Documentation:FIB book/Cervical Spine Injuries from Headfirst Diving in Shallow Water

Cervical spine injuries have the potential to be life-altering by causing paralysis or even death. Diving headfirst into shallow water is a very common way of incurring said injury as the energy of the entire body must be absorbed by the head and neck, which are not capable of withstanding large loads. For differing reasons, young males are much more likely to sustain such injuries, as found in epidemiological studies. It is also known that the majority of diving injuries occur in shallow water and natural environments such as lakes and rivers, which may be due to the lack of visibility.

Due to the complexity of researching diving-induced cervical spine injuries, in which one must deal with the safety of volunteers, the lack of muscle activation in post-mortem human surrogates, the speed gained by the diver in the air and the drag produced by the water, many different testing methods are used in conjunction to provide biomechanical data. The testing methods include: human volunteers diving with imaginary pool bottoms, final element analysis of the complete human cervical spine, and dive-like impacts applied to cadaveric spines.

The injuries that occur in diving are mostly localized to the C4-C6 vertebra. There are many different types of injuries, including wedge fractures, burst fractures, ligament tears, dislocations, etc. and various modes of injury. The most common is compression-flexion of the neck (in which the chin is pushed towards the chest). Although the presence of water creates drag on the subject, diving posture minimizes drag forces and therefore still results in a large impact speed. Many cadavers tests have been conducted to identify the impact force required to cause spinal injury, results of which are reflected in spine injury criteria.

Despite the severity and relatively common occurrence of cervical spinal injuries due to diving, this is a preventable injury. Engineering and regulatory design, diving technique, education and administrative regulation all have the ability to greatly reduce the occurrence of such injuries. Although there has been a sufficient amount of research done on these injuries, the need for a study involving all the most important factors of a diving injury (presence of water, full human model/cadaver, muscle activation, etc.) has yet to be addressed and should be focused on in future research.

Relevance

Headfirst diving injuries can range from concussions to debilitating neurological damage such as tetraplegia, or even death. Such diving accidents can damage both the spinal cord and the spinal column. This is cause for grave emotional and economic effects on those affected. Furthermore, it is estimated that diving accounts for up to 21% of all spinal cord injuries (SCI), depending on the country observed . In the United States, for example, diving is the fourth-largest cause of SCI, accounting for 8.5% of all SCIs ever since 1973 . This can be debilitating, particularly for the young. As reflected in results of the American Spinal Cord Injury Association (ASIA) Impairment Score, diving accidents tend to produce more severe and lasting effects.

Diving injury can affect the spine from the start of the cervical spine to the start of the thoracic spine. On top of the neurological injuries previously listed, other physiological damages due to SCIs include changes to the composition of muscle and fat, as well as changes to hormonal homeostasis. Though this article will not explore these secondary physiological changes, SCI ought to be carefully examined due to its impact on quality of life, and particularly diving-related SCI due to its high prevalence.

Epidemiology

Rate of Occurrence and Demographics

Cervical injury due to shallow-water diving can be seen in many areas of the world. The table below summarizes such occurrences in several representative regions. Some of these references have also been noted by Blanksby et al. which shows a wide-ranging variety of epidemiological figures. Though each study has a varying sample size, the most prevalent demographic is in males aged 20±5 years [table]. Having surveyed 1106 SCI victims resulting from diving, DeVivo et al. summarize that the average diving-related SCI patient in the US, is typically younger and more likely to be a white male, compared to other people with SCI in general.

Sample of Cervical Injury in Europe and North America

Country and Hospital	Years of Study	Number of Diving-Injuries	Most Common Demographic	Source
12 Hospitals throughout Sweden	1970-1979	17 cases (of 332 observed)	25-year old males (17%)	Ersmark et al.
Spinal Unit of Florence, Italy	1978-2002	1158 cases 65 resulted in spinal lesion	Males at mean age of 21	Aito et al.
Hospital Geral de Santo Antonio, Portugal	1960-1980	4 cases due to general watersports (of 162 observed)	Males aged 15-22	Gaspar et al.
Stoke Mandeville Hospital, United Kingdom	1944-1977	150 cases	Males (95%) aged 15-24 (67%)	Frankel et al.
Los Angeles County-University of Southern California Medical Center and the Rancho Los Amigos Hospital, USA	1970-1973	61 cases (of 356 observed)	Males aged 16-25 Males five times more at risk than females	Heiden et al.
Toronto General Hospital and Sunnybrook Hospital, Canada	1948-1973	38 cases (of 358 observed)	Males 4.5 times more frequent than females	Tator et al.

Furthermore, another study by Aito et al., reveals that out of all sports and leisurely activities, diving is that which affects young people most commonly, and also results in the majority of neurologically incomplete cervical SCI cases (i.e.: ASIA Injury Scale B). The authors suspect the cause is lack of visibility under the water, especially if diving outdoors. In the same paper, the authors estimate that in general, traumatic spinal cord injury (TSCI) affects 17.5 patients per million in Europe, and 40 patients per million in North America.

