- Email: [email protected]
The Epidemiology of Caving Injuries in the United States
WILDERNESS & ENVIRONMENTAL MEDICINE, 23, 215–222 (2012)
ORIGINAL RESEARCH
The Epidemiology of Caving Injuries in the United States Alejandro C. Stella-Watts, MD; Christopher P. Holstege, MD; Jae K. Lee, PhD; Nathan P. Charlton, MD From the Department of Emergence Medicine, University of Rochester, Rochester, NY (Dr Stella-Watts); and the Departments of Emergency Medicine (Drs Holstege and Charlton) and Public Health Science (Dr Lee), University of Virginia, Charlottesville, VA.
Objective.—Caving is a demanding sport practiced throughout the world. Currently, there are no collective data analyzing injury mechanism or type in these austere environments. This study is a retrospective analysis of caving incidents documented by the National Speleological Society (NSS)— American Caving Accidents (ACA) annual publication. Methods.—This study retrospectively analyzes 877 incident reports collected between 1980 and 2008 by NSS-ACA. For each victim, the month, year, location, age, gender, incident type, injury zone of the body, injury type, the result of the incident, and time intervals for rescue were extracted. Results.—A total of 1356 victims were identified; 83% of victims were male, 17% were female. Ages ranged from 2 to 69 years old, with an average of 27 years. The greatest number of events occurred in summer months, peaking in July. The most common incident leading to traumatic injury was a caver fall (74%), also contributing to 30% of caver fatalities. Lower extremities were most commonly injured (29%), followed by the upper extremities and head (21% and 15%, respectively). Fractures comprised 41% of injuries, followed by lacerations (13%), bruise, hematoma, and abrasions (12%), and sprains and strains (7%). Conclusions.—The majority of injuries were not life threatening; however, over the course of 28 years there were 81 documented fatalities. Similar to other studies of wilderness injuries, fractures, soft tissue injuries, and lacerations were prominent in this study. In general, the overall precipitating event leading to injuries is falling, leading to orthopedic trauma. To better prepare cave rescue teams we have attempted to describe the characteristics of caving injuries in the United States. Key words: caving, cave, injury, rescue, epidemiology
Introduction Caving is a demanding activity practiced throughout the United States and the world.1,2 In the United States alone there are tens of thousands of known caves, and while the exact numbers of caving visits are unknown, it is estimated that at least 2 million people in the United States visit caves each year.3 Although the majority of these visits would be considered low risk (eg, guided caving tours), the National Speleological Society has more than 10 000 members who are more likely involved in rigorous caving excursions.2 Many states have specific data pertaining to existing caves. For example, there are 4378 documented caves in Virginia, with only 1348 that have Disclaimer: The authors have no conflicts of interest to declare. No offprints are available from the authors. Corresponding author: Nathan P. Charlton, MD, University of Virginia, PO Box 800774, Charlottesville, VA 22903 (e-mail: [email protected]).
been mapped.4 Of these, there are a total of 498 miles of surveyed passage, with the longest being 22.5 miles, and the deepest cave is 1263 feet.4 Underground exploration has evolved into complex excursions, going deeper into unexplored terrain. Some of the most challenging projects involve weeks to months in complete darkness thousands of meters underground. This environment is dark and wet, requiring climbing and ropework in muddy and slippery conditions, and at times even underwater submersion.5,6 Because of the convoluted terrain, electronic communication is limited. Although the caving community considers safety a priority, there is an inherent level of risk involved.2 Rockfall, cold temperatures, and contaminated air are just a few of the potential hazards.5,6 During the past decade, there has been increasing research into appropriate rescue equipment and extrication techniques within cave environments.3,6 Although data are collected on
216
Stella-Watts et al
Table 1. Demographics of caving victims Demographics Gender Male Female Unknown Age (years) 0–9 10–19 20–29 30–39 40–49 50–59 60–69 ⬎70 Unknown Total victims
Methods Number (%)
967 (71.3) 203 (15.0) 186 (13.7) 8 (0.06) 258 (19.0) 228 (16.8) 109 (8.0) 73 (5.4) 25 (1.8) 6 (0.4) 0 (0) 649 (47.9) 1356
yearly caving accidents, there are no collective data available analyzing injury mechanism or type to help establish appropriate medical protocols in these austere environments.3,5,7 The purpose of this study is to describe the demographics, mechanisms, and characteristics of caving injuries occurring within the United States.
This study is a retrospective analysis of caving incidents in the United States from 1980 to 2008 collected and documented by the National Speleological Society (NSS)—American Caving Accidents (ACA) annual publication.8 –22 These incident reports are voluntarily provided to the NSS by groups such as local caving grottos and rescue personnel, such as professional cave rescuers, emergency medical service (EMS) providers, and police. The NSS requests that reports be submitted in their standard format available on the NSS website.2 This form mandates date and place of incident, the mechanism of incident (using predefined terminology), and result of incident (broadly defined as fatality, injury, or no consequence). Other details are optional including age, gender, and injury type.2 These incident reports are compiled and published in an annual NSS-ACA publication. In this study a single author reviewed all incident reports listed in each yearly publication. Data were extracted using predefined variables for type of injury and body zones based on commonly found wilderness injury patterns.23 The mechanism of the incident was classified based on the predefined terminology used on the NSS incident report form. For each victim, the month, year, location, age, gender, incident type, injury zone of the
Figure 1. Monthly distribution of caving incidents.
