1 Introduction

Aquatic injuries may result from accidents within bodies of water such as oceans, lakes, rivers, and swimming pools.¹ During the early twenty-first century, a significant rise in water-based recreation was observed in the United States (U.S.) with a proportional rise in aquatic injuries.^2,3 Aquatic and high-speed watercraft related accidents are a significant population health risk and are associated with drownings, traumatic brain injuries, infections, extremity fractures, and spinal injury. By 2020, U.S. watercraft sales surpassed $49 billion, a 14.2 % increase from the previous year.⁴ This contributed to the rise in water-based recreational activity, including personal watercraft (PWC), small water vehicles operated by an individual supported on the body of the vessel rather than within a protective hull, and high speed waterboats.⁵ In 2020, there were 1564 PWC accidents, resulting in 66 fatalities, a nearly 34 % year-over-year increase in fatalities.^5,6 The increase in aquatic injuries necessitates additional research in manufacturing safety, policy, use, and overall aquatic awareness.

Miami, Florida leads the U.S in marine expenditures, with over 24 million tourists contributing to over $4.3 billion in water recreation and commerce annually.^4,7 Tourists are often unfamiliar with safety laws required by Florida including licenses, age restrictions, and safety equipment, “No Wake” zones, boating channels, high traffic areas, and hidden sandbars.⁸ The authors’ institution is a Level-1 Trauma Center accepting the largest volume of local and non-local aquatic related injuries in the greater Miami area.⁹

Only one prior study has evaluated the association of spinal fractures and PWC injuries and none have evaluated the association of more general aquatic accidents and spine injuries.^10–13 The purpose of this study was to evaluate the prevalence of aquatic injuries presenting to a single academic level 1 trauma center, with a specific focus on spinal fractures. This present study spans 21-years and is the largest review of such injuries. The primary objective is to describe and clarify the mechanisms of injury (MOI), injury locations, and types of spine fractures associated. Secondary objectives are to identify patient characteristics, associated injuries, local versus non-local residence, and seasonal variability among patients. Information from this study may help policy makers identify trends in PWC use, highlight common spine and associated injuries and mechanisms with the eventual goal of increasing awareness and reducing future accidents and injuries.

2 Methods

3 Results

Of 498 patients admitted after an aquatic accident, 83 (16.7 %) had a spinal fracture. Most patients with spinal fractures were White (61.5 %) males (43.4 %), with a mean age of 38.2 ± 16.0 years. They resided within the greater Miami area (55.4 %) and were hospitalized for an average of 13.4 ± 27.2 days. Of the 83 who sustained a spinal fracture, 21 (25.3 %) had additional fractures. The demographic information of patients with spinal fractures alone or those with concurrent fractures showed no statistically significant difference (Table 1).

Of the 83 patients with a spinal fracture, 47 (56.6 %) returned for at least one follow-up visit. Greater Miami area residents were more likely to return for follow-up visits compared to non-locals (71.7 % versus 37.8 %, p = 0.013, Table 2) and for a greater duration (71.3 ± 123.3 versus 14.3 ± 44.0 days, p = 0.009, Table 2). Among the 21 patients with concurrent fracture(s), there was a trend towards increased local patients attending at least 1 follow-up visit (69.2 % versus 25.0 %, p = 0.080, Table 2. Locals were more likely to follow up for a greater duration compared to non-locals (81.4 ± 158.5 versus 6.1 ± 12.0 days, p = 0.010, Table 2). A total of 44 (53.0 %) patients required surgery for their injuries. Of those who sustained spinal fractures, there was a trend towards more non-locals requiring any surgery compared to locals (54.6 % versus 45.5 %, p = 0.052, Table 2). Non-locals were also more likely to require spine surgery compared to locals (80.8 % versus 19.2 %, p < 0.001, Table 2).