An interesting epidemiological comparison can be made between Aito et al.'s 2005 paper, and their 2014 paper, which were both conducted at the same hospital. In the 2014 paper, from the years 1981 to 2010, it was found that there were 102 cases of sports-related injuries. Out of that: 78 cases were of male patients, 24 were of female patient; the age range was from 14 to 51,with the median at 26; 74 of the cases resulted in tetraplegia, 28 in paraplegia; and, 26 cases came with associated injuries. In their 2005 paper, particularly in diving accidents, the mean age is 21, and most patients are male. Rarely is there an associated injury. Comparative ASIA injury scores are shown below.

Comparative Injury Severity Distributions of General Sports v. Diving

Score	Sport-related SCI in general 102 cases	Diving-related SCI 65 cases
ASIA A	42%	53%
ASIA B	22%	25%
ASIA C	8%	11%
ASIA D	18%	11%
ASIA E	10%	0%

Environmental Conditions

Distribution of injured divers per pool depth per Albrand and Walter . Plotted by Emilie Boras.

Environmental conditions greatly affect the risk of diving cervical spine injuries. The type of body of water and its depth are important factors in such injuries. McElhaney et al. reported 41 of 57 diving injuries documented occurred from the pool side into 4 feet of water or less, which was corroborated by DeVivo et al.. Blanksby et al. reported that 89% of all diving-related SCI cases happen in waters 1.52 m, or 5ft deep, or less. Albrand and Walter documented the distribution of injury per pool depth as seen in the adjacent figure. They found 72% of divers from their sample were injured in the natural environment (rivers, lakes, and reservoirs) while the remaining were injured in bodies of water intended for water sport (swimming pools or above ground pools).

Frankel et al. surveyed the British population. From their study, two-thirds of patients received an SCI because while diving into seas and rivers. Patients were generally unaware of submerged objects, and grossly misestimated water depths. They also note that many injuries are caused by people not heeding warning signs as many incidents occur during vacations; as such they recommend increased education to cultivate public awareness. Moreover, the researchers found that of the 150 cases observed, 27 happened explicitly in swimming pools; 13 cases happened at the shallow ends, 6 at the deep end, and 8 from other causes. Many of the cases from the shallow end happened because small children were first learning how to dive there.

DeVivo et al.'s survey of an American demographic's 1106 spinal injury cases found that there were select major factors which featured in the instances of injury, as listed in the table below.

Factors of Injury in DeVivo et al.

Condition Type	Occurence
Water Environment and Surroundings	57.1% of cases were in a natural body of water, whereas only 30.8% were in a swimming pool
Safety, Labelling, and Enclosures	57.2% of cases happened at a water depth not exceeding 4 feet, 87.4% of cases happened where no depth warning label was present
Substance Use	Alcohol consumption was reported 48.9% of the time, whereas drug usage was reported 2.2% of the time; this can be contrasted to the fact that 46.1% of injuries happened at parties
Timing	52.1% of cases occurred during daylight hours, 51.0% of cases occurred during the weekend, and 82.0% of cases occurred during summer vacation.
Familiarity with diving environement	In 44.0% of cases, the injury was at the person's first visit to the particular pool
Diving Technique	In 69.7% of cases, the patient entered the water using a normal diving technique

Biomechanical Study

Spinal injuries in diving occur when the momentum of the diver and resultant impact with the water’s bottom surface is translated through the diver’s head into the rest of their body. The sudden arrestation of the head’s motion causes the remainder of the forward moving body to “crash” into the neck and exert a moment and compressive force on the body’s cervical spinal column. This is a well defined situation between a diver, a body of water, and the body of water's bottom which lends itself relatively well to biomechanical testing and analysis.

However, studying the injury biomechanics of headfirst diving spinal injuries presents a unique challenge. We can study the kinematics of headfirst diving with human volunteers to non injurious depths and derive conditions leading up to injury. But to evaluate the forces and injuries produced from the dive we must use other human surrogates, such as cadavers. These surrogates cannot easily replicate the necessary dive conditions given their inability for muscle activation, which is necessary to maintain a diving form throughout the dive. This forces the division of the study into two parts: A kinematics portion with human volunteers and a a kinetics portion with other human surrogates.

Thus, in order to study headfirst diving injury mechanisms it is necessary to:

1. Extract headfirst dive conditions through past injurious incidents

2. Have human volunteers replicate injurious dives in deeper waters to measure kinematic data

3. Apply the kinematic data to cadaver or other human surrogate testing to produce kinetic data

4. Derive a relationship between injury and loads produced as a function of dive parameters

These items are explored in various papers but none has combined the four above steps.

