Wednesday, August 17, 2016

The use of stomach contents to estimate time of death

The present article focuses on the contents of the stomach and how they change with time, in the context of whether or not one can obtain useful information on the time of death.  The stomach uses a combination of mechanical action and hydrolytic enzymes such as pepsin to break food down into smaller particles and simpler chemicals.  Kong and Singh (2008) reviewed the physiology of the stomach, including factors that affect the rate at which the stomach empties.  A website at Colorado State has a helpful graphic showing the transit times of material in the digestive system.

 The digestive system potentially has more than one type of evidence that it can provide an investigator.  The rate at which digestive material moves from the stomach into the small intestine can suggest something about the time of death relative to the time of the last meal. This issue was the subject of a previous blog entry that considered the Knox/Sollecito case.  We will start with a relatively recent addition to the literature on stomach contents and time of death, and then examine individual cases, most of which come from older literature.

Patel et al. (2013)
Patel and coworkers coauthored a paper which correlated the contents of the stomachs of 100 deceased individuals with known times of death and times of last meal.  The authors divided the contents as follows:

Category I:      semi-digested food particles
Category II:     semi-digested unidentifiable food particles
Category III:    empty stomach

Category I was found more commonly in those dying within 0-2 hours after their last meal.  Category II was found more commonly in those dying within 2-6 hours after their last meal.
Category III was found more commonly in those dying more than 6 hours after their last meal.

19 cases fell into category I, of which 16 (84%) had consumed their final meal in 0-2 hours before death, 2 cases had eaten 2-4 hours before death (11%), and one person had eaten 4-6 hours before death (5%).  For category II, the most likely situation encountered was that the last meal was taken 2-4 hours before death (49%), but there were a number of cases in which it was 0-2 (13%), 4-6 hours (26%), or greater than 6 hours (11%).  The span of 2-6 hours covered 75% of the individuals.  For category III the most likely situation encountered was that the last meal was consumed more than 6 hour before death.

The authors discussed the composition of the meal, noting that a carbohydrate meal leaves the stomach most quickly, a protein meal leaves at an intermediate rate, and a fatty meal leaves the stomach most slowly.  I will use this paper as a guide to evaluate some earlier cases.

Horowitz and Pounder (1985)
Horowitz and Pounder wrote, “There is a significant variation in emptying rates between normal subjects, but under controlled conditions gastric emptying of a test meal is relatively reproducible in normal individuals.”  These authors note that many things can affect the rate of gastric emptying, such as certain drugs, stressful stimuli, and diseases.

They wrote, “In those occasional instances where the gastric contents provide the only practicable means of evaluating the time of death some guidelines can be offered.  First only the digestible solid component of the meal should be assessed…Second the weight of the solid component should be compared with the estimated weight of the known last meal…Third, the confidence limits of any opinion should take into consideration the many possible variables, so that the estimate given should cover a ranges of at least some hours.”  They conclude by saying, “Estimates to within half an hour clearly cannot be justified in the light of present knowledge of patterns of gastric emptying.  For forensic purposes the stomach is a very poor timekeeper.” 

Jaffe (1989)
Dr. Frederick Jaffe authored an article in which the abstract read in part, “Using it [stomach contents] as a guide to the time of death, however, is theoretically unsound and presents many practical difficulties, although it may have limited applicability in some exceptional instances.”  The article discussed three cases.  I will take the facts as presented in this article as givens, but a few of them are in dispute.  I will mention the controversial points of which I am aware without taking a position on them.

In the Steven Truscott case, the victim ate at 5:50 PM and was in the presence of Steven Truscott between 6:30 and 8:25 PM.  The pathologist put the time of death as being between 7:15 and 7:45 but later had misgivings about this range (Dr. Jaffe’s remarks indicate strong skepticism about this level of precision).  One source indicates that Steven Truscott was seen at 8 PM on the grounds of a school.

 In the David Hendricks case, Mr. Hendricks’ wife and three children were killed.  Fragments of vegetables were found in their stomachs.  One expert put the time of death at 2-4 hours after the consumption of pizza, the known last meal, and two put it at 1-3 hours.  Mr. Hendricks left his house on a trip 4.5 to 5.5 hours after the meal.

