Proceedings, International Snow Science Workshop, Banff, 2014

Transcription

1 BURIED ALIVE: EXAMINING ASSUMPTIONS CONCERNING PROLONGED BURIALS AND AVALANCHE RESCUE Scott Savage 1, Dale Atkins 2, Terry O Connor MD 3 Sawtooth Avalanche Center, Ketchum, Idaho, USA Recco AB, Avon, Colorado, USA St. Luke s Wood River Hospital, Ketchum, Idaho, USA ABSTRACT: The probability of surviving an avalanche burial decreases dramatically after minutes, and we teach students they are their partners best chance for survival in a burial event. Rescue education efforts have created two widespread rescue assumptions: prolonged burial searches are body recovery missions, and never leave the scene to go for help. In February 2014, an avalanche in Idaho completely buried a group of four snowmobilers. While one rider was killed, three riders survived; one dug himself out of the debris, the second survived a 60 minute burial, and the third survivor was extricated successfully 105 minutes after the avalanche released. She was recovered by trained recreationists who responded because survivors left the scene to seek assistance. This paper utilizes the Idaho case study to examine the validity and implications of common avalanche rescue assumptions. This winter, at least five prolonged burial survivals occurred in the United States. We review the environmental and physiologic factors contributing to survival during prolonged burial and emphasize that rescuers must operate urgently until victims are recovered and physically examined. We then examine the role and frequency of small-team rescuers : rescuers who are neither part of the involved party nor professionals but play integral roles. We propose a holistic, systems-based approach to avalanche rescue that incorporates individual, small-team, and professional segments in a non-linear, collaborative model. The authors ultimate goal is to encourage the avalanche community to further examine avalanche rescue-related assumptions that shape rescuers mentality and behavior. KEYWORDS: avalanche burial, avalanche rescue, prolonged, shallow 1. INTRODUCTION 1.1 Literature Review Each year, snow avalanches kill about 42 people in North America and 120 in Europe (Atkins 2011a,b). Studies indicate asphyxia is the cause of death in approximately 75% of avalanche fatalities and trauma is responsible for about 25%; hypothermia may also account for a small number of fatalities (Boyd et al 2009). Asphyxia is a condition in which a critical decrease in the amount of oxygen in the body (hypoxia) accompanied by an increase in carbon dioxide (hypercapnia) leads to death. Asphyxiation is caused by several nonexclusive mechanisms: Snow can block or obstruct a person s airway, preventing air exchange. Pressure exerted by debris may critically reduce chest expansion during breathing efforts, reducing the inhalation and exhalation volumes (Stalsberg et al 1989). * Corresponding author: Scott Savage, Sawtooth Avalanche Center, Ketchum, ID Phone (406) scott@sawtoothavalanche.com Avalanche debris contains 30-60% water and 40-70% air (McClung and Schaerer 2006), resulting in less air available for breathing. Rebreathing expired air in an enclosed space results in hypercapnia. Air mixing and gas diffusion and effusion are compromised in avalanche debris, causing CO 2 levels immediately surrounding victims to gradually increase. The rate of development of asphyxia in this scenario is variable, depending on the size of the air pocket and the dynamics of diffusion of expired CO 2 -rich air away from an avalanche burial victim (Radwin et al 2000, Brugger et al 2003). For example, when live people are buried in snow, the warm exhaled air condenses, freezes, and melts surrounding snow, often times creating an ice mask near the person s face that inhibits diffusion and accelerates asphyxia (Radwin et al 2000, Sommerfeld et al1993). Studies show that, even during prolonged burials, augmenting effusion of expired air with artificial air pockets or artificially excluding CO 2 creates an environment with O 2 levels adequate for survival 364

2 and delays asphyxia (Radwin et al 2000, Grissom et al 2000). As asphyxia is the major cause of death in avalanche burials, it intuitively follows that burial duration is highly correlated to mortality (Falk et al 1994, Brugger et al 2001, Haegeli et al 2011). For victims not killed by trauma, probability of survival is approximately 90% if recovered within minutes of burial. The probability quickly drops to ~33% between minutes as asphyxia affects most victims. Survival beyond 35 minutes requires both a patent airway and a sizeable air pocket or suitable air mixing. But even in the case of prolonged burials >35 minutes, survival further drops to near 10%. Hypothermia plays an intriguing role in prolonged burial survival. It has been hypothesized that hypothermia may augment survival rates under hypoxic stress as cooling