US Caving Injuries
217
body, injury type, the result of the incident, and time intervals from rescue call to victim contact and from victim contact to cave exit were recorded. Victims were included if they voluntarily entered a cave, an incident occurred for which a report was generated, and the incident occurred within the United States. Incidents were excluded if they did not occur within the United States or were the result of a victim entering a cave unintentionally. It is important to note that caving incidents are complex, and often involve multiple factors and multiple injuries. For each victim all injuries were recorded individually; therefore, multiple body zones injured and injury types were often included for the same caver. In addition, multiple factors sometimes led to the incident, so multiple mechanisms of incident could be recorded for the same incident. Traumatic injuries were separated into 10 injury zone categories. For each injury zone, the injury was classified into 11
Table 2. Yearly distribution of incidents Year
Incidents
Cavers requiring rescue
1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 Total Average
24 23 30 31 31 11 23 43 38 40 49 43 50 47 not available not available 44 35 29 42 36 30 30 29 20 25 32 23 19 877 32
43 52 51 53 38 18 31 61 54 59 110 64 69 77 not available not available 71 52 43 67 48 40 40 51 25 30 43 36 30 1356 50
Table 3. Mechanism of incident Incident
Number (%)
Caver fall Mechanical falla Equipment failure Anchor failure
339 (24) 308 (22) 19 (1) 12 (1)
Loss of cave integrity Rockfallb Entrapmentc Buried
152 (11) 85 (6) 45 (3) 22 (2)
Unable to exit cave Stranded Lost Flooding Light failure Difficulty on rope Stuck/wedged in rock Exhaustion Equipment problem
765 (54) 185 (13) 178 (13) 117 (8) 107 (8) 81 (6) 58 (4) 33 (2) 6 (⬍1)
Cave environment Hypothermia Medical illness during cavingd Poisoning by toxic gas Drowning Animal attack Illness acquired by cavinge Burn (gas lamps)
133 (9) 45 (3) 30 (2) 23 (2) 16 (1) 10 (1) 6 (⬍1) 3 (⬍1)
Injured while rescuing Harness hang syndrome Cannot categorize Total
13 (1) 3 (⬍1) 18 (1) 1423
a
Human error; excludes equipment and anchor failure. Injury caused by falling rocks; excludes being buried or entrapped. c Trapped in a location by falling earth but no injury occurred. d Medical condition such as seizure or heart attack occurring while caving. e Illness such as infectious disease related to the cave environment. b
injury types. The rescue classification was identified for each victim to determine whether an incident resulted in needing rescue. Injury and medical outcomes, including the need for transport to medical care, were often unavailable. Results From 1980 to 2008 a total of 877 incident reports were identified with a total of 1356 individual cavers involved in these incidents. Results are divided into month of the year, demographics, incident type, injury zone of the body, injury type, rescue classification, and time inter-
218
Stella-Watts et al
Table 4. Mechanism of medical problem Medical problem
Number (%)
Traumatic injury Animal attack Burn (gas lamps) Injured while rescuing New-onset medical conditions Exhaustion Hypothermia Medical illness during cavinga Poisoning by toxic gas Harness hang syndrome Drowning Illness acquired by cavingb Total
446 (72) 10 (2) 3 (⬍1) 13 (2) 33 (5) 45 (7) 30 (5) 23 (4) 3 (⬍1) 16 (3) 6 (⬍1) 628
victims and young adults (age ⬍20 years) constituted a large percentage of the victims (37%). Yearly trends revealed an increased number of incidents during summer months, with a peak in July (Figure 1). The average number of caving victims was 50 per year, with an average of 32 caving incidents per year (Table 2).
MECHANISM OF INCIDENT
a Medical condition such as seizure or heart attack occurring while caving. b Illness such as infectious disease related to the cave environment.
vals from rescue call to victim contact and from victim contact to cave exit. DEMOGRAPHICS AND OVERALL TRENDS Gender was documented for 1170 (86%) victims; age was documented in 707 victims. Results are listed in Table 1. Age ranged from 2 to 69 years old. Pediatric
From the 1356 incident victims, there were 1423 identified inciting factors identified that led to the incident— these were termed mechanism of incident and are listed in Table 3. Some of these incidents led to a caver not being injured but requiring rescue, whereas some led to illness such as hypothermia or traumatic injury. The most common incident leading to a rescue effort was being unable to exit cave, such as being stranded or lost, (54%), which rarely led to medical illness or injury. The second most common mechanism of incident was caver fall, which led to 24% of incidents, and was responsible for 74% of traumatic injuries. Only 628 incidents resulted in a medical problem, listed in Table 4. Illness during caving, such as heart attack or seizure, was an uncommon reason for needing rescue, only occurring in 2% of total incidents (Table 3) and constituting 5% of all medical problems (Table 4).
Table 5. Injury type by injury zone Injury type Injury zone
Fracture, number (%)
Dislocation, number (%)
Sprain/strain, number (%)
Laceration, number (%)
Amputation, number (%)
Bruise/hematoma, number (%)
Head Spine/back Upper extremitya Shoulder to wrist Hand Lower extremityb Hip to lower leg Foot/ankle Thorax Abdomen Pelvis/genitalia Otherc Total Percentaged
20 (19) 31 (48) 51 (34) 41 (36) 10 (28) 101 (48) 56 (42) 45 (58) 26 (53) 0 (0) 13 (56) 2 (2) 244 41
0 (0) 0 (0) 32 (21) 31 (27) 1 (3) 6 (3) 5 (4) 1 (1) 0 (0) 0 (0) 0 (0) 0 (0) 38 6
0 (0) 3 (5) 5 (3) 5 (4) 0 (0) 30 (14) 11 (8) 19 (25) 1 (2) 0 (0) 1 (4) 0 (0) 40 7
42 (39) 2 (3) 19 (13) 10 (9) 9 (25) 11 (5) 10 (8) 1 (1) 1 (2) 0 (0) 1 (4) 0 (0) 76 13
0 (0) 0 (0) 1 (⬍1) 0 (0) 1 (3) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 1 ⬍1
6 (6) 9 (14) 16 (11) 14 (12) 2 (6) 24 (11) 21 (16) 3 (4) 8 (16) 1 (11) 3 (13) 4 (4) 71 12
a
Upper extremity is further stratified into 2 categories: 1) shoulder to wrist, 2) hand. Lower extremity is further stratified into 2 categories: 1) hip to lower leg, 2) foot and ankle. c Injury zone— other refers to when a caver is reported to suffer an injury such as a fracture or bruise without indicating a body zone. d Diagnosis percentage is calculated excluding injuries that did not report the diagnosis. b
US Caving Injuries
219
INJURY ZONE AND INJURY TYPE
Table 6. Rescue classification
Of the cavers needing rescue, a total of 459 incidents resulting in 720 distinct traumatic injuries were identified and categorized into injury zones (Table 5). The most common injury zone was the lower extremities, followed by the upper extremities and head. Injury type was also recorded for each injury zone. From the data, classification of diagnosis was made for only 588 of the injuries (Table 5). The 132 injury zones for which the distinct injury was not reported were not included in this calculation. Spine or back was the least accurately reported zone, with 16 victims (25%) not having a distinct injury listed. Of the injuries reported, fractures were the most common type of injury, making up 41% of all injuries (Table 5). The head deserves specific mention as it is the only injury zone that regularly receives protection from a helmet. Skull and facial fractures are less common, contributing to 19% of head injuries, whereas lacerations constitute 39% of head injuries.