The distribution of associated injuries is available in Table 3. More patients sustained upper extremity fractures (14.5 %) compared to lower extremity fractures (8.4 %) and pelvic ring fractures (8.4 %). There were 12 (14.5 %) upper extremity fractures (2 scapula, 2 clavicle, 3 humerus, 4 radius, 4 ulna, 1 hand/finger), 7 (8.4 %) lower extremity fractures (4 femur, 1 patella, 2 tibia, 2 fibula, 2 foot), 7 (8.4 %) pelvic ring fractures, 3 (3.6 %) acetabulum fractures and 2 (2.4 %) open knee joints. Many spinal fractures occurred in multiple or junctional segments (20 patients, 24.1 %). In descending order, 17 (20.5 %) were classified as subaxial cervical, 17 (20.5 %) lumbar, 14 (16.9 %) thoracic, 11 (13.3 %) upper cervical (C1-2), and 4 (4.8 %) sacral fractures. The upper cervical fractures were sub-classified according to the Gehweiler Classification for Atlas (C1, Table 3a)¹⁴ and Anderson and D'Alonzo classification of odontoid process (C2, Table 3b)¹⁵fractures. The most common C1 fracture was a Type 3 fracture involving the anterior and posterior arch (Jefferson burst fracture). The most common C2 fracture involved the vertebral body or multiple aspects of the C2 vertebra. The subaxial cervical spine fractures (C3-C7) were classified according to the injury morphology aspect of the Subaxial Injury Classification System and Severity Score (SLICS)¹⁶ classification and was most commonly a distraction (29.4 %) injury morphology (Table 3c). Thoracic and lumbar fractures were classified according to the injury morphology aspect of the Thoraco-Lumbar Injury Classification and Severity Sore (TLICS)¹⁷ classification and most commonly a burst (35.3 %) injury morphology (Table 3d).

All patients were sub-classified according to MOI, showing any correlation between injuries sustained and aquatic MOI (Table 4). The most common MOI was boating or personal watercraft (B/PWC) accident (n = 33, 39.8 %), shallow water dive (SWD) (n = 20, 24.1 %), fall/ejection (FE) (n = 16, 19.3 %), and direct collision (DC) (n = 14, 16.9 %). Operative spine injuries were significantly associated with subaxial cervical spine injuries (p < 0.001) and SWD mechanism (p < 0.001). Subaxial cervical spine injuries were most common after SWD accidents (n = 12, 70.6 %), comprising 53.9 % (n = 14) of operative spine cases. Upper cervical spine injuries were frequently observed in B/PWC accidents and SWD (p < 0.001). Thoracic and lumbar injuries showed a significant association with B/PWC accidents (p = 0.017). B/PWC accidents had a higher likelihood of causing thoracic or lumbar injuries (p < 0.001) compared to both direct collision (DC) and fall or ejection (FE). B/PWC accidents were also more likely to result in pelvic ring and acetabulum injuries (p = 0.008) compared to FE MOI and upper extremity injuries (p = 0.045) compared to DC MOI. SWD mechanisms resulted in more operative spine injuries (p < 0.001), upper cervical injuries (p = 0.002), and subaxial cervical injuries (p = 0.020) compared to B/PWC MOI. (Table 4).

Most (n = 56, 67.4 %) patients sustained aquatic associated spinal fractures occurring equally during the spring and summer months. Specifically, May (n = 14; 16.9 %) and July (n = 12; 14.4 %) were the months with the most injuries. The fall season demonstrated the lowest incidence of aquatic related spine injuries (n = 12; 14.4 %) (Fig. 1). Fig. 2 displays the mean age of patients by injury year and demonstrates a positive correlation (R² = 0.2773; p = 0.036). Fig. 3 provides a case example of a thoracolumbar fracture-dislocation sustained in a surfing accident.

Close

Fig. 1 Spine Injuries by Season and Month. Bar and line graph demonstrating the number of patients sustaining spine injuries by season and month.

Export to PPT

Close

Fig. 2 Mean Age by Injury Year. Graphical representation of mean age of patients by injury year.

Export to PPT

Close

Fig. 3 Case example of a thoracolumbar fracture-dislocation sustained in a surfing accident. A 37-year-old man was admitted to our level 1 trauma center on February 21, 2015 following a surfing accident. The patient sustained a fracture-dislocation at L2-L3, with T9-T12 spinous process fractures, left T9-T12 transverse process fractures demonstrated on Sagittal CT (A) and a traumatic durotomy with severe conus medullaris and cauda equina compression (B). Motor examination demonstrated full strength (5/5) in the bilateral upper and lower extremities, with diminished sensation to light touch in the anterior and anterolateral left thigh and leg, otherwise sensation remained intact. He had a negative Hoffman's sign, no clonus, [downgoing babinski,] symmetric reflexes, with tenderness present from lower cervical to lumbar spines. [Rectal tone remained intact.] With a TLICS[citation] score of 9 (translation/rotation morphology 3, injured PLC 3, incomplete cord injury 3), the patient underwent open reduction of the L2-L3 fracture, posterior spinal decompression and instrumented fusion from T11 to L4 with repair of the traumatic durotomy. Postoperative radiographs demonstrate reduction and stabilization T11 to L4 posterior spinal fusion (C and D). The patient sustained the following associated injuries and treatment: Intraventricular hemorrhage with placement of a C1-C2 subarachnoid drain, and multiple rib fractures treated conservatively. Postoperatively, the patient remained stable with stable neurologic function. He remained in the hospital for a total of 10 days with 2 ICU days. Ultimately, the patient was transferred to an inpatient rehab center and pursued additional follow up care near his home in the US Virgin Islands.