of the human body decreases oxygen consumption by 7-10% / deg C (Wood 1991). Even though, malignant arrhythmias and cardiac arrest can certainly develop with prolonged burial and hypothermia (Strapazzon et al 2012), the hypothermic stress itself may prove to be neuroprotective. This is reflected in some recent cases in which victims of profound hypothermia subjected to prolonged resuscitative efforts made full neurologic recovery (Gilbert et al 2000, Oberhammer et al 2008). Previous studies show a correlation between burial duration and burial depth with deeper burials generally resulting in longer burial duration and decreased survival (Falk et al 1994, Haegeli et al 2011) but Burtscher (1994) argued that burial depth (independent of burial duration) does effect probability of survival. Limited data has made rigorous analysis of burial depth versus survival difficult. Finally, the majority of the world wide avalanche accident data comes from accidents involving at least one fatality; accidents with fully buried victims but no fatalities often go unreported and are underrepresented, likely introducing a bias towards poorer survival for given burial durations (Haegeli et al 2011). Since the early 1970s (Atkins 2008), avalanche rescue has been separated into three distinct and sequential phases: self-rescue, compancompanion (small-team) rescue, and organized (professional) rescue. If a buried or injured person could not escape on their own, they have to rely on companions or others nearby (Fig 1). If no companions were available, then the person must rely on organized rescue. If companions could not affect a successful rescue, an organized rescue team was called. Within each phase, the focus lies on rescue tools. While professional rescuers are better equipped to locate and treat avalanche victims once on-site, delays arising from notification and travel put avalanche victims relying on professional rescue at a distinct disadvantage. However, new and improved notification technologies and expanded use of mechanized transport have significantly decreased professional rescue response time since 2000 (Atkins 2008). Figure 1. Traditional approach to avalanche rescue. The International Commission for Alpine Rescue has defined an avalanche accident as a medical emergency (ICAR 1999), and any acute incident that produces a risk to life or long-term health requires early notification of rescue services to achieve the best possible outcome. In situations where complexity and uncertainty dominate, and especially in time-sensitive settings, pursuing multiple approaches simultaneously produces the best outcomes (Rashid et al 2013). This simultaneous, multi-path approach to solving complex and timesensitive problems is counter to traditional avalanche rescue thinking and actions. A modern approach to avalanche rescue (Atkins 2012) uses a holistic, systems approach that focuses on the process rather than only individual tools (Fig 2). When time is of the essence, the best outcome to find and care for the avalanche victim is achieved by applying multiple phases and tools in unison because the rescuer does not always know which path will succeed most quickly. The simultaneous union of strategies, tactics, and tools produces optimum results. A linear, sequential approach will result in wasted time. Saving a life is a process that involves the buried victim and possibly their companions, others nearby, professional rescuers, and medical professionals in both pre-hospital and hospital settings. 365

3 Figure 2. Modern approach to avalanche rescue (Atkins, 2012). 1.2 Idaho 2014 Accident In February 2014 in south central Idaho, all four members in a snowmobiling party were completely buried in a destructive force size 3 (D3) avalanche (Savage 2014, Fig 3 photo). The slide buried all four riders with their heads less than 30cm deep. One rider extricated himself in approximately 45 minutes. This first survivor was able to hear and then successfully recovered a second survivor approximately 60 minutes after burial. After the first two survivors were freed from the debris, they immediately began searching for the two missing members of their party. The survivors were experienced riders who were well-versed in avalanche rescue but had never triggered an avalanche or been involved in a rescue. Following multiple transceiver signals, they quickly saw an exposed head and arm. This rider had cleared snow away from part of his helmet but was ashen and determined to be deceased. At this point, the survivors were unable to effectively use their cold hands and were struggling with the close-proximity double burial transceiver scenario using their older single antennae transceivers. The accident site does not have cellular phone coverage and they did not have other communication devices. Recognizing they were unable to affect a timely rescue, they decided to return to the trailhead 