Rescue classification
Number (%)
Fatality Injury and rescuea Rescue, no injury Injury, no rescue No injury, no rescueb Total
81 (6) 351 (26) 764 (56) 92 (7) 68 (5) 1356
a
Rescue refers to a caver requiring rescue assistance from EMS or cave rescuers. b No injury, no aid refers to scenarios where an incident occurred and a caver decided to escape a cave only to realize there were no injuries sustained.
minor sprains or abrasions. Incidents in which no injury or no rescue was recorded refers to when a potentially dangerous incident occurred, such as flooding or cave collapse, where no one was injured or required external rescue.
RESCUE CLASSIFICATION
RESCUE TIME INTERVALS
All incident victims were assigned to 5 categories to classify rescue requirement (Table 6). Including fatalities, 88% of incidents required rescue from either other climbers or rescue services. Some cavers were injured but were able to exit without aid, such as those with
Although the NSS often did not have access to the time intervals of rescues, some data were obtained. Only 256 (19%) of incidents reported the time interval between when a rescue call was made (to either EMS or the National Cave Rescue Commission) to when rescuers
Table 5. Continued Injury type Internal organ Burn, Minor injury not Diagnosis not Total number (including Total percent for damage, number (%) number (%) specified, number (%) reported, number (%) diagnosis not reported) injury zone 5 (5) 0 (0) 0 (0) 0 (0) 0 (0) 1 (⬍1) 1 (⬍1) 0 (0) 8 (16) 7 (77) 0 (0) 3 (3) 24 4
3 (3) 0 (0) 8 (5) 1 (⬍1) 7 (19) 2 (1) 2 (2) 0 (0) 0 (0) 0 (0) 0 (0) 1 (1) 14 2
9 (8) 3 (5) 4 (3) 3 (3) 1 (3) 12 (5) 8 (6) 4 (6) 3 (6) 0 (0) 3 (13) 45 (42) 80 14
22 (20) 16 (25) 14 (9) 9 (8) 5 (14) 22 (10) 18 (13) 4 (5) 2 (4) 1 (11) 2 (9) 53 (49) 132
108 64 150 114 36 209 132 77 49 9 23 108 720
15% 9% 21% 16% 5% 29% 18% 11% 7% 1% 3% 15% 100%
220
Stella-Watts et al
Table 7. Fatalities Variable Gender Male Female Mechanism Caver fall Drowning Jammed Cardiac disease Hypothermia Other Total
Number (%)
68 (84) 13 (16) 24 (30) 24 (30) 15 (18) 10 (12) 5 (6) 3 (3) 81
established contact with the victim. Of those, 113 (44%) took 0 to 2 hours for rescuers to reach the victim, 143 (56%) took longer than 2 hours, and 13% required longer than 24 hours. Individual cavers or caving groups who were able to rescue themselves were not counted, as the rescuers were already present at the time of the incident. However, if a victim required aid from cavers of another group, time was recorded if available. The rescue duration, defined as the time between when rescuers established contact with the victim to when the victim reached the cave entrance, was documented in 444 (33%) of reported incidents. Of those reports, 190 (43%) required 0 to 2 hours for the victim to exit the cave, whereas 244 (57%) required longer than 2 hours, and 160 (36%) required longer than 4 hours. An important time interval we were not able to extrapolate was the time from when the incident occurred to when a rescue call was made—an important variable that can in some instances take hours or days, particularly if the caver did not leave a trip plan with someone on the surface. FATALITIES Fatality occurred in 81 (6%) individuals during the study period (Table 7). The incidence of fatality was 3 deaths per year. Water-related incidents (eg, drowning and flooding) and caver falls tied for the highest percentage of deaths. Medical illness (such as cardiac disease and hypothermia) resulted in a minority of deaths compared with accident-related deaths. Discussion Before this study, there had been no extensive analysis of injuries involved in underground recreation and exploration. The leading experts in speleology wrote much of
the previously published work on the medical aspects of caving incidents; however, most recommendations were anecdotal. Cave rescue is often a complex and timeconsuming endeavor. The time intervals between when an incident occurs, a rescue call is made, and when rescue teams establish contact with a victim can take hours and even days, requiring numerous rescuers. Not only do these rescuers need to prepare for injuries they will encounter and anticipate further complications throughout a rescue operation, rescuers must also protect themselves from injuries. Understanding the etiology of underground injuries and outcomes is of utmost importance when executing a cave rescue operation. In this study the incidence of caving incidents was reported as 32 per year, accounting for an average of 50 victims per year. This likely underestimates the true number of incidents as this number takes into account only those reported to the NSS. We would speculate that incidents resulting in no injury or only minor injuries and those requiring no rescue or rescue only from other cavers would be those that are most likely underreported. It was found that an increased number of incidents occurred during summer months, with a peak in July. This is likely related to the increased incidence of caving excursions during the summer months. Most injuries were not life threatening. However, some fatalities did occur. Although the focus of this study was not primarily on fatalities, caver fall, drowning, becoming physically jammed, and cardiac disease were common causes of fatal outcomes. With the exception of becoming jammed, these etiologies are consistent with other studies on the epidemiology of mortality in the wilderness.24,25 The total number of caving excursions and total caving participants per year are unknown. Because of this, the rate of total caving incidents per year based on those reported to the NSS is likely an underestimate. However, because caving fatalities are often better publicized and better documented, our calculated rate of 3 deaths per year may be more accurate. Similar to other epidemiologic studies of wilderness injuries, fractures, soft tissue injuries, and lacerations were prevalent in this study.23 Fractures were the most common injury requiring rescue and were the most common injury for almost all injury zones, contributing to 48% of spine and back injuries, 48% of lower extremity injuries, and 56% of pelvis and genitalia injuries. Lower extremity, spine, and pelvis injuries are medically important as these victims are less likely to be able to perform self-rescue and may require significant rescue resources. Upper extremity injuries also constituted a large number of injuries, and although in some situations these victims may be more likely to self-rescue, difficult terrain will pose a challenge to any victim with a signif-
