Thank you, ladies and gentlemen. Please be seated. Doctor.
Lakshmanan Sathyavagiswaran, the witness on the stand at the time of the noon recess, resumed the stand and testified further as follows:
Doctor, we were talking about how core temperature is measured and you have told us about the liver and the rectum and you mentioned the brain. From your own experience and review of any literature do you have an opinion as to whether the core temperature from each of those sources is identical?
The brain temperature would reflect the core temperature the best, and the degrees would correspond better with the postmortem--as a postmortem interval progresses. The rectum would be the next in line and liver is compatible to the rectal temperatures.
No, is better than the other two because it reflects more of the core temperature.
Is there a--well, actually let me ask Mr. Fairtlough to put a graph on the board. This is from page 30 of Henssge's text. Ask if this is material that you referred to, considered and relied upon in part in forming opinions regarding the estimation of the range for time of death? And if Mr. Fairtlough could just pull it back just a little so that we can see the bottom of the graph as well. Little more. Thank you, Mr. Fairtlough. Doctor, what does this graph represent?
This shows the temperature drop in relationship to postmortem interval and you see the graph for each area being used to see the drop in temperature. The no. 3 refers to the rectal temperature which is between the liver temperature which is no. 4, and brain temperature is no. 5.
Doctor, there is a legend that Mr. Fairtlough is pointing to which has a number corresponding to the source for any particular temperature reading; is that correct?
And then we have three different and distinct lines to represent the relationship between the time in hours following death and the temperature as measured; is that correct?
And I don't know if Mr. Fairtlough can perhaps put a big "L" or something associated with that line.
Mr. Fairtlough didn't like his initial "L." And could we put an "R," Mr. Fairtlough, in some form.
Could Mr. Fairtlough put a "B." And your Honor, I will ask when this is printed out, if we could, please, to have this marked as exhibit 367-B, as in boy.
Doctor, from this photograph does it appear that the actual temperature recorded in relationship to the time since death differs depending upon the source used?
Does this difference between rectal, liver and brain as the source affect how precise one can be, given that you are using one source, the liver, rather than all three sources?
Yes. That is one of the main problems in this kind of estimation of time range since death, because your variability in what kind of method you used to gauge the body temperature.
Now, doctor, going on in this first chart, you've got the--you've got the probe identified and I think you have already told us that is the type of instrument used.
And in fact is this a photograph, two photographs of the actual probe and dial that were used--that was used with respect to the liver temperatures of Nicole Brown Simpson and Ronald Goldman?
Let's go to item 3, talking about how range for time of death is calculated based on the measurement of core temperature. Before we get to what is described as "Assumptions required to apply formula," are there various approaches mathematically that are used to use temperature as a basis to estimate a range for time of death?
Basically in--in the past there has been a formula which has been used by Moritz, I think, where they used to take the time interval between death and the temperature which is obtained. And let's say that the temperature you obtained is for assumption 80, and you assumed the normal body temperature to be 98, you have an 18-degree difference. They used to divide it by 1.5 and you have a one and a half degree drop per hour after death and--but this formula didn't take into account the plateau phase that you have after death wherein the temperature really doesn't drop soon after death. This is a period, depending on where the temperature is taken from, you may or may not have a period of time when the temperature may not drop, so there is that variable fact to which doesn't--this formula doesn't take into account. There is one other formula which has been used in the past, and there have also been other formulas which have been used, but this other group of people, Marshall and Hoare, they have a chart wherein you can use that chart to estimate the range, so there are different formulas which can be used. I alluded to one particular nomogram earlier of Henssge which he has taken some of the variable factors, factored it in in the graph to estimate, so bottom line is it is still a statement of range. It doesn't give you a precise estimate of the data.
For the benefit of the ladies and gentlemen of the jury and the reporter, I don't believe we picked up, at least I didn't, the name that you associated with the first formula?
Thank you, doctor. Now, doctor, the formula you described of roughly 1.5 degrees per hour, is that a formula that in your opinion yields a precise period for the postmortem interval to the time when the temperature was measured?