4km away via snowmobile to seek assistance by stopping a passing motorist. Fortunately, a motorist was able to quickly get to a nearby lodge/restaurant that is frequented by snowmobilers and skiers. Two recreationists who had recent avalanche rescue training, knew the accident site well, and had accessible snowmobiles quickly responded to the accident site and recovered a third survivor 105 minutes after the avalanche occurred, saving a life. Figure 3. View of the 16 Feb 2014 Idaho avalanche from the top of the debris field. Victim locations are numbered 1-4 (Sawtooth Avalanche Center photo). The chief author of this paper was at the trailhead interviewing the second survivor when the final survivor arrived and also performed the ensuing accident investigation for the Sawtooth Avalanche Center; in the days and weeks following the accident and investigation, two prominent questions arose: The probability of surviving a complete burial longer than 35 minutes is low - were the 3 survivors simply lucky or were there underlying reasons for their survival? Non-professional rescuers who weren t part of the involved party saved a person s life are these small-teams effectively integrated into current rescue algorithms and education efforts? 2. CASE STUDY: FACTORS IN IDAHO SURVIVALS In the February 2014 Idaho accident, three people survived prolonged >35 minute burials. In the authors opinion, two factors appear to have promoted their survival: shallow burial depth and small-team neither companion nor organized rescue. All four victims heads were buried less than 30 cm deep. All of the survivors reported a crushing sensation on their chests and were unable to breathe initially due to the pressure exerted by the debris. However, all three were able to wiggle enough to create space for chest inhalation and exhalation soon after burial. Additionally, the survivors airways were not packed with snow and they were able to create air pockets by clearing 366

4 some snow from the area surrounding their faces. The survivor recovered at 60 minutes was buried on her side and reported that the dense, crushing debris extended to her left armpit while she was able to move her right arm somewhat. The survivor recovered at 105 minutes was able to scratch the snow in front of her face enough to feel a little cold air come in. Her rescuers easily cleared the snow from her face and head using only their hands. These observations indicate that the upper 20-40cm of debris was significantly softer and less dense than the lower portions of the debris. The third survivor s rescuers were an excellent example of a small-team rescue that fits neither companion nor professional rescue definitions. The initial two survivors quickly recognized their physical and equipment limitations and did not waste time, choosing to leave the site to seek assistance soon after their extrication. The smallteam rescuers rapidly responded to the scene, assuming they were looking for a live person. They possessed the skill and training to perform a textbook hasty response long before more organized, professional rescuers arrived at the scene. 3. DISCUSSION Radwin and others (2000) found that human subjects buried in kg/m3 snow produced CO 2 faster than diffusion could keep up, resulting in CO 2 levels dangerous to humans within 10 minutes of burial. As elevated CO 2 levels caused by exhalation appear to be the primary mechanism driving asphyxiation in many avalanche fatalities, maintaining CO 2 levels in the burial environment close to those in ambient air is crucial to survive prolonged burials. Most avalanche debris is a porous material consisting of approximately 50% air and 50% water (500kg/m3). For any given gas, diffusion and effusion from higher concentration to lower concentration in a porous material are largely governed by two material properties: porosity, a measure of the empty spaces in a material, and tortuosity, a characterization of the connectedness of the pores in a material. Porosity and tortuosity are dependent on particle size, orientation, and structure and are difficult to study in avalanche debris; for this reason, most investigators use density as a surrogate for porosity and tortuosity. The authors hypothesize that the burial environment in the upper 30-50cm of debris may differ significantly from deeper portions of the debris in some avalanches, resulting in delayed onset of asphyxia and increased probability of surviving prolonged shallow burials. The chief author has, on many occasions over 20 years as a field-based avalanche practitioner, observed two scenarios that have produced debris with markedly different characteristics in the upper cm relative to the deeper debris: hard slab avalanches and less dense, softer surficial debris resulting from secondary debris flows that produce a final wave of lower density debris. The surface and near-surface debris from hard slab