US Caving Injuries icant injury. Although generally not immediately life threatening, these fractures can greatly impede rescue and increase victim exposure to the environment. EMS and rescue providers should be prepared with adequate knowledge and supplies to splint and bandage the injured victim and should be prepared for a completely dependent victim requiring extrication. In addition, rescuers should be able to perform spinal immobilization as head and spine injury were common in this study. Cavers can become trapped in the cave by injury, illness, and many types of mechanical incidents. Of the incidents reported, the majority of victims required some form of aid to leave the cave. Because of the arduous nature of caving, this is expected as any impairment in performance or gear may lead to an insurmountable obstacle. Any incident resulting in a caver being unable to exit the cave will result in increased exposure to the environment. With many caves harboring a cold and wet environment, rescuers should be knowledgeable of the signs and symptoms of hypothermia, and be prepared to remove wet clothes and provide adequate warmth during the rescue operation. Burial occurred in 22 victims during this study. Rockfall and burial poses the potential problem of crush injury, which, given the duration of most rescues, could lead to rhabdomyolysis, crush syndrome, and rapid cardiovascular collapse once the victim is freed. If available, rescuers should consider early administration of intravenous fluids if crush syndrome or rhabdomyolysis is of concern.26,27 Harness hang syndrome or suspension trauma (cardiovascular shock induced by passively hanging in a harness) could potentially result in similar pathology; early intravenous fluids have recently been advocated for this syndrome as well.28 In this study suspension trauma was rare with only 3 reported cases. LIMITATIONS There are a number of limitations that are present with this analysis. Incident reports were often generated from nonmedical personnel (eg, volunteer rescue groups, police, and fire fighters) and, therefore, the injury reported in the incident report may not accurately reflect the final medical diagnosis. As the reports are voluntary they likely underreport the true incidence of caving accidents and injuries. Gaps, including demographics and injury descriptions, often existed in the reported data. In addition, we were not able to obtain data for the years 1994 and 1995. The NSS-ACA report form requires that mechanism of incident be recorded to submit an incident report; however, the mechanism of incident does not necessarily identify the mechanism of the injury. Therefore, the
221 mechanism of the injury could at times be misleading. For example, equipment failure while on a rope can cause a caver to fall and fracture a lower extremity; the mechanism of incident for the fractured leg would then be equipment failure, although the mechanism leading to the injury was falling. As only 1 author recorded the data, his interpretation of reported mechanism and injury pattern in some instances may have resulted in misclassification of the data. This report does not include data from outside the United States; therefore, injuries incurred by use of equipment or techniques more common outside of the United States may be missed. Additionally, large caving expeditions are more common outside the United States; therefore, these data may not include injuries more common on large expeditions. In this study it was not possible to identify the level of caver experience; therefore, it was not possible to separate novice caver injuries from those of experienced cavers. Novice cavers may have different mechanisms of injury and injury patterns from those of experienced cavers or those on caving expeditions. Conclusions Underground exploration is a difficult undertaking requiring experience, resources, and a high level of autonomy. When injury occurs in these austere environments, rescue is difficult, requiring lengthy rescue times and extensive resources. Cavers themselves are the key to proper rescue. They must be well prepared, and make smart decisions to orchestrate and initiate rescue efforts. The NSS’s extensive and well-organized network of grottos throughout every region has allowed them to maintain documentation of underground injuries in the United States for 3 decades. It is our intention that with this initial evaluation and with continued reevaluation, data collection and medical knowledge of caving injuries can be enhanced to provide greater benefit to rescuers and cavers in need. Acknowledgements The authors are grateful for the help and expertise of Ray Keeler, Editor of American Caving Accidents, and the National Speleological Society. References 1. International Union of Speleology. National and World Region Speleological Organisations. Available at: http:// www.uis-speleo.org/people/natspel.html. Accessed December 23, 2011.
222 2. National Speleological Society. Available at: http:// www.caves.org/. Accessed September 1, 2011. 3. Hooker K, Shalit M. Subterranean medicine: an inquiry into underground medical treatment protocols in cave rescue situations in national parks in the United States. Wilderness Environ Med. 2000;11:17–20. 4. The Virginia Speleological Survey. Available at http:// www.virginiacaves.org. Accessed September 1, 2011. 5. Hempel JC, Fregeau-Conover A, eds.On Call: A Complete Reference for Cave Rescue. 2001 edition. Huntsville, AL: The National Speleological Society Inc; 2001. 6. Hudson SE, McCurley LH. Caving and cave rescue. In: Auerbach PS, ed. Wilderness Medicine: Management of Wilderness and Environmental Emergencies. 5th ed. St Louis, MO: Mosby Elsevier; 2007. 7. Ashford DA, Knutson RS, Sacks JJ. Injury among cavers: results of a preliminary national survey. J Sports Med Phys Fitness. 1999;39:71–73. 8. Knutson S. American caving accidents 1980 –1981. Natl Speleol Soc News. 1983;41:99 –118. 9. Knutson S. American caving accidents 1982. Natl Speleol Soc News. 1983;41:255–265. 10. Knutson S. American caving accidents 1983. Natl Speleol Soc News. 1985;42:347–357. 11. Knutson S. American caving accidents 1984. Natl Speleol Soc News. 1985;43:355–376. 12. Knutson S. American caving accidents 1985. Natl Speleol Soc News. 1986 44:399 – 409. 13. Knutson S. American caving accidents 1986. Natl Speleol Soc News. 1987;45:386 –398. 14. Knutson S. American caving accidents 1987. Natl Speleol Soc News. 1988;46:475– 490.
Stella-Watts et al 15. Knutson S. American caving accidents 1988. Natl Speleol Soc News. 1989;47:314 –334. 16. Knutson S. American caving accidents 1989. Natl Speleol Soc News. 1990;48:331–349. 17. Knutson S. American caving accidents 1990. Natl Speleol Soc News. 1990;48:330 –357. 18. Knutson S. American caving accidents 1992. Natl Speleol Soc News. 1993;51:346 –370. 19. Putnam WO. American caving accidents 1996 –1998. Natl Speleol Soc News. 2000;58:1–57. 20. Putnam WO. American caving accidents 2002–2003. Natl Speleol Soc News. 2005;63:1– 48. 21. Putnam WO. American caving accidents 2004 –2005. Natl Speleol Soc News. 2007;65:1– 45. 22. Keeler R. American caving accidents 2007–2008. Natl Speleol Soc News. 2010;68:1– 44. 23. Forrester JD, Holstege CP. Injury and illness encountered in Shenandoah National Park. Wilderness Environ Med. 2009;20:318 –326. 24. Goodman T, Iserson KV, Strich H. Wilderness mortalities: a 13-year experience. Ann Emerg Med. 2001;37:279 –283. 25. Montalvo R, Wingard DL, Bracker M, Davidson TM. Morbidity and mortality in the wilderness. West J Med. 1998;168:248 –254. 26. Better OS, Abassi ZA. Early fluid resuscitation in patients with rhabdomyolysis. Nat Rev Nephrol. 2011;7: 416 – 422. 27. Haojun F, Jianqi S, Shike H. Retrospective, analytical study of field first aid following the Wenchean earthquake in China. Prehosp Disaster Med. 2011;26:130 –134. 28. Mortimer RB. Risks and management of prolonged suspension in an alpine harness. Wilderness Environ Med. 2011;22:77– 86.
