% female and 30% male, while male and female injury rates involving roping were nearly equal. Of the 15 males injured where a rope was involved, 5 occurred while roping an animal or during roping competitions.
Being bitten by a horse was rare in this series (4%). There are several studies in the literature of horse bite injuries. In one study, there were 10 horse bites 24 out of 117 injuries. A different study 25 described 24 horse bite injuries in a series of 622 patients. Of these 24, 21 healed uneventfully. However, 1 female sustained a serious bite on the cheek, and 2 men sustained bites on their forearms with extensive muscle damage in both. Another study 26 of 921 animal-related injuries noted 373 animal bites, with 68 due to horses. In a series of 132 animal bites to the head, 27 2 were caused by horses and 1 by a donkey. Both horse and donkey bite injuries 28 were described in yet another study: 24 cases of horse and 5 donkey bite wounds. A breast amputation due to a horse bite 7 has been described as well as an incidental finding of a mucinous carcinoma of the breast 29 in a patient with a history of a horse bite to the breast. In a series of 94 penetrating injuries to the chest in children, 30 1 was due to a horse bite. In our study, there were 5 cases of horse bites; 4 resulted in finger amputations/avulsions, and 1 a thumbnail avulsion with phalangeal fracture. There was 1 breast injury: a skin avulsion from hitting a branch while riding.
Regarding the avulsions in this study, 52% (38 of 73) involved the digits, with the remainder at other locations. The avulsions of the digits, when the location was described, mostly involved the nail/fingertip (18 of 23 or 78%), and fingertip avulsions/injuries can result in morbidity,31,32 due to their sensory and aesthetic functions. Unrecognized nail bed injuries, along with distal phalangeal fractures, 31 can lead to significant long-term morbidity with delayed and secondary procedures 32 for fingernail sequelae with unpredictable final results.
Avenues for injury prevention could be guided by this study's findings. The 76% involvement of a rope in the amputation group is an opportunity for educational programs. “The risk of being dragged if the reins are wrapped around the hand” 12 is noted in the Riding and Road Safety Manual of the British Horse Society. Regarding rodeo/calf roping, counsel could be given as to the most proper method of using the rope, as well as formal guidelines from equestrian societies or event venues. The Pony Club of America Safety manual 33 states, “Never tie or wind the lead rope [the rope attached to a halter and used to lead a horse] around any part of you in any way”. A recent Barrel Horse News internet post 34 stated: “Safety around horses is extremely important. Honestly, it doesn’t matter if you’ve been around them your whole life, or if you’re new to it. Accidents can happen, so be aware of things and try to prevent a wreck or accident before it happens.” They also note that “coils of rope can get wrapped around your hand and result in burns or injury if your horse bolts or spooks.” 34 In 8 finger/hand injuries of intercollegiate rodeo athletes, 35 53% of the injuries used a glove for protection. However, protective gloves may be ineffective 5 against the huge force exerted. The forces encountered during a midshaft femur fracture range from 3.07 to 14.7 kN. 36 These forces to fracture a femur are all less than the 16.54 kN for “heading” (roping the seen with during roping activities. 5 Such forces would likely overwhelm any protective effect of a glove. These injury prevention avenues are simply postulates; it is not known if they would have any impact, as such data is not available in the database.
There are several limitations of this study. First, as it is an ED-based study, the treatment and the final outcomes are unknown. Second, this only involves the acute injury, and it is known that amputation revision is a very common procedure, so the level of amputation described in this study may not be the final amputation level for the patient. Third, the narrative comments were not always adequate to determine the level of acute amputation. Fourth, there may have been other cases of amputations or avulsions that were not coded as such and/or mentioned in the narrative comments and thus would not have been discovered in the data search. There may also have been a case of amputation that was not confirmed. However, this is very unlikely, as the data entered into the NEISS is reviewed daily and then immediately available to Consumer Product Safety Commission staff for further review and analysis. Potential errors, such as illogical coding combinations, are noted by CPSC staff who contact the coder either via computer or telephone to verify and/or correct the entries. Prior studies have indicated NEISS database reports consumer product–related injuries with 89%–98% accuracy.37,38 Fifth, when using the narrative comments, a fall of the rider from the horse vs falls of both the horse and rider cannot be differentiated from one another.” The latter, which often involves being struck or crushed by a 1200 lb animal with steel shoes, will likely result in more severe injuries. Finally, the question of missing data exists. However, in this study, the only missing data were for the patient's race (known for 84 of the 120 patients) and the position on the horse (known for 111 of 120 patients), which does not impact the major findings of the study. The strength of this study is that it is the largest series of amputations/avulsions due to horse-related injuries that gives health care providers an excellent overview of such injuries. Finally, it is baseline data to evaluate the outcomes and/or efficacies of future prevention programs that might be instituted.
In conclusion, this is the largest study to date of amputations and/or avulsions due to horse-associated injuries. For these 120 equine-associated amputations/avulsion injuries, the most common injury mechanism was riding the horse (31%), followed by equipment issues (19%), horse being spooked (18%), being bucked/thrown/kicked off the horse (15%), falling off the horse (11%), and others 6%. A rope/chain was involved in 24%. There were 55 amputations involving the finger (40), thumb (13), and ear and toe (1 each); 15 involved fingertips and the remainder more proximal. A rope was more likely involved in those sustaining amputations compared to avulsions, with males having more rope injuries. This baseline data could be useful to evaluate the outcomes and/or efficacies of future prevention programs.
Footnotes
Ethics Approval Statement
Use of this publicly available de-identified data was considered exempt by our local Institutional Review Board.
Author Contribution(s)
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The authors received no financial support for the research, authorship, and/or publication of this article.
Data Accessibility Statement
These data are publicly available online at https://www.cpsc.gov/cgibin/NEISSQuery/home.aspx. Further details regarding the NEISS data and the acquisition and guidelines for its use are available at 
.