Forensic Disaster Response: The Crash of Comair 5191
Challenges, issues, and solutions of identification in mass disasters differ with the type and scope of the catastrophe.
At approximately 6:10 A.M. on Sunday, August 27, 2006, Comair Flight 5191 taxied into takeoff position at Blue Grass Airport in Lexington, Kentucky. Neither tower controllers nor the crew on the flight deck noticed the aircraft had turned on to unlit secondary Runway 26 — only half as long as intended 7000 ft. primary Runway 22.
Too soon after the Bombardier CRJ-100ER began its takeoff roll the 50 seat commuter aircraft ran out of concrete. With the crew fighting to get the doomed plane airborne, the plane’s landing gear clipped an 8 ft. perimeter fence at 158 mph with its tail dragging the ground. The plane failed to clear a nearby grove of trees; the tail separating from the fuselage. The aircraft then slammed into a hillside no more than 1000 ft. from the end of the runway and exploded.
A PLANE IS DOWN
Less than 90 minutes later, University of Louisville dentistry professor Mark Bernstein, DDS, and a diplomat of the American Board of Forensic Odontology, home asleep in nearby Borden, IN, is informed by Tracey Corey, MD, the Chief Medical Examiner (CME) of the Commonwealth of Kentucky, to assemble his odontology team. A plane is down at Bluegrass Airport, 50 onboard, one survivor.
“Although I lecture on mass disaster preparedness as a component of a University of Louisville Bioterrorism grant, and recently participated in a mock disaster exercise at the Greater Cincinnati Airport, the immediate reaction to such news is momentary confusion and inaction,” Bernstein said.
The shock is transient. Bernstein immediately passes the word to faculty colleague Ryan Noble, DDS. Within minutes the remaining members of Bernstein’s odontology team—Drs. James Woodward, William Lee, and Corky Deaton—are also notified, as is Virginia Woodward, the team’s dental hygienist. All are to report to the medical examiner’s facility in Frankfort, 18 miles from the crash site.
Bernstein and Noble first swing by the University School of Dentistry to gather supplies and equipment that have been stored for just such events. There is one snag. The mobile x-ray unit is not exactly portable—it weighs 500 pounds and must be dismantled, then somehow maneuvered into Bernstein’s SUV. Campus police help hoist the unit. Other gear in the emergency cache includes antemortem and postmortem dental charting forms, a computer, a Dexis digital x-ray probe and software package, miscellaneous containers, and a digital camera.
The odontologists depart for the state forensic laboratory in Frankfort, arriving by noon, before the bodies.
“A perimeter of security surrounds the building but we are anticipated, and promptly admitted,” Bernstein said.
Meanwhile, Corey and Fayette County coroner Gary Ginn have activated the remainder of the mass disaster team, including 22 coroners, seven pathologists, and Emily Craig, Ph.D., the state anthropologist.
Members of the Kentucky Mass Disaster Team (KMDT), a group of county coroners from the Kentucky Coroners Association formed in the early 1990s to respond to state mass fatality incidents, are already at the crash scene, removing bodies from the wreckage. First members of KMDT are on the scene within 30-45 minutes.
“Coroners do the body recovery,” said Lee, who happens to be the coroner of Hardin County (and the only coroner-dentist in Kentucky).
Corey said Kentucky is lucky that it already had in place a mass fatality response team of experienced death investigators.
“When I arrived at the scene many of them were already on hand,” she said.
Over the years, the medical examiner’s office and KMDT had trained and formulated disaster plans and procedures. Plans must be flexible.
“Every mass fatality incident is going to be unique, and although you can have policies and procedures you are always going to have to modify those procedures to fit the situation that arises,” Corey said.
Unlike Hurricane Katrina, where victims were widely scattered, here crash victims are confined, facilitating recovery. By 4 P.M., remains begin arriving at the Frankfort morgue, where they are logged, placed on gurneys, and rolled into refrigerated trucks positioned earlier in a secure area.
“Sometimes, you might have to work in a tent or temporary facility, but here we were fortunate to be so close to the Medical Examiner’s Office in Frankfort,” Lee said.
Once the morgue operations are ready, bodies are brought in groups, tagged, photographed, and personal effects catalogued.