Export to PPT

4 Discussion

Aquatic accidents and associated injuries represent significant healthcare challenges. These injuries occur in both local and non-local patients, favoring coastal regions with large bodies of water. Aquatic injuries may involve tourists, high speed watercraft accidents, falls and ejections from PWCs, and may result from improper use of boating and personal watercraft equipment.^18–20 The decision for stabilization versus definitive fixation relies on the treating provider discretion, and represents a significant decision that contributes to treatment complexity, with additional considerations for non-local patients. Aquatic injuries may result in open fracture, posing increased risk to uncommon organisms not often treated in landlocked cities or may result in multiple injuries, require multiple surgeries, with extended lengths of stay.¹³ A recent study reported patients with aquatic acquired spinal fractures required on average 1.2 operations per hospitalization, highlighting the importance of understanding the injury burden and risk factors associated to improve healthcare system evaluation and management and minimize error.^13,18 The current study contributes to the understanding of aquatic injuries, focusing on spinal fractures. The emphasis on patient and seasonal characteristics, MOI, and anatomical distribution of injuries may benefit manufacturers and policymakers in accident prevention, and may assist healthcare workers in optimizing identification, evaluation, and management in these patients.

The results of the current study demonstrate White and Hispanic users in their thirties, residing in the Greater Miami Area and participating in aquatic activities in Spring and Summer are at higher risk for aquatic accidents. Discordantly, prior studies report most aquatic spinal fractures occur in males with mean age in their late twenties.^13,21 Our study demonstrated a significant rise in mean age of patients who have sustained aquatic spinal injuries since 2000, suggesting older individuals participate in recreational aquatic activities more frequently and/or with riskier behavior. Consistent with previous literature, boating and PWC accidents were responsible for most observed injuries.²²

Aquatic spine injuries are relatively rare, with some studies showing 3 % of traumatic spinal injuries from aquatic activity.²³ These incidents are typically overshadowed by more prevalent spinal injuries caused by land motor vehicle accidents and accidental falls.²⁴ Similar to non-aquatic injuries, the most frequent spinal regions injured in our study were thoracic and lumbar, consistent with the most common non-aquatic epidemiological studies demonstrating accidental falls and motor vehicle accidents result in thoracic and lumbar compression fractures or cervical spine injuries, respectively.^21,24 Thoracic and lumbar spine compression or burst fracture morphology is significantly associated with boating and personal watercraft injuries. Prior studies have demonstrated vertebral body injuries are the most devastating injuries in PWC use and are often associated with abdominal and genitourinary injuries.^13,25 This data demonstrates the mechanism of aquatic and non-aquatic spinal injuries are similar when involving a motor vehicle. Therefore, similar mechanisms of passenger protection present in land motor vehicles may be applied to PWC and motorboats, to reduce aquatic injuries.

The current study also demonstrated, compared to thoracic and lumbar spine injuries, cervical spine injuries are less common, result in more variable morphology, and are more likely to result in operative intervention. Nearly three-quarters of the subaxial, and half of the upper cervical spine injuries occurred as a result of shallow water dives.²⁶ Prior literature has demonstrated cervical spine injuries are present across many different aquatic activities, including skimboarding, surfing, and are closely linked to shallow water diving.^26–31 The current paper corroborates prior literature and provides additional evidence that aquatic recreation and motorized activity contribute to cervical injuries at high rates. Such injuries have the potential for severe consequences including neurological deficit and possible paralysis.^{21,28–30,32–34} Although many shallow water spinal injuries are preventable, regional factors may play a role in the observed incidence.³⁵ Cities that provide extensive shallow water recreation (i.e. surfing, skimboarding, sand bars, shallow beaches, etc.) may predispose individuals to such injuries and should be properly identified and considered in state or city policy surrounding water recreation.^36,37