Testing Methods

Diving spinal cord injuries have been studied using a variety of human surrogates for biomechanical study including human volunteers in non-injurious tests, post-mortem human surrogates (PMHS), and computational models. The use of ATDs and animal tests do not feature heavily in the diving injury research literature.

Human Volunteers

Ethical restrictions greatly limit the scope in which human volunteers can be used for testing but they still provide a valuable supplement to other forms of research.

McElhaney et al. re-created 9 diving scenarios representative of 57 injurious diving incidents using anthropometrically similar human volunteers diving into deeper waters. The divers were video captured set against a grid in order to extract dive velocities and positions. Dive results were used to validate a mathematical model predicting dive trajectory and velocity.

Albrand and Walter studied diving kinematics in a similar fashion. Divers were filmed diving into a pool set against a grid by a 24-frames per second film (which was considered high speed in 1975). Diving height was varied from pool side, a 1 meter platform, a 1 meter springboard, a 3 meter springboard, and a 5 meter platform. The frames captured were used to reconstruct the velocities and deceleration curves.

Perrine et al. conducted a study in which human volunteers performed shallow-entry dives into a deep pool that contained a marking at 107cm (the depth of most above-ground pools and the depth of the shallow end of most in-ground pools). The aim of the study was to assess how Blood Alcohol Concentration (BAC) effects a person's ability to perform a safe shallow-entry dive. It concludes that the diving performances of participants was significantly degraded under the influence of alcohol. The dives were filmed underwater and classified as "no contact", "light contact" or "heavy contact" with the imaginary 107cm-deep pool bottom. Light contact dives were assumed to indicate a minor risk of injury and heavy contact dives a substantial risk of injury, however, the justification for this distinction is not included in the article. It is also stated that "the specific mechanisms underlying the performance impairment remain unknown and the epidemiological data necessary to evaluate the accident prevalence rates adequately are limited".

Computer Modelling

There are currently two numerical methods that are used in research to simulate the mechanical behavior of the human spine: the finite-element method (FEM) and multibody simulation (MBS) . For each method, one paper that is relevant to diving injuries is reviewed.

Finite Element Method

Tchako and Sadegh used ANSYS 5.6 to create a complete and detailed three-dimensional finite element model of the human cervical spine (C1-C7), including soft and hard tissues. The model was used to assess modes of injury in sport, including diving. Geometric data was obtained from segmenting CT scans and material property data was obtained from other literature. The muscles are created with three-dimensional tension-only cable elements to simulate passive function. This was done because, according the authors, active function of the muscles is an intentional act that could not occur during sudden and unexpected loading. To simulate diving injury, the model was used to evaluate stress due to compression-flexion in two cases: free movement of the head after ducking and locking of the upper vertebrae (C1-C2). In both cases, the model was constrained at T1, C7 and C6 in the anterior-posterior and lateral directions (but had freedom to translate in the inferior-superior direction). A compressive load of 450N was quasi-statically applied to T1 in progression.

This study highlights the advantages and disadvantages of finite-element models. The model is highly detailed with regard to geometric accuracy and it allows for the observation of stress/strain concentrations before failure. However, there are several substantial sources of error. First the material properties are constant and do not account for subject-subject variation. Secondly, the boundary conditions and initial conditions are not well-defined. And third, the model is tested with quasi-static loading rather than impact loading. The latter two sources of error are a result of the simulations being performed to align with a cadaveric study performed by Bauze and Ardan in 1978 . The goal of Bauze and Ardan’s work was to produce forward dislocation in a cadaver spine due to compression-flexion loading as it occurs in a diving accident. This study assessed a scenario in which the cadaver is vertical with the head in a “ducking” position (the degree of ducking was not quantitatively defined) under a quasi-static increasing compression load. This does not represent a true diving accident, where the victim would not be completely vertical, the position of the head would be case-specific, and the load would be an impact rather than static. The most appropriate boundary conditions are unknown, and Tchako and Sadegh found that the maximum stress in the spine was 50% lower when the model was constrained at C1-C2 and T1-C6 compared to only constrained at T1-C6.

Multibody Simulation

Silvestros et al. use a multibody musculoskeletal approach to model the human cervical spine during axial impacts. The magnitude of impact is intended to replicate sports injuries. This study aims to address the issue that current impact specific musculoskeletal models are lacking because cervical spine joint dynamics are often informed by unrepresentative quasi-static or static experiments (this limitation is also seen in the finite-element model by Tchako and Sadegh). For instance, Nightengale et al. used a musculoskeletal model to investigate axial impacts with a peak force of 2000N reached in 0.01s , however, the model was developed from experiments with a peak force of 200N reached in 2.0s . It is unclear if the data may be extrapolated to that degree.