No. It is just an estimated range. So let's take the example I gave you, 98.6, 1.5 degrees an hour, and let's assume that the temperature you take when you see the remains to be--at that time to be 80 degrees. You are talking about 18 degrees and you divide it by 1.5, you are talking about roughly for twelve hours, and so you cannot assume--let's say the 80 degree temperature was taken at 1:00 P.M. in the afternoon. You cannot tell that the person died at 1:00 A.M. using that formula, because of the variable factors, one of them being the plateau phase, and no. 2, the postmortem drop in temperature doesn't occur at a constant rate, because there is variability to the drop in temperature itself and the formula you use assumes that the postmortem temperature is going to drop at a constant rate and it doesn't happen that way, because you have environmental factors which can play a role, like wind and where the body is. The clothing on the body can play a role. There are a number of variable--variables in the environment, on the body itself, clothed or not clothed, where the body is in contact, also the preexisting disease process in the body. There are so many variables in this so you really can't use this mathematical formula to say this is the time of death. There is going to be variability in the time you get. And you are to couch your opinion keeping these variable factors in mind and always give an estimated range, because you can't be precise in estimating a time.
Now, doctor, in this section, "Assumptions required to apply formula," does each of the forumlas you've described make an assumption as to what the actual core temperature was of the decedent at the moment of death?
Yes. What we do is usually everybody thinks everybody has got a normal temperature, 98.6, and as I just gave you in my example, that may not be the case, so the first assumption which may be wrong is that assumption that the body temperature was 98.6 when the person died, and that may be the first assumption which would be wrong.
Doctor, if you are not physically with the decedent at the exact moment of death and at the exact moment of death take the decedent's core temperature, is there any medical way of which you are aware that the actual core temperature at time of death can be determined?
Does that assumption limit the preciseness with which one can estimate a range for time of death?
Because even in a normal person you can have variability in the temperature. You may have a higher temperature in the afternoon than in the morning and that variation is .5 degrees just in an individual variation. And also the normal temperature for different populations. If you take a population of persons, you have a variability in--in the temperature. If you take about a hundred people and you measure their body temperature, of course excluding the individual variability, you will see variability in the normal body temperature. What is one person's body temperature may be 37, another person's may be 36.5 and there is a variation which is known in the population also.
Doctor, you have been using numbers some of which are 98.6 and others which you have just said like are 37.
98.6 I'm using the temperature when you use Fahrenheit which is one method of determining temperature, and the other one is 37 degrees, I use for centigrade which is the other method of measuring temperature.
And doctor, is there a standard formula that is used to convert temperature that is expressed in Fahrenheit to centigrade and visa versa?
And in fact in going through these materials have you attempted to identify temperatures by both centigrade and Fahrenheit?
In the book, this book by Henssge and the others, is it primarily referring to centigrade readings?
Now, doctor, in your own experience have you found that people have different normal temperatures?
That is, if you took my temperature and your temperature and temperatures from other people, that we would not necessarily, even if we were all healthy and this was our normal temperature, we would not necessarily have the same temperature?
Well, that could be their normal body temperature and also the time you take it, you could be taking it at a time when they have their individual variability in temperature. As I said, that in a normal day you may have variability in your temperature. Mornings are usually lower than afternoons and then also certain times of the month, especially in the female population, you have temperature rise during the menstrual period, so depending on which phase of the month or what day and even time, you can have variability between persons. I just gave you some examples.
Let me invite your attention and ask if this is part of what you read, considered and relied in part on in forming your opinions from page 10 of the Henssge book under "Body temperature at the time of death." "It is very difficult to specify a normal body temperature as this value can vary considerably between individuals. Rectal temperatures in a group of healthy subjects can vary between 34.2 and 37.6 degrees centigrade with a mean of 36.9 degrees centigrade." Doctor, are you familiar with the term, "Mean," m-e-a-n, as it is used in this sentence?
That will be the average temperature if you--if you take the hundred people and you take a temperature and you do a standard deviation graph, it will be the mean which you get for the hundred people, the average, average temperature.