A majority of patients who underwent operative intervention for a spine fracture were involved in a SWD mechanism associated with subaxial cervical spine location of injury. This is substantiated in the morphology of injury with cervical spine injuries resulting in greater incidence of unstable distraction, translation or rotation, and stable or unstable burst fracture patterns. In this setting, while more conservative treatment modalities may be used in stable lumbar spine injuries in the absence of neurologic compromise, cervical spine injuries exhibit a lower threshold for operative reduction and or decompression when involving the spinal cord.³⁸

Few prior studies have evaluated aquatic spine injuries. Similar to one current study by Donnally et al., thoracic and lumbar injuries were most prevalent, and the axial load mechanism was most associated with vertebral body burst fracture. There was a significant association between direct collision mechanism of injury and increased Injury Severity Score (ISS) in patients with spine fractures with and an associated increased risk of visceral organ damage.¹³ Finally, the frequency of aquatic spine injuries was associated with time of year with two-thirds of mortality events occurring in the spring, largely attributed to spring break vacations.¹³

Correspondingly, we evaluated the impact of tourism and season on aquatic activity and injury incidence in our population. Our findings revealed more than 60 % of injuries occurred in spring and summer months and nearly 45 % of patients were non-local. Contrary to our hypothesis in which locals would have a lower risk and incidence of aquatic injury, our analysis demonstrated no significant difference between groups. In contrast, non-local patients were more likely to require operative intervention for spinal fractures compared to local patients. This contributes to treatment complexity as patients are often visiting for a short period and as a result of injury, require extended hospital lengths of stay, medical transportation to their primary residence, and limit follow up with their operative surgeon. Despite these findings, locals have more exposure events in the aquatic environment, and thus may be less likely to sustain injury per exposure. Similarly, it may be worthwhile to evaluate a learning curve for operating PWC, navigating various aquatic situations, as well as the common local and non-local behavioral elements when participating in aquatic recreation.

5 Conclusion

Aquatic associated spinal injuries in an environment with high tourist density pose a significant challenge in treatment algorithms. Spinal injuries from aquatic activities depend on the mechanism of injury with thoracic and lumbar injuries most common from boating and personal watercraft accidents, and operative subaxial cervical spine injuries most commonly occurring as the result of shallow water dives. Regardless of residence, a similar proportion of injuries occur in local and non-local individuals with the greatest percentage of injuries in white and Hispanic people in their late thirties occurring in Spring and Summer. Non-local individuals who sustain spinal fractures are more likely to require surgery and less likely to follow-up with their primary surgeon. The observed increase in aquatic injuries necessitates additional research regarding safe manufacturing, policy, use, and overall aquatic awareness. Future studies are worthwhile: evaluating effects of local residence exposure events and injury per exposure between based on residence, a learning curve for PWCs, and local and non-local behavioral elements with aquatic recreation and their impact. Lastly, information from this study may help policy makers emphasize education, awareness, safety, and personal protective equipment with the eventual goal of reducing future accidents, consequent spinal injuries, and improving treatment.

CRediT authorship contribution statement

Dustin H. Massel: Conceptualization, Methodology, Supervision, Writing – original draft, Writing – review & editing. Jaxon D. Wagner: Conceptualization, Data curation, Investigation, Writing – original draft, Writing – review & editing. Jonathan Weinerman: Formal analysis, Writing – review & editing. Joseph P. Costello: Data curation, Investigation, Writing – original draft, Writing – review & editing. Aziz T. Shittu: Investigation, Writing – original draft, Writing – review & editing. Alexander J. Butler: Writing – review & editing. Augustus J. Rush: Writing – review & editing. Chester J. Donnally: Conceptualization, Methodology, Supervision, Writing – review & editing.

Guardian/patient's consent

All procedures performed in this study involving human participants were conducted in accordance with relevant institutional and national research committee standards. Institutional review board approval was obtained prior to initiation of the study (IRB# 20190977, approved February 7, 2020). As this study involved a retrospective review of de-identified patient data, the requirement for informed consent was waived by the institutional review board.

Data availability

The data that support the findings of this study are available from the corresponding author upon reasonable request. The data are not publicly available due to restrictions imposed by the Institutional Review Board (IRB) approval, which ensures the confidentiality and privacy of the participants. Researchers interested in accessing the data for verification of results, replication of the analysis, or conducting secondary analyses should contact the corresponding author directly.

Ethical statement

This study was conducted in compliance with relevant institutional and national research committee standards. Institutional review board approval was obtained prior to the initiation of the study (IRB# 20190977, approved February 7, 2020). As the study involved a retrospective review of de-identified patient data, the requirement for informed consent was waived by the institutional review board.

Funding statement

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.