To avoid such uncertainty, Silvestros et al. used porcine cervical spine specimens to estimate the visco-elastic properties of the spine’s joints at the relevant loads. However, this assumes similarity between human and porcine spines which introduces a new source of error. Furthermore, the loading parameters used are based on rugby collisions, which can only be cautiously applied to diving. To simulate the impact, an 80N load was dropped from a height of 0.5m (resulting in an impact speed of 3.1m/s) and the peak resultant forces were 3.0-4.8kN. As seen in the Impact Forces section below, these values are within the same range of what is commonly tested for diving injuries, but the head was in a neutral position rather than in flexion.

Discussion

Computer simulations are a highly valuable tool for fast and inexpensive experiments. However, their usefulness is limited by the biomechanical data that is available to build the model. Nightingale et al. determined that muscle activation and flexion angle substantially influence load magnitude in impacts to the cervical spine, which are two factors that are not considered in the studies above. For example, the influence of flexion angle on the maximum compression force in C4-C5 vertebrae is plotted below.

Maximum compression force in the C4-C5 vertebrae plotted against flexion angle of the neck. Adapted from Nightingale et al.

This provides insight into the complexity of replicating cervical spine dynamics and how a wide variety of clinically observed injuries may result from seemingly similar loading conditions. The next step in creating a comprehensive cervical spine model is to combine a multibody simulation with the finite element method . The multibody simulation would be used to determine mechanical response on a macro-scale, while the finite-element method would be applied to specific regions of interest (such as common fracture locations). In this way, the multibody simulation would provide the boundary conditions and initial conditions for the local finite element analysis. Ultimately however, the success of this approach still depends on the quality of data obtained from other means of testing.

Post-Mortem Human Surrogates & Cadaveric Spines

Testing full Post-Mortem Human Surrogates (PMHS), or cadavers, can prove the most biofidelic recreation of a scenario but may also prove difficult and impractical in practice. In order to accurately model the true response of the cervical spine to impacts experienced in diving, it is important to consider which components of the cadaver are important to include in order to produce a biofidelic response. Nusholtz et al. used an entire cadaver in order to produce proper responses from the head, neck and thorax. However, due to the little additional value presented by and the difficulty of using full cadavers, many studies have opted to refine study to a head-spine complex with additional supports to model proper anthropometry. Cadaveric cervical spines are produced by transecting the spine at an upper thoracic vertebra (usually T2-T4), and casting the the bottom most vertebra into resins held by metal enclosures . As an alternative to a metal enclosure, an acrylic sleeve around the spine along with a pin through the lowest vertebral body to prevent rotation has been utilized . The muscles and ligaments surrounding the spine, down until the upper thoracic vertebra, remain intact for the testing in order to maintain the biofidelity of the spine, even though the muscles will not be engaged. The head is occasionally removed as well. This dissection of the head-spine complex provides a component that is independent of body parts which would not affect the kinetics of the cervical spine, while maintaining essential components intact. Different studies implemented different cadaveric test setups and orientations, such as the head being inverted and attached to a track or being held in place by a spring. However, the cadaveric head-spine complex used in testing is generally quite similar between experiments investigating the response of the human cervical spine.

Different methods are used to support and pre-condition cadaver spines for testing, as the most accurate method has not yet been established. Researchers identify what may be an appropriate physiological response in given situations and prepare the specimens based off these assumptions in an attempt to increase the biofidelity of the test. For example, Nightingale et al. maintained the neutral position of the cervical spine through the use of sutures and the cervical spine was manually moved through flexion and extension of 60 degrees for 50 cycles in order to accurately model an alive human . On the other hand, Pintar et al. placed a "halo ring" around the skull with a weight attached by a string over a pulley to the front of the skull and a spring attached to the back of the skull, thus producing a model of the musculature in the head and neck . In the same study, the table that the cadaver was situated on, was adjusted to remove lordosis (inward curve of the lumbar spine) in the column . In another study, the bottom of the cadaveric spine (T2-T4) was rigidly attached while the top of the cadaveric spine (basiocciput) was allowed to move along a lubricated surface and a flexed position of the cadaveric spine was achieved through a jack . This allowed for the specific physiological response desired as forward dislocation of the cervical vertebra was achieved as vertical compression, flexion and horizontal shear forces were produced . These different methods attempting to produce a biofidelic response are a source of controversy in this field of research.

Despite currently providing the most biofidelic results, there are important shortcomings associated with using post-mortem human surrogates. Muscles are important for stabilization of the neck, head, and spine during impacts, and a lack of muscle activation and muscle degradation in PMHS makes this is a shortcoming of cadaveric spines. There are two factors which allow for accurate use of cadaveric spines in research without need for accurate muscle response: a short injury occurrence and compressive loading . The role of muscles in strengthening the neck is lowered in compressive loading and the time for injuries to occur is much lower than the typical muscle reflex time, thus suggesting that the muscles would not be active at the time of injury and therefore have less effect on the response to the impulse. Due to the lack of muscle activation, PMHS specimens are often held in position with sutures that maintain the alignment of the head to provide a repeatable impact orientation . Additionally, an important aspect to consider in PMHS testing is the age of the cadavers used. Nightingale et al. point to the fact that the older age of cadavers tends to potentially skew results, and they suggest a peak loading force for young males which is 20-30% higher than that which they postulate for the older male cadavers tested.