Mistaken identity of the dead at mass fatality incidents is always a risk. In April, 2006, a terrible mistake was made after two Indiana college students who resembled each other were involved in a car crash near Fort Wayne, Indiana. One died, one survived comatose. The family of the deceased was told their daughter had died in the crash, only to learn five weeks later that the victims had been misidentified at the scene.
Lexington circumstances are different, but Corey takes no chances; she requires that a coroner accompany each victim to each station during identification, eliminating any potential for mislabeling or loss. Days later, as the bodies are released to families, Corey double checks each body bag, just to be sure.
“All victims display varying degrees of fire effect,” Bernstein said. “Many have blunt force trauma.”
Clothing is removed first and autopsies performed by the pathologists, who not only look for identification clues but attempt to determine cause of death. While all deaths are related to the crash, specific cause for each individual—whether from injuries from the impact or smoke inhalation—must be ascertained for the death certificate.
“A person might have had a heart attack and be dead before the crash,” Lee said.
After autopsy, the remains go to dental processing, where Bernstein exposes the dentition, noting fracture patterns, dental damage, airway soot, color of tissues.
“These factors may help reconstruct the cause of death,” he said. The hygienist, who serves a scribe, records the findings. Odontology teams photograph the dentition and chart the dental findings on standard postmortem forms.
Victims are then returned to the trucks.
The forensic teams work until 9:30 P.M. Sunday night and return at 7 Monday morning. By Monday evening all 49 decedents have been examined.
“Remarkably, almost no teeth or jaw fragments are lost or co-mingled,” Bernstein said. One segment of an upper jaw not present with its remains is found at the scene two days later.
Elsewhere, Comair arranges for the victims’ families to gather at a Lexington motel. FEMA’s Disaster Mortuary Operational Response Team (DMORT) family assistance crew, along with the Fayette County coroner’s office, are busy collecting medical and dental records of those on Flight 5191’s manifest.
Family assistance is critical.
“Someone must sit down with 49 emotionally charged families and ask them detailed questions to help identification—color and length of hair, color and length of fingernails, and so forth,” Lee said. “You need people who are trained for this.”
All crash victims are adults, most are locals, and most have considerable dental work—factors that contribute to rapid acquisition of antemortem dental records. The local medical community responds compassionately, Bernstein said.
“Indeed, 47 antemortem records were received by Tuesday,” he said. By Tuesday evening antemortem records and postmortem records are matched and tentative identifications are being made.
Bernstein organizes the antemortem records into two groups, male and female. The most characteristic feature of each victim is selected, then postmortem records from that gender group are visually scanned to find matching characteristics.
“If only one postmortem record showed a similar characteristic, it was considered a tentative identification,” Bernstein said. If more than one victim shared the characteristic, a second characteristic is selected for comparison.
Bernstein said that for each tentative identification made, confirmation is established by comparing all charted dental features, noting any dental similarities while accounting for any discrepancies. Most identifications are straightforward.
“Some are difficult, either because of little dental work or the antemortem records are old and temporal changes have occurred,” he said. In these instances, digital postmortem radiographs are used to compare with antemortem films.
“Radiographs provide specific dental silhouette patterns and show subtle anatomic features that can be objectively compared with precision to ensure identification,” Bernstein said.
In the end, dental comparisons establish 47 identifications. Of the two individuals on whom no dental records are obtained, one is identified by fingerprints and the other by a radiographic comparison of an antemortem x-ray of a previously damaged finger bone.
All 49 victims are positively identified by Wednesday evening.
The loss of Flight 5191 is the first mass fatality incident in Kentucky since a 1988 bus crash.
“Comparison of each accident illustrates why it is axiomatic that no single protocol for managing mass fatalities is appropriate in all situations,” Bernstein said.
In both incidents, victims were burned but remained relatively intact, and dental records were quickly recovered. However, because the bus victims were children, there was very little dental work.
“Children lose teeth and new ones erupt so rapidly, their antemortem radiographs of only a few months earlier appeared much changed from current dentitions,” Bernstein said. This mandated making dental radiographs on each victim and copying antemortem radiographs on each patient so that subtle anatomic features of teeth and bone could be compared, a difficult and time consuming task.