avalanches often times consists of large blocks, resulting in significant pore space and air channels between blocks within the debris and the free air above the surface. The decreased tortuosity (increased connectedness of the pore spaces) would allow faster diffusion and effusion and the larger direct channels to the ambient air would promote some bulk flow of air via advection; this would significantly increase air mixing in the upper portions of the debris as advection is much more efficient than simple diffusion and effusion. The second scenario involves relatively softer, less-dense deposits on the surface of the debris field. From the chief author s experience, avalanches involving very soft, lower density slabs that exhibit smooth, laminar flows (not turbulent) produce softer debris, especially near the surface. Less dense debris enhances diffusion and effusion relative to denser, less porous debris and would promote adequate chest movement for inhalation. Additionally, the chief author has observed several larger avalanches that have smaller secondary flows that run down the track after the main debris flow; these avalanches can create very dense debris fields with a shallow, less dense layer on the surface that would promote air exchange, air pockets, chest movement, and human survival. The Idaho victims ability to move their hands and bodies in the upper 20-40cm of the debris while being unable to move body parts at all in the deeper, denser debris suggests this particular avalanche may be an example of this proposed scenario. If the authors hypothesis is correct, very shallow burials would result in increased survival relative to deeper burials during the minute time period when asphyxia dramatically impacts survival. A thorough statistical investigation is warranted and planned (personal communication, Haegeli, Logan, Brugger). Independent of burial depth, the 2014 Idaho case demonstrates the importance and critical help that others nearby can provide when an accident occurs. Following their extended burials, the initial two survivors were unable to use their hands to grasp shovels or probes. They had one opera- 367

5 tional snowmobile at their disposal as the other three machines were buried. If the initial two survivors had remained on-site and continued with the rescue, the process of locating, extricating, and transporting the third survivor would likely have been delayed and may have resulted in a poor outcome. While this case involved a small-team of other recreationists, the responding team could also have involved a nearby rescue team; Sawtooth Avalanche Center forecasters were in the adjacent drainage less than 5 linear km from the burial site at the time of the accident. Winter time backcountry recreation is booming. According to the Snowsports Industries of America (2011), human-powered snowsports is the fastest growing segment of winter recreation. Snowmobile use also continues to rapidly increase (ISMA 2014). In many locales this explosion in popularity equates to many more users; while this increases the likelihood of accidents, it also increases the chances that capable people may be nearby to help if an accident occurs. It also creates a greater potential to quickly notify professional rescuers. Rapidly evolving communication apps, devices, and technology are making it easier to notify rescuers, including both other recreationists nearby and professional rescuers. 4. CONCLUSIONS The 2014 Idaho accident illustrates several important points. First, avalanche burial victims must be assumed to be alive until confirmed deceased. The avalanche community has done an excellent job educating recreationists, professionals, and educators that we must find buried victims in the initial minutes or they probably will not survive. While this tactic encourages swift rescue, rescue operations extending beyond 30 minutes can be falsely assumed to be body recovery missions. The authors experience is that rescuers attitudes, intensity, and speed change as rescues become protracted. Statistical analysis shows that a buried person who is alive after 35 minutes has a relatively stable chance of surviving the ensuing 60 minutes (Haegeli et al 2011). Since 2000, two victims in the United States have survived burials of nearly 24 hours (Judd 2004) and at least five others have survived burials longer than 5 hours (Atkins 2010). Factors that contribute to prolonged burial survival (patent airway, adequate air pocket, lack of significant trauma, mild hypothermia) cannot be adequately assessed and triaged into treatment algorithms until victims are exposed. We should stress to avalanche rescuers, educators, and students that victims are alive until proven otherwise, even if they have been buried for a prolonged time. Secondly, avalanche rescue and rescue training would be well-served to follow a holistic, systems-based approach that applies