Anthropometric Test Devices

Anthropometric test devices (ATDs) are largely intended for use in motor vehicle testing however they have been used alongside PMHS tests in order to confirm or deny biofidelity. Given the complexity of the spine is hard to properly replicate in ATDs and the difference in loading during car crashes versus head first diving, ATDs have not been very useful in testing diving impact injuries thus far. Pintar et al. did test Hybrid III ATDs in their study of compressive impact loads but only alongside cadaveric spine datums. They later suggested a corridor for spine displacement versus force for design of ATDs.

Injury Biomechanics

Injury

Region

The Human Spinal Column with Labelled Vertebrae

Spinal fractures due to headfirst diving generally occur in the lower cervical vertebrae from C4-C6 . These spine vertebrae are evident in Figure 2. This is as C4 to C7 serve as the junction from the pliable cervical spine to the stiffer thoracic spine. The bending induced in this region causes fracture primarily from C4 to C6. McElhaney et al. presented the locations of 67 cervical spine injuries sustained in diving and sliding accidents, 60 of which were diving accidents. DeVivo et al. presented the location of 192 diving spinal injuries as percentages per vertebral level. Injuries at C4-C6 accounted for 79% of those sustained. Both distributions are evident in the table below.

Level of Fractures

Vertebral Level	Number of Injuries (sample of 67)	Percentage of Injuries (sample of 192)
C1	0	1
C1-2	2	-
C2-3	1	0.5
C2-3-4	1	-
C4	1	17
C4-5	10	-
C4-5-6	1	-
C5	19	35.1
C5-6	13	-
C5-7	1	-
C6	8	26.8
C6-7	7	-
C7	1	12.4
C8	-	3.1
T4, 5, 6, 7	1	-

These lower cervical spine injuries can result in permanent tetraplegia or death.

Mode and Type

Head impact on the rigid pool bottom at some anterior or posterior angle from the spine vertex causes compression as well as potential flexion or extension. Lateral rotation may also occur, but it is not the primary rotation of focus, as it is smaller in magnitude. The adjacent figure illustrates the possible types of loading given a certain dive technique.

Dive position variations and resulting loading modes effects on spinal injury modes. Created by Emilie Boras.

A compression-flexion or hyperflexion loading is produced when the diver enters at an angle and tucks their head towards their chest. This load path usually results in wedge fractures and dislocation ,. A more vertical entry and upright posture will result in purer compression and can result in burst fractures, a fracture noted as "characteristic of diving". Finally, a compression-extension loading, or hyperextension, occurs when the diver's head is up and tilted backwards. This can also result in fracture, albeit Albrand and Walter did not cite which might specifically occur.

McElhaney et al. reported 63 of the 67 injuries sustained they surveyed were flexion-compression injuries. They remarked on the similarity of the type of injuries in this region due to diving. Albrand and Walter also found this to be the most common injury mode. This is due to the angle of dive entry and diver's tending to tuck in their head. Impact with the pool bottom in this position causes neck compression and flexion as the body's inertia continues forward but is stopped via loads transmitted through the neck. The specific diving scenario can result in any variety of structural injuries, such as wedge fractures, burst fractures, tear-drop fractures, articular facets fractures, dislocation, distraction, intervertebral disc rupture and herniation, ruptured ligaments, and damage to muscle tissue. Nightingale et al. discuss how the kinematic complexity of cervical spine dynamics contributes to the broad variety in injury locations and compressive loading mechanisms. Fracture type distributions are evident in the table below.

Prevalence of Different Fracture Types

Injury Types	Aito et al. 65 cases	Kewalramani et al. 45 cases
Wedge or teardrop Fracture	61%	80%
Burst Fracture	21%	20%
Other Fracture:	14%	-
Pure Dislocation:	2%	-
No Visible Fracture:	2%	-

Aito et al. found injuries in addition to primary spinal injuries were rare, with only 6% of cases suffering from head trauma as well. 36 of the 65 patients studied required surgery. Of the patients who required surgery, 47% developed kyphosis, whereas of those who did not receive surgery, 62% developed kyphosis. McElhaney et al. describe how injury to the spinal column can result in subsequent functional spinal cord injury:

"An anterior fragment may displace forward producing a "teardrop" fracture while the posterior fragment displaces back ward into the spinal cord resulting in cord compression".

Although spinal fractures and injury can occur without resultant spinal cord injury and permanent neurological damage, diving injuries appear to usually result in spinal cord injury as well, as McElhaney et al. found 60/67 injuries resulted in tetraplegia, 2/67 resulted in permanent neurological deficiency and only 4/67 resulted in no permanent neurological damage. High incidences of spinal cord damage is due to the types of spinal column injuries sustained in diving. Certain fractures, such as burst fractures, and dislocations which are common in incidences of diving spinal injury impinge on the spinal cord and are to blame for neurological damage.