Flight 5191 response was managed differently.
The decision to delay making dental x-rays and not to copy all of the antemortem x-rays was made because clinical attributes quickly documented by digital photography rendered so much dental information that these comparisons alone allowed unconditional identification in most cases. In cases where there was any uncertainty, radiographic comparisons were performed.
“This modification saved time while not compromising accuracy, thus optimizing the identification effort,” Bernstein said.
Dental results were presented to Corey, who cross-checked them against suspected identification based on personal effects, medical records, and anthropologic data.
“The rapidity and effectiveness of the response allowed us to release the remains to families as quickly as possible without compromising accuracy,” Bernstein said.
Corey said her strongest advice to other CMEs or disaster planners who may have to direct a response is to find the right people.
“The most important thing in planning is to have contacts with forensic people with experience and expertise who you will need to call on,” she said.
Identification Clues, Like the Phoenix, Rise from the Ashes
Forensic science can sometimes fail to identify disaster victims.
Explosions and intense fires leave precious little to work with. Traditional dental x-ray comparison may be unsuccessful if structural relationships of the jaw are lost. When airplanes crash soon after takeoff with a full load of fuel, ensuing fires are long and hot. Teeth warp, making traditional forensic identification impossible.
In a new development, University at Buffalo (UB) forensic dental researchers have found that evidence still exists among the ashes when all else – flesh, bones, teeth, DNA – is lost. Researchers there have demonstrated that inorganic resins that make up the central matrix of dental fillings not only withstand temperatures of 1,800°F, they can be recovered and identified.
“When fire burns hot enough and long enough, all other clues are destroyed,” said Mary Bush, DDS, assistant professor of Restorative Dentistry, UB’s School of Dental Medicine. Identification at this point can only be made from nonbiological evidence found with the victim. Dental prosthesis, such as crowns and partial dentures, are examples of nonbiologicals that survive inferno conditions, she said.
Bush found that not only are resin fillings retrievable, but they can be identified by brand or brand group. Brand groups are important because some manufacturers use the same inorganic elemental formulations.
“Inorganic elements stay virtually unchanged even after cremation conditions,” Bush said.
These materials can then be checked against dental charts to provide identifying information when other means of identification are exhausted. Or, it can provide ancillary certainty if other clues remain.
Odontologists have noticed this work. “Forensic dentists have all seen the effects of high heat on filling materials, but this research documents these effects for use by odontologists,” said forensic odontologist Ann Norrlander, DDS, a University of Minnesota assistant professor of Diagnostic and Biological Sciences.
In most cases where destruction of dental evidence is minimal to moderate, odontologists will continue to rely on dental radiographs alone for comparisons, but with severe dentition destruction, information collected through the Bush method may make the difference between a possible or positive identification and one where determination is impossible, Norrlander said.
Arnold S. Hermanson, a Prairie Village, Kansas, dentist, and member of the American Society of Forensic Odontology, said Bush has created a taxonomic chart of the composite kingdom.
“This database will allow investigators to speciate every resin ever placed in a tooth,” Hermanson said. “With portable equipment, on-site brand identification can be made.”
So far, Bush’s technique works well in the lab, but further work remains before it appears in the field. In a new paper (Journal of Forensic Science, Jan 2007) Bush reports the possibility of using a portable x-ray fluorescence detector to perform field analysis.
“This hair-dryer sized machine will show a spectrum in about six seconds,” Bush said. The spectrum is based on elemental composition, which are compared to known compounds.
Identification of restorative material is rendered useless, however, if dentists don’t accurately record type and brand of material used.
“Quality dental record-keeping has been a historic problem facing forensic odontology,” said UB researcher Raymond Miller, a UB clinical assistant professor of Oral Diagnostic Sciences.
Miller said dental education needs to emphasize the importance of maintaining accurate dental records, including not just procedure but materials and brands.
“Restorative dentists need to understand the forensic value of this information,” he said. “The circumstance may be rare, but the results may bring justice or closure to a complex case.”
Douglas Page writes about forensic science and medicine from Pine Mountain, California. He can be reached at firstname.lastname@example.org.