ORIGINAL RESEARCH
The Epidemiology of Caving Injuries in the United States Alejandro C. Stella-Watts, MD; Christopher P. Holstege, MD; Jae K. Lee, PhD; Nathan P. Charlton, MD From the Department of Emergence Medicine, University of Rochester, Rochester, NY (Dr Stella-Watts); and the Departments of Emergency Medicine (Drs Holstege and Charlton) and Public Health Science (Dr Lee), University of Virginia, Charlottesville, VA.
Objective.—Caving is a demanding sport practiced throughout the world. Currently, there are no collective data analyzing injury mechanism or type in these austere environments. This study is a retrospective analysis of caving incidents documented by the National Speleological Society (NSS)— American Caving Accidents (ACA) annual publication. Methods.—This study retrospectively analyzes 877 incident reports collected between 1980 and 2008 by NSS-ACA. For each victim, the month, year, location, age, gender, incident type, injury zone of the body, injury type, the result of the incident, and time intervals for rescue were extracted. Results.—A total of 1356 victims were identified; 83% of victims were male, 17% were female. Ages ranged from 2 to 69 years old, with an average of 27 years. The greatest number of events occurred in summer months, peaking in July. The most common incident leading to traumatic injury was a caver fall (74%), also contributing to 30% of caver fatalities. Lower extremities were most commonly injured (29%), followed by the upper extremities and head (21% and 15%, respectively). Fractures comprised 41% of injuries, followed by lacerations (13%), bruise, hematoma, and abrasions (12%), and sprains and strains (7%). Conclusions.—The majority of injuries were not life threatening; however, over the course of 28 years there were 81 documented fatalities. Similar to other studies of wilderness injuries, fractures, soft tissue injuries, and lacerations were prominent in this study. In general, the overall precipitating event leading to injuries is falling, leading to orthopedic trauma. To better prepare cave rescue teams we have attempted to describe the characteristics of caving injuries in the United States. Key words: caving, cave, injury, rescue, epidemiology
Introduction Caving is a demanding activity practiced throughout the United States and the world.1,2 In the United States alone there are tens of thousands of known caves, and while the exact numbers of caving visits are unknown, it is estimated that at least 2 million people in the United States visit caves each year.3 Although the majority of these visits would be considered low risk (eg, guided caving tours), the National Speleological Society has more than 10 000 members who are more likely involved in rigorous caving excursions.2 Many states have specific data pertaining to existing caves. For example, there are 4378 documented caves in Virginia, with only 1348 that have Disclaimer: The authors have no conflicts of interest to declare. No offprints are available from the authors. Corresponding author: Nathan P. Charlton, MD, University of Virginia, PO Box 800774, Charlottesville, VA 22903 (e-mail: [email protected]).
been mapped.4 Of these, there are a total of 498 miles of surveyed passage, with the longest being 22.5 miles, and the deepest cave is 1263 feet.4 Underground exploration has evolved into complex excursions, going deeper into unexplored terrain. Some of the most challenging projects involve weeks to months in complete darkness thousands of meters underground. This environment is dark and wet, requiring climbing and ropework in muddy and slippery conditions, and at times even underwater submersion.5,6 Because of the convoluted terrain, electronic communication is limited. Although the caving community considers safety a priority, there is an inherent level of risk involved.2 Rockfall, cold temperatures, and contaminated air are just a few of the potential hazards.5,6 During the past decade, there has been increasing research into appropriate rescue equipment and extrication techniques within cave environments.3,6 Although data are collected on
216
Stella-Watts et al
Table 1. Demographics of caving victims Demographics Gender Male Female Unknown Age (years) 0–9 10–19 20–29 30–39 40–49 50–59 60–69 ⬎70 Unknown Total victims
Methods Number (%)
967 (71.3) 203 (15.0) 186 (13.7) 8 (0.06) 258 (19.0) 228 (16.8) 109 (8.0) 73 (5.4) 25 (1.8) 6 (0.4) 0 (0) 649 (47.9) 1356
yearly caving accidents, there are no collective data available analyzing injury mechanism or type to help establish appropriate medical protocols in these austere environments.3,5,7 The purpose of this study is to describe the demographics, mechanisms, and characteristics of caving injuries occurring within the United States.
This study is a retrospective analysis of caving incidents in the United States from 1980 to 2008 collected and documented by the National Speleological Society (NSS)—American Caving Accidents (ACA) annual publication.8 –22 These incident reports are voluntarily provided to the NSS by groups such as local caving grottos and rescue personnel, such as professional cave rescuers, emergency medical service (EMS) providers, and police. The NSS requests that reports be submitted in their standard format available on the NSS website.2 This form mandates date and place of incident, the mechanism of incident (using predefined terminology), and result of incident (broadly defined as fatality, injury, or no consequence). Other details are optional including age, gender, and injury type.2 These incident reports are compiled and published in an annual NSS-ACA publication. In this study a single author reviewed all incident reports listed in each yearly publication. Data were extracted using predefined variables for type of injury and body zones based on commonly found wilderness injury patterns.23 The mechanism of the incident was classified based on the predefined terminology used on the NSS incident report form. For each victim, the month, year, location, age, gender, incident type, injury zone of the
Figure 1. Monthly distribution of caving incidents.