Kinematics

Speeds

The typical headfirst diving scenario sees the subject gain some velocity from a jump or fall through air before they enter the water and are slowed by hydrodynamic drag. The more streamlined the diver's form, the more drag is minimized. Minimizing drag results in impacting the pool bottom at the fastest possible speed and increases chance of injury. This worst case minimal drag form is characterized by the diver traveling with their sagittal plane perpendicular to the pool bottom and their coronal plane at some angle of entry with the pool bottom. This orientation causes the resultant head impact with the pool bottom to exert a compressive force along the diver's spine as well as a flexion or extension moment due to the spine's inherent eccentricity and the diver body's inertia. McElhaney et al. described this stop in head motion while the remainder of the body continues forward and downward as a "snap-roll motion".

McElhaney et al. attempted to describe the dive kinematics with a mathematical model predicting dive trajectory and velocity. They described how subjects first gain velocity by falling through the air after the initial jump, gaining velocity per

$v^{2}=2gh$

where v is body velocity, g is acceleration due to gravity, and h is the fall height. The height of the diver center of gravity's fall through air ranged from 1.2 m to 2.2 m.

Air drag imparts minimal effects at these small heights. Subsequent impact with the water begins to impart drag forces, effectively slowing diver descent. Pressure drag is the primary drag component and so diver position greatly affects deceleration. Velocity is dictated by

$V=V_{o}e^{-0.22iqs}$

where q is body cross section, i is a form coefficient, s is the path length, and $V_{o}$is the initial velocity.

From this they found velocity estimates per water depth. A flat body dive only retained 1% of initial speed at a depth of 4 meters while a head first dive retained 59% of its speed at the same depth. This highlights the importance of diver form with respect to impact velocities.

The estimated head impact speeds for poolside volunteer dives ranged from 311 cm/sec to 655 cm/sec. Springboard dives ranged from 380 cm/sec to 810 cm/sec.

They postulate tolerable impact velocities for headfirst dives are less than 311 cm/sec. The sustained forces from these impacts are less than those necessary for head trauma but are sufficient to cause neck injury. This is why so few divers with spinal cord injuries also have skull fractures.

Two limitations of this models are the fact it does not account for changing form coefficient during the dive as the diver changes position and angle as well as the fact it does not account for the time at which the body is partially immersed, which is the most significant in shallow dives.

Albrand and Walter found the expected velocities at time of impact for recorded injurious dives as a function of initial height above the water and water depth. They predicted that divers impacting depths of 2 feet would do so at 15 ft/sec or 4.6 m/sec and those impacting depths of 4 feet would do so at speeds of approximately 13 ft/sec or 4 m/sec. This generally agrees with McElhaney et al. 's postulation that 3.1 m/sec is a sufficiently large speed of impact to induce cervical spine injury and the fact most injurious dives occur at depths of 4 meters or less.

Impact Forces

The forces experienced within the vertebral column during an injurious diving incident cannot be ethically measured in human volunteers and cannot always be properly replicated using whole cadavers, computer models, or animal models. Therefore, extracting forces experienced upon diving impact is usually best done thus far by testing partial human cadaveric spines with loading conditions akin to those experienced during headfirst diving. However, most cadaveric spine loading experiments are not event-specific and offer generalized loading results. Therefore such experiments must be selected for analysis on the basis of their similitude to diving conditions. Important aspects to replicate are the rates and orientation of loading. Diving spinal injuries generally occur with the spine in a posture which induces compression-flexion (CF) and the velocities of impact generally range from 3.1 m/sec to 4.6 m/sec . The table below compares results from various spine impact experiments which can be applied to diving.

Comparison of Cadaveric Spine Impact Loading Simulating Diving and Resultant Loads and Injuries