multiple strategies and tactics integrating individual, smallteam, and professional segments in a non-linear, collaborative model. Avalanche educators should include information and training on how companions, small-teams, and professional rescuers should interact with each other. The sum of their combined efforts can be greater than their individual efforts. Lastly, shallow burials <30-50cm in specific debris conditions may produce more hospitable environments for victims. If true, these burial scenarios may result in a higher than previously reported probability of survival in the case of shallow prolonged burials. The rescuer rarely knows how deeply a victim is buried or whether the victim has an air pocket or patent airway. Since survival is uncertain, rescuers should not ask Is the person dead? Instead, rescuers should always ask Could the person still be alive? ACKNOWLEDGEMENTS The authors wish to thank the survivors of the 16 Feb 2014 Idaho accident for their forthright accounts and encouragement on this project; Pascal Haegeli and Spencer Logan for thoughtful discussions on accidents and avalanche burial and survival; the Sawtooth National Forest and Sawtooth Avalanche Center for latitude to work on this project; and Simon Trautman for reviewing this paper. REFERENCES Atkins D: Colorado Avalanche Information Center: US & world avalanche accident stats [serial on- line]. Available: (accessed Aug 20th, 2014). Atkins D, 2008: Time to change rescue attitudes for a new generation. Proceedings of the 2008 International Snow Science Workshop, Whistler, BC, Atkins D, 2010: Avalanche rescue: the United States experience. Proceedings of the 2010 International Snow Science Workshop, Squaw Valley, CA, Atkins D, 2011: US Avalanche Accidents2001/ Student Handbook National Avalanche School 2011, Snowbird, UT. Atkins D, 2011: Avalanche rescue report. International Commission for Alpine Rescue to the Mountain Rescue Association, Oct Atkins D, 2012: Vision zero: applying road safety to avalanche safety. Oral presentation at the 2012 International Snow Science Workshop, Anchorage, AK, 17 Sept

6 Boyd J, Haegeli P, Abu-Laban RB, et al, 2009: Patterns of death among avalanche fatalities: a 21 year review. CMAJ, 180, Brugger H, Durrer B, Alder-Kastner L, et al, 2001: Field management of avalanche victims. Resuscitation, 51, Brugger H, Sumann G, Meister R, et al, 2003: Hypoxia and hypercapnia during respiration into an artificial air pocket in snow: implications for avalanche survival. Resuscitation, 58, Falk M, Brugger H, Adler-Kastner L, 1994: Avalanche survival chances. Nature, 368, 21. Haegeli P, Falk M, Brugger H, Etter H, Boyd J, 2011: Comparison of avalanche survival patterns in Canada and Switzerland. CMAJ, DOI: /cmaj Gilbert M, Busund R, Skagseth A, et al, 2000: Resuscitation from accidental hypothermia of 13.7 degrees C with circulatory arrest. Lancet, 355, Grissom C, Radwin M, Harmston C, Hirshberg E, Crowley T, 2000: Respiration during snow burial using an artificial air pocket. JAMA, 283, Grossman M, Saffle J, Thomas F, Tremper B, 1989: Avalanche trauma. J Trauma, 29, Judd R, 2004: An avalanche and miracle that saves two students lives. Seattle Times, 23 Feb International Snomobile Manufacturers Association (ISMA), 2014: Available: (accessed 21 Aug 2014). Locher T, Walpoth B, 1996: Differential diagnosis of circulatory failure in hypothermic avalanche victims: retrospective analysis of 32 avalanche accidents. Schweiz Rundsch Med Prax, 85, McClung D, Schaerer P, 2006: The Avalanche Handbook. 3 rd ed The Mountaineers, p 245. McIntosh S, Grissom C, Olivares C, et al, 2007: Cause of death in avalanche fatalities. Wilderness Environ Med., 18, 293. Oberhammer R, Beikircher W, Hormann C, et al, 2008: Full recovery of an avalanche victim with profound hypothermia and prolonged cardiac arrest treated by extracorporeal rewarming. Resuscitation, 76, Radwin M, Grissom C, Scholand M, Harmston C, 2000: Prolonged oxygenation during snow burial; the role of carbon dioxide exclusion. Proceedings of the 2000 International Snow Science Workshop, Big Sky, MT, Rashid F, Edmonson A, Leonard, H, 2013: Leadership lessons from the Chilean mine rescue. Harvard Business Review, Jul-Aug, Savage S, 2014: Sawtooth Avalanche Center accident report on 16 Feb 2014 Frenchman Ck fatality. Available: Accident pdf. SIA, 2011: 2011 snowsports participation report. Snowsports Industries of America, 11 Jun Sommerfeld RA, et al, 1993: C02, CH4, N20 flux through a Wyoming snowpack and implications for global budgets. Nature, 361, Stalsberg H, Albretsen C, Gilbert M, et al, 1989: Mechanism of death in avalanche victims. Virchows Arch A Pathol Anat Histopathol, 414, Strapazzon G, Beikircher W, Procter E, Brugger H, 2012: Electrical heart activity recorded during prolonged avalanche burial. Circulation, 125, Wood S, 1991: Interactions between hypoxia and hypothermia. Annu Rev Physiol, 53,