In the Crimmins case, the two children were fed at 7:30 and an abduction would have taken place about 9 hours later, if Mrs. Crimmins’ account were accurate.  Yet, recognizable food was found in one of the children’s stomachs.  Assuming that their meal was indeed manicotti and string beans (Mrs Crimmins stated that it was veal), Dr. Jaffe indicated that this was not reasonable.  This is the only case in which Dr. Jaffe implies that stomach contents were useful in ascertaining an approximate time of death.

Davis (1989)
Dr. Joseph Davis wrote a letter in response to Dr. Jaffe’s article, focused on the Hendricks case.  Dr. Davis discussed how the length of time for the stomach to empty increased with increasing size of the meal.  With respect to the three children he wrote, “Although the evidence was not as close to absolute as in the Crimmins case, which involved an 8 ½ h postprandial period, it is well within accepted probability that a coincidental delay in gastric emptying time would not occur in all three victims.”

Pope (2012)
During an attempted robbery in Eugene, OR, the suspect was killed in an exchange of fire with a barista.  Examination of his stomach contents revealed contents suggesting the consumption of a bacon cheeseburger and French fries.  The size of the French fry was a clue that helped narrow down the identity of the fast food restaurant.  This information prompted investigators to examine security footage from a local fast food restaurant, where they identified the deceased suspect and a second suspect, who was later arrested.  About an hour elapsed between the time that the two suspects were in the restaurant and the time of the attempted robbery. 

In the Truscott case, the victim probably died within 2 hours of consuming her last meal, but she might have died up to about four hours after her last meal.  In addition, there is some uncertainty about the time that Truscott was later seen.  There is no support in the literature for the narrow 30-minute window.  If the victim ate near 5:50 and Mr. Truscott left her presence before 8 PM, it is quite possible that she died afterwards.  Therefore, the stomach contents were overinterpreted in his case.  In the Hendricks case the TOD was unlikely to have been more than four hours beyond the consumption of the last meal, and the similarity among the contents of the three stomachs means that any factor that might have slowed digestion for one child would probably have to have slowed it for all three.  In the Crimmins case the contents of the stomach were correctly interpreted as ruling out an abduction later than 4 AM, and the only uncertainty is in what the children ate for dinner.

Dr. Jaffe correctly cautioned his readers about dangers of overinterpreting stomach contents as they relate to TOD.  Yet there are clearly examples in which information on the time of death or time of last meal based upon contents of the stomach was helpful to investigators.  It seems that stomach contents are more useful under some circumstances than others.  The estimate will be in the form of a range of several hours.  If the time of the last meal is not known, then only a relative time of death can be ascertained.  If the time of a meal is known, but the type of food is not known, then some uncertainty exists with respect to whether or not it is the last meal that the deceased person ate.  If the time and nature (size and type of food) of the last meal is known, then if identifiable material is found, the last meal was probably consumed within two hours, with an outside chance that it was consumed between two and six hours before death, unless some exceptional circumstances are at work.  If the stomach is empty, then the time of death is likely to be more than 6 hours after consumption of the last meal, but it could be 2-6 hours.

Davis JH, “Gastric Emptying Time” (1989) American Journal of Forensic Medicine and Pathology 10(3) 271-272.
Horowitz M, Pounder DJ, “Is the stomach a useful forensic clock?” (1985) Austr NZ J Med 15, 273-276.
Jaffe F, “Stomach Contents and Time of Death:  Reexamination of a Persistent Question” (1989)
American Jounal of Forensic Medicine and Pathology 10(1) 37-41.
Kong F, Sing RP “Disintegration of Solid Foods in the Human Stomach” (2008) Journal of Food Science 73(5) R67-R80.
Patel V, Silajiya D, Shah K, Menat A, Tandel M, and Raloti S, “Estimation of time since death by gastric contents” (2013) IJCRR 5(11), 125-129.
Pope L “Identification of a Second Suspect via Stomach Contents at Autopsy” (2012) J Assoc Crime Scene Reconstr. 18(2) 13-15.