Study	Test Setup	Cadaver Posture	Linear Speed of Impact (m/s)	Peak Forces (N)	Resultant Injuries (Focus on compression and flexion-compression)
Biodynamics of the Total Human Cadaveric Cervical Spine	Isolated T2 – head spine Preload weights and spring to simulate musculature	Vertical No lordosis Akin to compression posture	Range: 2.95-8.13	Actuator force: 5856 (3 m/s)  - 19205 (7 m/s) Distal Z force (measured at T2): 1177 (5.5 m/s) - 6193 (7 m/s)	From C3-C6: Vertical fractures Burst fractures Wedge fractures Spinous process fractures Compression fractures Reduction in disc space Dislocation Cord compression
Cervical Spine Injury Mechanisms	Full Cadaver Free-fall drop onto padded rigid structure Low-drop and high-drop tests	Supine position Upside-down Impact along sagittal plane at angle to produce CF	Range: 1.4 – 5.9	300 – 10 800	C6, C7,T2, T3, and T4: Various fractures Ligament damage (Note: Re-used cadavers for multiple tests)
Dynamic Characteristics of the Human Cervical Spine	Isolated cadaver head neck complexes Subject to crown impact	11 compression orientations 3 extension orientations due to slight lordosis 5 flexion orientations due to slight kyphosis	Range: 2.5-8	Mean force at failure: 3326 Compression and flexion type range at failure: 744-5179	Primarily C3 – C5: Wedge fractures Burst fractures Vertical fractures Ligament tears (posterior ligamentous damage in flexion)
Dynamic Responses of the Head and Cervical Spine to Axial Impact Loading	Isolated cadaver head spine complexes Transected at T3-T4 T3-T2 constrained Ear and nose sutures to support head Dropped from 0.33m, 0.53m, and 0.63m	Anatomically neutral position	Range: 2.43-3.51 Mean: 3.2	Mean neck force:1727+/-387 Ranged: 1289-4189 Mean rigid impact head force:8306+/-1796   Mean padded impact head force: 2691+/-968	2.43 m/s, 3.14m/s, and 3.28 m/s “large head motion” flexion tests: No injury 3.07-3.26 m/s flexion tests: C1 fractures and C2 hangman fractures due to local extension C1 fracture due to compression C2-C4 tears and chips No "major head flexion" produced CF injuries
Injury Biomechanics of the Human Cervical Column	Cadaveric neck head complexes Transected at T2-T3 Thoracic end constrained Preload weights and spring to simulate musculature	Vertically erect Dynamic impact resulted in CF	Range: 3.2-5.7	1700 - 3200	C3 vertical fracture C5 burst fracture C4 wedge fracture
The Dynamic Responses of the Cervical Spine: Buckling, End Conditions, and Tolerance in Compressive Impacts	Cadaveric head and neck complexes Upwards of T1 Impacted with compliant/rigid surfaces	Anterior, vertex-aligned, and posterior positions Aimed to create extension and flexion	3.14+/- 0.19	Mean head impact load: 5800+/-2500   Mean neck failure load: 1900+/-700 Female Mean: 1061+/-273 Male Mean: 2243+/-572 Sugg. young male mean:3790+/-150	C3 burst fracture C6-7 facet dislocation C1 fractures Posterior ligament damage C5 & C7 burst fractures

The location of the force measurement affects the reading, whether it is recorded at the impactor, at a specific vertebrae, or at the end of the spine segment tested. Failure loads generated in the above tests generally ranged from approximately 1 to 4 kN, depending on the circumstances of loading and the cadaver used. All but few resulted in severe injuries. The tests resulting in compression or compression-flexion injuries were focused on in the resultant injuries. There is some variation within the vertebral location injured compared to epidemiologic data, as there are injuries at C1, C2, C3, and in the thoracic vertebra. However these were largely either due to local extension within the spine within a larger flexion-compression injury or due to solely compressive injuries bordering on extension . Otherwise the injury locations agree with those seen in the literature and the injury types, such as burst and wedge fractures, replicate those seen in real world cases.

Pertinent Injury Criterion

Forces and moments experienced in headfirst diving can be translated into risks of injury using injury criteria. Though there is significant controversy as to the best predictive neck injury criteria and thresholds, the injury criterion most pertinent to cervical spine injuries due to headfirst diving appears to be the Neck Injury Criteria (Nij). Although Nij is intended to predict injury for car crashes, it is still useful in studying diving spinal injuries. However Nij was not applied in any of the biomechanic papers analyzed. HIC was occasionally cited and is a decent means of quantifying the force applied to the cadaver spine setup through the head but does not allude to chance of neck injury. The recorded forces for failure do appear reasonable with respect to the AIS 3+ injury envelope developed for Nij. The peak compressive force envelope at approximately 3200-3600 N for adults is within the same order of magnitude as the forces measured above. Albeit not directly measured in most of the studies above, Nij sets flexion and extension moment limits of approximately 210-410 Nm and 60-125 Nm for adults. Per the Nij theory, the addition of flexion and extension moments in the above studies would tend to lower the required compressive force for injury, which does appear to trend below the compressive envelope limit in most studies,,, tending to confirm this theory.

Prevention

Diving injuries are very preventable. In contrast to other sport trauma injuries, diving injuries occur due to distinct decisions and actions which are not intrinsically part of the activity itself (such as with football). These injuries can be prevented by a combination of design elements, diver instruction and technique, education, and regulation.

Prevention by Design

Changes in pool depth design and regulation can place engineering designs that prevent the physical conditions necessary to produce serious spinal injuries. Albrand and Walter recommend re-evaluating safe diving depths requirements.