Let me continue: "Rectal temperature is often referred to as `deep central temperature' similar in value to that of the brain, heart, lungs and abdominal organs." Is that part of what you read, considered and relied upon?
Continuing on: "Many factors influence body temperature. Most individuals show a diurnal rhythm in witch the body temperature fluctuates plus or minus .5 degrees centigrade around the person's normal mean temperature. These cyclic patterns persist regardless of activity or disease states." Again part of what you read, considered and relied upon?
Doctor, what is the effect, if any, on the preciseness of being able to estimate a time of death based on temperature if you cannot know in fact what the person's normal temperature is and if you cannot know in fact where within a range, due to the diurnal cycle, the person's temperature is on a particular day?
Because you are relying on a particular temperature and you do not know these other variable factors, naturally your estimation would be off by a couple of hours, so naturally your estimated range may not be accurate because the data you are using is not accurate.
Your Honor, I have another board and I just lost--could I ask Mr. Fairtlough to help me out for one moment?
And your Honor, I have a board entitled "Some limitations to the preciseness of estimating postmortem interval, PMI, from body temperature measurements." May this be marked as exhibit 367-C.
And a second board "Relationship between postmortem interval, PMI, and body cooling" as D.
Doctor, let me put up this board C, and again with the Court's permission ask you to step down. Doctor, if you could use the pointer for just a second so I can show you what area I want to have you discuss. There appears to be an area here marked 99.7 degrees Fahrenheit, 37.6 degrees centigrade and an area down below 93.6 degrees Fahrenheit, 34.2 degrees centigrade, and there appears to be a lighter blue rectangle that is covering the distance between those two indicated areas. Do you see that?
It basically represents the range of values for normal core temperature in living human beings as found in the literature which we just discussed. This is a chapter by Nokes on a study where he has done and found this variability in several people. He measured the core body temperature and they range from 34.2 centigrade to 37.6 centigrade. The range is pretty wide. I think the variability in my experience is more within a degree of 98.6 either way if you take the average normal population.
Average normal temperature in your experience is one degree Fahrenheit either way of 98.6?
Doctor, just employing the 1.5 degree per hour cooling rate that you indicated with that initial formula, what does the difference between having a 99.6 normal temperature have versus having a 97.6 temperature have?
Because let's take the same example I gave, 80, let's say the liver temperature you take is 80. If you had 99.7 as the real core temperature, the drop of temperature is 20 degrees and you decide by 1.5, just take the rough formula, you get the number of 13 hours. But let's take the 97.5, take the same example of 80 degrees and you divide by 1.5, I'm just giving the rough example we discussed, and you divide by 1.5, you come to a number of hours as 11 and a half. And just in the simple example you see there is a one and a half degree variation in your estimation and again that is not a precise time of death, but there is a difference in the calculation just using the simple formula we just used.
I think you may have misspoken. Did you say one and a half degrees and intend to say one and half hours?
I'm sorry. Is the difference, though, between the example you gave of 80 with 99, roughly, 6, and 80 degrees and 97.6, a difference when you get between 13 hours and 11 and a half hours?
All right, doctor. If you could just--is it in fact correct medically to say that a person may not have in fact what is a normal core temperature due to illness or any other factors at the time of death?
That is another factor one must keep in mind, because if somebody has preexisting diseases, like thyroid disease or hyperthyroidism, they may have a higher temperature than what is normally expected or they have an infection, they have a fever from a flu, the temperature would be higher. So when you make an assumption that somebody had a normal temperature and when you find the body has already cooled off, that variable factor you may not be able to speculate unless you have obvious pneumonia or meningitis at autopsy.
Unless you have some obvious infection even at autopsy, like pneumonia or meningitis.
Thank you. Doctor, let me put up this next board, 367-D. Let me just put it up here for the moment and ask if in general terms you are familiar with what is shown in this?
Let's start with the blue dash line and there appears to be a legend indicating that this is a rectilinear curve. Is that what that is?
Yes. This graph assumes that the temperature drops at a constant rate after somebody dies and this is--this photograph would represent the rough formula which I just alluded to earlier.