US Caving Injuries
217
body, injury type, the result of the incident, and time intervals from rescue call to victim contact and from victim contact to cave exit were recorded. Victims were included if they voluntarily entered a cave, an incident occurred for which a report was generated, and the incident occurred within the United States. Incidents were excluded if they did not occur within the United States or were the result of a victim entering a cave unintentionally. It is important to note that caving incidents are complex, and often involve multiple factors and multiple injuries. For each victim all injuries were recorded individually; therefore, multiple body zones injured and injury types were often included for the same caver. In addition, multiple factors sometimes led to the incident, so multiple mechanisms of incident could be recorded for the same incident. Traumatic injuries were separated into 10 injury zone categories. For each injury zone, the injury was classified into 11
Table 2. Yearly distribution of incidents Year
Incidents
Cavers requiring rescue
1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 Total Average
24 23 30 31 31 11 23 43 38 40 49 43 50 47 not available not available 44 35 29 42 36 30 30 29 20 25 32 23 19 877 32
43 52 51 53 38 18 31 61 54 59 110 64 69 77 not available not available 71 52 43 67 48 40 40 51 25 30 43 36 30 1356 50
Table 3. Mechanism of incident Incident
Number (%)
Caver fall Mechanical falla Equipment failure Anchor failure
339 (24) 308 (22) 19 (1) 12 (1)
Loss of cave integrity Rockfallb Entrapmentc Buried
152 (11) 85 (6) 45 (3) 22 (2)
Unable to exit cave Stranded Lost Flooding Light failure Difficulty on rope Stuck/wedged in rock Exhaustion Equipment problem
765 (54) 185 (13) 178 (13) 117 (8) 107 (8) 81 (6) 58 (4) 33 (2) 6 (⬍1)
Cave environment Hypothermia Medical illness during cavingd Poisoning by toxic gas Drowning Animal attack Illness acquired by cavinge Burn (gas lamps)
133 (9) 45 (3) 30 (2) 23 (2) 16 (1) 10 (1) 6 (⬍1) 3 (⬍1)
Injured while rescuing Harness hang syndrome Cannot categorize Total
13 (1) 3 (⬍1) 18 (1) 1423
a
Human error; excludes equipment and anchor failure. Injury caused by falling rocks; excludes being buried or entrapped. c Trapped in a location by falling earth but no injury occurred. d Medical condition such as seizure or heart attack occurring while caving. e Illness such as infectious disease related to the cave environment. b
injury types. The rescue classification was identified for each victim to determine whether an incident resulted in needing rescue. Injury and medical outcomes, including the need for transport to medical care, were often unavailable. Results From 1980 to 2008 a total of 877 incident reports were identified with a total of 1356 individual cavers involved in these incidents. Results are divided into month of the year, demographics, incident type, injury zone of the body, injury type, rescue classification, and time inter-
218
Stella-Watts et al
Table 4. Mechanism of medical problem Medical problem
Number (%)
Traumatic injury Animal attack Burn (gas lamps) Injured while rescuing New-onset medical conditions Exhaustion Hypothermia Medical illness during cavinga Poisoning by toxic gas Harness hang syndrome Drowning Illness acquired by cavingb Total
446 (72) 10 (2) 3 (⬍1) 13 (2) 33 (5) 45 (7) 30 (5) 23 (4) 3 (⬍1) 16 (3) 6 (⬍1) 628
victims and young adults (age ⬍20 years) constituted a large percentage of the victims (37%). Yearly trends revealed an increased number of incidents during summer months, with a peak in July (Figure 1). The average number of caving victims was 50 per year, with an average of 32 caving incidents per year (Table 2).
MECHANISM OF INCIDENT
a Medical condition such as seizure or heart attack occurring while caving. b Illness such as infectious disease related to the cave environment.
vals from rescue call to victim contact and from victim contact to cave exit. DEMOGRAPHICS AND OVERALL TRENDS Gender was documented for 1170 (86%) victims; age was documented in 707 victims. Results are listed in Table 1. Age ranged from 2 to 69 years old. Pediatric
From the 1356 incident victims, there were 1423 identified inciting factors identified that led to the incident— these were termed mechanism of incident and are listed in Table 3. Some of these incidents led to a caver not being injured but requiring rescue, whereas some led to illness such as hypothermia or traumatic injury. The most common incident leading to a rescue effort was being unable to exit cave, such as being stranded or lost, (54%), which rarely led to medical illness or injury. The second most common mechanism of incident was caver fall, which led to 24% of incidents, and was responsible for 74% of traumatic injuries. Only 628 incidents resulted in a medical problem, listed in Table 4. Illness during caving, such as heart attack or seizure, was an uncommon reason for needing rescue, only occurring in 2% of total incidents (Table 3) and constituting 5% of all medical problems (Table 4).
Table 5. Injury type by injury zone Injury type Injury zone
Fracture, number (%)
Dislocation, number (%)
Sprain/strain, number (%)
Laceration, number (%)
Amputation, number (%)
Bruise/hematoma, number (%)
Head Spine/back Upper extremitya Shoulder to wrist Hand Lower extremityb Hip to lower leg Foot/ankle Thorax Abdomen Pelvis/genitalia Otherc Total Percentaged
20 (19) 31 (48) 51 (34) 41 (36) 10 (28) 101 (48) 56 (42) 45 (58) 26 (53) 0 (0) 13 (56) 2 (2) 244 41
0 (0) 0 (0) 32 (21) 31 (27) 1 (3) 6 (3) 5 (4) 1 (1) 0 (0) 0 (0) 0 (0) 0 (0) 38 6
0 (0) 3 (5) 5 (3) 5 (4) 0 (0) 30 (14) 11 (8) 19 (25) 1 (2) 0 (0) 1 (4) 0 (0) 40 7
42 (39) 2 (3) 19 (13) 10 (9) 9 (25) 11 (5) 10 (8) 1 (1) 1 (2) 0 (0) 1 (4) 0 (0) 76 13
0 (0) 0 (0) 1 (⬍1) 0 (0) 1 (3) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 1 ⬍1
6 (6) 9 (14) 16 (11) 14 (12) 2 (6) 24 (11) 21 (16) 3 (4) 8 (16) 1 (11) 3 (13) 4 (4) 71 12
a
Upper extremity is further stratified into 2 categories: 1) shoulder to wrist, 2) hand. Lower extremity is further stratified into 2 categories: 1) hip to lower leg, 2) foot and ankle. c Injury zone— other refers to when a caver is reported to suffer an injury such as a fracture or bruise without indicating a body zone. d Diagnosis percentage is calculated excluding injuries that did not report the diagnosis. b
US Caving Injuries
219
INJURY ZONE AND INJURY TYPE
Table 6. Rescue classification
Of the cavers needing rescue, a total of 459 incidents resulting in 720 distinct traumatic injuries were identified and categorized into injury zones (Table 5). The most common injury zone was the lower extremities, followed by the upper extremities and head. Injury type was also recorded for each injury zone. From the data, classification of diagnosis was made for only 588 of the injuries (Table 5). The 132 injury zones for which the distinct injury was not reported were not included in this calculation. Spine or back was the least accurately reported zone, with 16 victims (25%) not having a distinct injury listed. Of the injuries reported, fractures were the most common type of injury, making up 41% of all injuries (Table 5). The head deserves specific mention as it is the only injury zone that regularly receives protection from a helmet. Skull and facial fractures are less common, contributing to 19% of head injuries, whereas lacerations constitute 39% of head injuries.