The body of water bottom surface properties might also be modified to prevent diving spinal injuries in non-natural bodies of water. Padding of surfaces may actually increase forces experienced. Nightingale et al. hypothesizes that "pocketing of the head" in impacts is more conducive to resulting in a cervical spine injury. For rigid impact surfaces, the specimens were able to move out of the way and reduce the probability of injury, whereas, for padded impact surfaces, the neck is strained for a longer time, thus allowing for the inertia of the body to increase the compressive axial load on the neck. Due to this information, it is suggested that materials without a damping characteristic should be used for surfaces in which head impacts may occur. Allowing the head to continue translating reduces the amount of momentum imparted on the cervical spine, so surface with low coefficients of friction or rigid surfaces are preferable. Additionally, many residential pools have a steep upslope from the deep end to shallow end called the “spinal wall”, which accounts for the majority of deep-water injuries . In order to prevent this mode of injury, it is recommended that lifelines are used to delineate this point in all pools or as an alternative solution, designers could seek to eliminate “spinal walls” .

Prevention by Technique

Teaching proper diving technique to all potential divers would be beneficial in reducing risks on injury. The findings of McElhaney et al. regarding the "snap roll" motion and dive depths achieved per technique recommend keeping the head and hands up and the back arched in shallow water diving. This way the dive depth is not so great and the diver would impact their hands on the water bottom before their head. In addition to using this technique, divers should be instructed to steer-up when diving as this limits the depth of the dive and impact would not occur on the crown of the head, thus reducing the occurrence of severe cervical spinal injuries . Instead of the full momentum of the body impacting the surface of the pool bottom, some of the momentum will be horizontal thus reducing the peak impact force and reducing the probability of injury.

Prevention by Education

Both children learning to swim, and pool owners, ought to be taught about diving safety. DeVivo et al. suggest that local schools might teach children about the risks of diving at a young age. Frankel et al. agree with this point, suggesting that young children should learn about the risks of diving since the age they learn how to dive. Furthermore, pool owners should should learn about proper ways in keeping their pools from being trespassed at improper hours, providing adequate safety features when the pool is being used, and having clear depth indicators placed noticeably around their pool .

Prevention by Regulation

Administration controls could also be establish to curb the frequency of diving in unsafe conditions. Adding lifeguards, imposing bylaws or pool rules on swimming and diving in certain regions or in an inebriated state would likely curb injury frequency. Improving awareness of the severity of spinal injuries in diving and warning the public of the dangers of headfirst diving in waters of unknown depth could work to reduce the rates of crippling cervical spine injuries. Albrand and Walter suggest a public safety program on the hazards of diving. They suggest warning of the dangers of diving into natural bodies of water with varying depth and to prohibit diving into above ground pools.

Current Limitations and Future Research

Cervical spine injuries due to head first diving in shallow water are moderately well researched if the expanse of literature considered includes non-diving specific cervical spine injury research. The research thus far has been very valuable, however, as with most biomechanical study, it is difficult to determine any definite thresholds given the high situational variability and variability between subjects. This is the reason for most of the controversy in the literature, which primarily revolves around the disagreement between the force values returned in research findings. The literature also counts some important limitations, which ought to be addressed in any further work.

As mentioned in biomechanical study, the study of diving injuries based on existing research must be divided in two parts: a kinematics segment with human volunteers and a kinetics segment with other human surrogates. This segmentation of the problem could introduce uncertainties into the derived injury results as the kinematic initial conditions may not be properly translated into the load and injury testing. Perhaps future research could bridge this gap with a means of sustaining diver form and speeds in a cadaver and having the same cadaver experience headfirst impact. A sort of launching mechanism and cadaver reinforcement might achieve this but no such testing has been performed and would likely be quite demanding. A more feasible method of achieving better translation between segments would simply to have the same research group perform both the kinematics study and the kinetics study, as previous studies have only focused on one segment.

Nightingale et al. also observed how most spinal cord studies fail to draw a distinction between different populations' results. They found the failure load for female spines were approximately half the magnitude as that of male spines. They also speculated on the difference between sustainable forces in young males, the primary victims of diving cervical spine injury , and the older populations tested in cadaver testing. Accounting for both these dimensions in in quantifying the ability of spines to sustain loads is important as not to generalize results and reflect these generalizations in applications of results. Future research should attempt to draw these distinctions, which can be difficult given biomechanical testing sample sizes tend to be smaller given the difficulty on acquiring many cadavers but hopefully prior art values can be used in addition to account for low sample sizes.

Using prior art to quantify injury tolerances of a larger range in population also requires a certain degree of standardization in testing methods. Although some of the above mentioned tests have reason for particular setups, many could've further harmonized their experiment setup with others so that the data might be better compared.

Overall, the existing body of literature on this subject is quite expansive and has returned valuable results, but there has been a lack of all-encompassing testing, breadth of population study, and standardization. Further testing should seek to remedy these insufficiencies to better prevent these life-altering injuries.