I misspoke. A one and a half degree drop per hour. What is represented here, if any, which accurately at least in a qualitative sense represents the reality of the drop in temperature in relationship to the time since death?
The red graph, the red line, which is more like a sigmoidal curve, would represent the real correct type of temperature drop which you would normally see, because after death you have a phase when the temperature may not drop because of heat produced by dying tissues or intestinal bacteria and the temperature may even go up after death sometimes. But then the temperature doesn't drop at a constant rate, as is reflected in the blue line, but rather it is a slow process depending on the various other factors which affect the temperature drop which we will be discussing.
Doctor, there is a yellow line with arrows on this graph. What does that refer to?
That refers to a period after death wherein really the temperature has not dropped and this is called the plateau phase which is due to this postmortem--could be from heat production. And also really, you touch the skin surface, you will find the skin surface is cooler than the core body temperature, so it takes time for the inner body temperature to equilibrium with the skin temperature before the temperature starts dropping. So there are a number of reasons for this plateau phase, but one of them is alluded to, postmortem heat production.
And doctor, both from your review of the literature and in your own experience, is there any way to determine with reasonable medical certainty whether, A, there will be a plateau stage?
No, we cannot determine it because the plateau phase can be anywhere from a few minutes to a few hours and you cannot really estimate how long a plateau phase lasts in a particular person.
That was my second question. If you get out to a scene and you find a body and a probe shows a core temperature of 98.6--which you have indicated is the assumed normal core temperature at the time of death, correct?
If that is the situation, doctor, can you tell how long that temperature has been at that level on a plateau?
You cannot tell, but using general experience, plateau usually lasts anywhere from two to four hours depending, but you cannot be specific.
And doctor, what is the effect, if you cannot tell whether you have a plateau that is only a stage lasting a few minutes, as you indicated in your earlier answer, or goes two to four hours? What is that variability's impact on your ability to be precise with respect to the estimation of the range for time of death?
It is obvious that you cannot be precise from what we just discussed, because you have the same temperature for two to three hours, and how could you use it to estimate a time to be precise in those three hours? You can't.
Doctor, let me invite your attention back, if I could, to the 367-C and let me borrow the pointer for one moment. Assuming, hypothetically, you have a deceased individual and at the time the Coroner's investigator takes a core temperature reading the core temperature reading is 98.6 and assuming further that this person has experienced a temperature plateau, as you have already testified, is there any way medically that you would be able to tell where along this yellow line of the plateau stage you are actually at when you drop down to try to evaluate how long it has been since time of death?
And doctor, let me ask you now to assume in a different situation that you have an individual and the investigator takes a core temperature reading of 97.5 degrees Fahrenheit. First of all, doctor, is that within the range of your own experience as to variable difference from what is the expected, normal temperature of 98.6?
Assuming that you have such a temperature reading, doctor, is there any way you can tell, and assuming further that that is the person's actual core temperature at time of death--
--just assuming, is there any way you can tell, with respect to that individual, where along that person's plateau stage you are at?
Now, doctor, would it be correct to expect that if a person had a 97.5 degree normal core temperature at time of death, that that person's cooling curve will appear different than the cooling curve for a person who had a 98.6 degree Fahrenheit temperature that was the normal core temperature at time of death?
Now, doctor, assuming hypothetically you have a person who is found dead and the Coroner's investigator takes a core temperature reading of 97.5, is there any way that you as a forensic pathologist can tell whether that reading of 97.5 represents the person's actual normal core body temperature somewhere along the plateau or rather represents that the person had a higher temperature at time of death, but that the body has cooled from, for example, 98.6 to 97.5?
I can't tell the difference, and the problem is, if you assume the former, you are going to give a longer estimate as a range between the time--range for the time of death. And if it is really the latter, you are giving an incorrect answer.
Well, let's start with the assumption that it could be the person's normal core body temperature, actual core temperature at time of death, 97.5. If you obtained that temperature reading by your investigator, where along that person's cooling curve could you possibly be in terms of how long since the person died?
Now, doctor, if you assume it was the person's body temperature at the time the investigator took the core temperature, but that in fact the person's actual core temperature at time of death was this assumed normal 98.6, where would you be on that person's cooling curve?