Rescue classification
Number (%)
Fatality Injury and rescuea Rescue, no injury Injury, no rescue No injury, no rescueb Total
81 (6) 351 (26) 764 (56) 92 (7) 68 (5) 1356
a
Rescue refers to a caver requiring rescue assistance from EMS or cave rescuers. b No injury, no aid refers to scenarios where an incident occurred and a caver decided to escape a cave only to realize there were no injuries sustained.
minor sprains or abrasions. Incidents in which no injury or no rescue was recorded refers to when a potentially dangerous incident occurred, such as flooding or cave collapse, where no one was injured or required external rescue.
RESCUE CLASSIFICATION
RESCUE TIME INTERVALS
All incident victims were assigned to 5 categories to classify rescue requirement (Table 6). Including fatalities, 88% of incidents required rescue from either other climbers or rescue services. Some cavers were injured but were able to exit without aid, such as those with
Although the NSS often did not have access to the time intervals of rescues, some data were obtained. Only 256 (19%) of incidents reported the time interval between when a rescue call was made (to either EMS or the National Cave Rescue Commission) to when rescuers
Table 5. Continued Injury type Internal organ Burn, Minor injury not Diagnosis not Total number (including Total percent for damage, number (%) number (%) specified, number (%) reported, number (%) diagnosis not reported) injury zone 5 (5) 0 (0) 0 (0) 0 (0) 0 (0) 1 (⬍1) 1 (⬍1) 0 (0) 8 (16) 7 (77) 0 (0) 3 (3) 24 4
3 (3) 0 (0) 8 (5) 1 (⬍1) 7 (19) 2 (1) 2 (2) 0 (0) 0 (0) 0 (0) 0 (0) 1 (1) 14 2
9 (8) 3 (5) 4 (3) 3 (3) 1 (3) 12 (5) 8 (6) 4 (6) 3 (6) 0 (0) 3 (13) 45 (42) 80 14
22 (20) 16 (25) 14 (9) 9 (8) 5 (14) 22 (10) 18 (13) 4 (5) 2 (4) 1 (11) 2 (9) 53 (49) 132
108 64 150 114 36 209 132 77 49 9 23 108 720
15% 9% 21% 16% 5% 29% 18% 11% 7% 1% 3% 15% 100%
220
Stella-Watts et al
Table 7. Fatalities Variable Gender Male Female Mechanism Caver fall Drowning Jammed Cardiac disease Hypothermia Other Total
Number (%)
68 (84) 13 (16) 24 (30) 24 (30) 15 (18) 10 (12) 5 (6) 3 (3) 81
established contact with the victim. Of those, 113 (44%) took 0 to 2 hours for rescuers to reach the victim, 143 (56%) took longer than 2 hours, and 13% required longer than 24 hours. Individual cavers or caving groups who were able to rescue themselves were not counted, as the rescuers were already present at the time of the incident. However, if a victim required aid from cavers of another group, time was recorded if available. The rescue duration, defined as the time between when rescuers established contact with the victim to when the victim reached the cave entrance, was documented in 444 (33%) of reported incidents. Of those reports, 190 (43%) required 0 to 2 hours for the victim to exit the cave, whereas 244 (57%) required longer than 2 hours, and 160 (36%) required longer than 4 hours. An important time interval we were not able to extrapolate was the time from when the incident occurred to when a rescue call was made—an important variable that can in some instances take hours or days, particularly if the caver did not leave a trip plan with someone on the surface. FATALITIES Fatality occurred in 81 (6%) individuals during the study period (Table 7). The incidence of fatality was 3 deaths per year. Water-related incidents (eg, drowning and flooding) and caver falls tied for the highest percentage of deaths. Medical illness (such as cardiac disease and hypothermia) resulted in a minority of deaths compared with accident-related deaths. Discussion Before this study, there had been no extensive analysis of injuries involved in underground recreation and exploration. The leading experts in speleology wrote much of
the previously published work on the medical aspects of caving incidents; however, most recommendations were anecdotal. Cave rescue is often a complex and timeconsuming endeavor. The time intervals between when an incident occurs, a rescue call is made, and when rescue teams establish contact with a victim can take hours and even days, requiring numerous rescuers. Not only do these rescuers need to prepare for injuries they will encounter and anticipate further complications throughout a rescue operation, rescuers must also protect themselves from injuries. Understanding the etiology of underground injuries and outcomes is of utmost importance when executing a cave rescue operation. In this study the incidence of caving incidents was reported as 32 per year, accounting for an average of 50 victims per year. This likely underestimates the true number of incidents as this number takes into account only those reported to the NSS. We would speculate that incidents resulting in no injury or only minor injuries and those requiring no rescue or rescue only from other cavers would be those that are most likely underreported. It was found that an increased number of incidents occurred during summer months, with a peak in July. This is likely related to the increased incidence of caving excursions during the summer months. Most injuries were not life threatening. However, some fatalities did occur. Although the focus of this study was not primarily on fatalities, caver fall, drowning, becoming physically jammed, and cardiac disease were common causes of fatal outcomes. With the exception of becoming jammed, these etiologies are consistent with other studies on the epidemiology of mortality in the wilderness.24,25 The total number of caving excursions and total caving participants per year are unknown. Because of this, the rate of total caving incidents per year based on those reported to the NSS is likely an underestimate. However, because caving fatalities are often better publicized and better documented, our calculated rate of 3 deaths per year may be more accurate. Similar to other epidemiologic studies of wilderness injuries, fractures, soft tissue injuries, and lacerations were prevalent in this study.23 Fractures were the most common injury requiring rescue and were the most common injury for almost all injury zones, contributing to 48% of spine and back injuries, 48% of lower extremity injuries, and 56% of pelvis and genitalia injuries. Lower extremity, spine, and pelvis injuries are medically important as these victims are less likely to be able to perform self-rescue and may require significant rescue resources. Upper extremity injuries also constituted a large number of injuries, and although in some situations these victims may be more likely to self-rescue, difficult terrain will pose a challenge to any victim with a signif-
US Caving Injuries icant injury. Although generally not immediately life threatening, these fractures can greatly impede rescue and increase victim exposure to the environment. EMS and rescue providers should be prepared with adequate knowledge and supplies to splint and bandage the injured victim and should be prepared for a completely dependent victim requiring extrication. In addition, rescuers should be able to perform spinal immobilization as head and spine injury were common in this study. Cavers can become trapped in the cave by injury, illness, and many types of mechanical incidents. Of the incidents reported, the majority of victims required some form of aid to leave the cave. Because of the arduous nature of caving, this is expected as any impairment in performance or gear may lead to an insurmountable obstacle. Any incident resulting in a caver being unable to exit the cave will result in increased exposure to the environment. With many caves harboring a cold and wet environment, rescuers should be knowledgeable of the signs and symptoms of hypothermia, and be prepared to remove wet clothes and provide adequate warmth during the rescue operation. Burial occurred in 22 victims during this study. Rockfall and burial poses the potential problem of crush injury, which, given the duration of most rescues, could lead to rhabdomyolysis, crush syndrome, and rapid cardiovascular collapse once the victim is freed. If available, rescuers should consider early administration of intravenous fluids if crush syndrome or rhabdomyolysis is of concern.26,27 Harness hang syndrome or suspension trauma (cardiovascular shock induced by passively hanging in a harness) could potentially result in similar pathology; early intravenous fluids have recently been advocated for this syndrome as well.28 In this study suspension trauma was rare with only 3 reported cases. LIMITATIONS There are a number of limitations that are present with this analysis. Incident reports were often generated from nonmedical personnel (eg, volunteer rescue groups, police, and fire fighters) and, therefore, the injury reported in the incident report may not accurately reflect the final medical diagnosis. As the reports are voluntary they likely underreport the true incidence of caving accidents and injuries. Gaps, including demographics and injury descriptions, often existed in the reported data. In addition, we were not able to obtain data for the years 1994 and 1995. The NSS-ACA report form requires that mechanism of incident be recorded to submit an incident report; however, the mechanism of incident does not necessarily identify the mechanism of the injury. Therefore, the
221 mechanism of the injury could at times be misleading. For example, equipment failure while on a rope can cause a caver to fall and fracture a lower extremity; the mechanism of incident for the fractured leg would then be equipment failure, although the mechanism leading to the injury was falling. As only 1 author recorded the data, his interpretation of reported mechanism and injury pattern in some instances may have resulted in misclassification of the data. This report does not include data from outside the United States; therefore, injuries incurred by use of equipment or techniques more common outside of the United States may be missed. Additionally, large caving expeditions are more common outside the United States; therefore, these data may not include injuries more common on large expeditions. In this study it was not possible to identify the level of caver experience; therefore, it was not possible to separate novice caver injuries from those of experienced cavers. Novice cavers may have different mechanisms of injury and injury patterns from those of experienced cavers or those on caving expeditions. Conclusions Underground exploration is a difficult undertaking requiring experience, resources, and a high level of autonomy. When injury occurs in these austere environments, rescue is difficult, requiring lengthy rescue times and extensive resources. Cavers themselves are the key to proper rescue. They must be well prepared, and make smart decisions to orchestrate and initiate rescue efforts. The NSS’s extensive and well-organized network of grottos throughout every region has allowed them to maintain documentation of underground injuries in the United States for 3 decades. It is our intention that with this initial evaluation and with continued reevaluation, data collection and medical knowledge of caving injuries can be enhanced to provide greater benefit to rescuers and cavers in need. Acknowledgements The authors are grateful for the help and expertise of Ray Keeler, Editor of American Caving Accidents, and the National Speleological Society. References 1. International Union of Speleology. National and World Region Speleological Organisations. Available at: http:// www.uis-speleo.org/people/natspel.html. Accessed December 23, 2011.
222 2. National Speleological Society. Available at: http:// www.caves.org/. Accessed September 1, 2011. 3. Hooker K, Shalit M. Subterranean medicine: an inquiry into underground medical treatment protocols in cave rescue situations in national parks in the United States. Wilderness Environ Med. 2000;11:17–20. 4. The Virginia Speleological Survey. Available at http:// www.virginiacaves.org. Accessed September 1, 2011. 5. Hempel JC, Fregeau-Conover A, eds.On Call: A Complete Reference for Cave Rescue. 2001 edition. Huntsville, AL: The National Speleological Society Inc; 2001. 6. Hudson SE, McCurley LH. Caving and cave rescue. In: Auerbach PS, ed. Wilderness Medicine: Management of Wilderness and Environmental Emergencies. 5th ed. St Louis, MO: Mosby Elsevier; 2007. 7. Ashford DA, Knutson RS, Sacks JJ. Injury among cavers: results of a preliminary national survey. J Sports Med Phys Fitness. 1999;39:71–73. 8. Knutson S. American caving accidents 1980 –1981. Natl Speleol Soc News. 1983;41:99 –118. 9. Knutson S. American caving accidents 1982. Natl Speleol Soc News. 1983;41:255–265. 10. Knutson S. American caving accidents 1983. Natl Speleol Soc News. 1985;42:347–357. 11. Knutson S. American caving accidents 1984. Natl Speleol Soc News. 1985;43:355–376. 12. Knutson S. American caving accidents 1985. Natl Speleol Soc News. 1986 44:399 – 409. 13. Knutson S. American caving accidents 1986. Natl Speleol Soc News. 1987;45:386 –398. 14. Knutson S. American caving accidents 1987. Natl Speleol Soc News. 1988;46:475– 490.
Stella-Watts et al 15. Knutson S. American caving accidents 1988. Natl Speleol Soc News. 1989;47:314 –334. 16. Knutson S. American caving accidents 1989. Natl Speleol Soc News. 1990;48:331–349. 17. Knutson S. American caving accidents 1990. Natl Speleol Soc News. 1990;48:330 –357. 18. Knutson S. American caving accidents 1992. Natl Speleol Soc News. 1993;51:346 –370. 19. Putnam WO. American caving accidents 1996 –1998. Natl Speleol Soc News. 2000;58:1–57. 20. Putnam WO. American caving accidents 2002–2003. Natl Speleol Soc News. 2005;63:1– 48. 21. Putnam WO. American caving accidents 2004 –2005. Natl Speleol Soc News. 2007;65:1– 45. 22. Keeler R. American caving accidents 2007–2008. Natl Speleol Soc News. 2010;68:1– 44. 23. Forrester JD, Holstege CP. Injury and illness encountered in Shenandoah National Park. Wilderness Environ Med. 2009;20:318 –326. 24. Goodman T, Iserson KV, Strich H. Wilderness mortalities: a 13-year experience. Ann Emerg Med. 2001;37:279 –283. 25. Montalvo R, Wingard DL, Bracker M, Davidson TM. Morbidity and mortality in the wilderness. West J Med. 1998;168:248 –254. 26. Better OS, Abassi ZA. Early fluid resuscitation in patients with rhabdomyolysis. Nat Rev Nephrol. 2011;7: 416 – 422. 27. Haojun F, Jianqi S, Shike H. Retrospective, analytical study of field first aid following the Wenchean earthquake in China. Prehosp Disaster Med. 2011;26:130 –134. 28. Mortimer RB. Risks and management of prolonged suspension in an alpine harness. Wilderness Environ Med. 2011;22:77– 86.