You would be in this dotted line area, (indicating), so the time of death estimate makes it actually longer.
And doctor, is there any way medically that you can--let me withdraw the question and ask it preliminarily this way: Would it be accurate to say, doctor, that the distance in time of hours from this axis point where the vertical and horizontal axes met to the dotted line--
Mr. Kelberg, I don't think the jurors at the end of the box can see what you are pointing at there.
How about I bring the chart over here and perhaps with the elmo the jurors will be able to see it more clearly.
--it would work better if you stand on that side and I will try and stand on this side. Again, what I was starting to ask you is would it be accurate to say that in the situation where we are assuming the 97.5 degree reading actually reflects a cooling of the person's body from the reading at time of death of 98.6, that the time in hours since the time of death is going to be longer as represented from the distance where the two axes meet to where the dotted line is, then the time in terms of hours from time of death, if in fact this temperature that is obtained of 97.5, was the person's actual body temperature at time of death?
Doctor, medically, is there any way that you can tell whether you are--at this point where I'm pointing there is an intersection of the vertical dash line and the horizontal solid line--that is the time since death or you are at this point where the end of the green box applies?
And would it also be the case that you cannot tell whether you are at the point where the dotted vertical line meets the horizontal axis for time since death and any point between the origin of this graph where the vertical and horizontal solid lines meet and where the end of the green vertical box is?
If that temperature of 97.5 was truly the person's core temperature at time of death?
Doctor, how does this situation affect your ability to be precise with respect to the estimated range of time of death?
This is the reason I said from the beginning, it is an imprecise science and you can only give an estimated range, because if I use the assumption that the normal body temperature is 98.6, I'm going to estimate more number of hours since death. And if the true picture is actually that the temperature was 97.5 and I was seeing it at this point in the plateau phase, that means the hours since death was after shorter time frame than what I estimated when I made this original assumption which we all make usually that the body temperature is always 98.6.
Plus also you are to keep in mind--the plateau phase always in mind, which is a variable fact to which you cannot pinpoint saying that it is going to be two hours all the time or a few minutes all the time or three hours all the time.
Doctor, if we assume further, based on your own experience and Dr. Nokes' literature citation, that a normal core temperature can actually and above 98.6, and for example as you said, as much as a degree above, 99.6, how would that affect the difference, if the person's true normal core temperature at time of death was 99.6, the reading obtained by your investigator is 97.5, where is this horizontal line going to intersect with the cooling curve for the person with that higher normal temperature?
For this I will draw an imaginary cooling curve. The cooling curve will come like this, something like this, and it will be somewhere here, (indicating), which is in the same area, but you also have a longer estimated time interval estimated since death.
And doctor, would that create a wider window, if you will, within which you cannot with preciseness pinpoint the actual time of death?
Let me just take this one down so we can see if there is any additional information that we need to cover.
Doctor, these sub-parts of 3B, basically this is what you've already told us about?
All right. Your Honor, I have another board entitled "Algor mortis continued." May this be marked as 367-E?
We have covered--we have covered 1 and 2 and 3, I briefly covered it, but we could go over it.
All right. Before we go over that, I want to also ask you if in your review of the literature this was included in the material you read, considered and relied at least in part on in forming opinions about the inability to determine the actual normal core temperature from page 74 of Dr. Knight's forensic pathology book affecting the cooling curve initial body temperature. "This cannot be assumed to be 37 degrees centigrade and in fact is incapable of being measured in retrospect." "Retrospect" meaning going back to the time of death; is that correct?
Continuing on. "Not only is there a difference between the rectal, liver, brain axillary"--what is axillary, doctor?
It is an imprecise science and you can only give an estimated range, because if I use the assumption that the normal body temperature is 98.6, I'm going to estimate more number of hours since death.
The plateau phase can be anywhere from a few minutes to a few hours and you cannot really estimate how long a plateau phase lasts in a particular person.
I can't tell medically at which point it is.
You cannot tell that the person died at 1:00 A.M. using that formula, because of the variable factors.