Hearing transcripts

31 January 2008 - Morning session

1 Thursday, 31st January 2008
2 (9.30 am)
3 (Jury present)
4 LORD JUSTICE SCOTT BAKER: Mr Burnett, I have had a message
5 from the jury that the problem that prevented sitting
6 for them on Friday 29th February has now been overcome,
7 and I think, in the circumstances, unless anybody has
8 any very convincing reason for not, we should pencil in
9 that day as a day where we very possibly will be
10 sitting.
11 MR BURNETT: Yes. Thank you, sir.
12 LORD JUSTICE SCOTT BAKER: Mr Keen?
13 MR KEEN: It would appear that we have no French witness,
14 sir, and therefore, if I may, I will proceed with my
15 cross-examination of Professor Oliver.
16 LORD JUSTICE SCOTT BAKER: Yes. The position with the
17 French witness is we are still trying to get the
18 French -- and we ourselves are trying to trace her.
19 PROFESSOR JOHN OLIVER (continued)
20 Questions from MR KEEN
21 MR KEEN: Good morning, Professor Oliver.
22 A. Good morning.
23 Q. I wonder if I could go back very briefly to your
24 experience and background. You explained that you had
25 been professor of forensic toxicology at the University

1

1 of Glasgow. With regard to matters of research, do
2 I understand that one of your principal areas of
3 research is and was alcohol and driving?
4 A. That is correct.
5 Q. Is it the case that, up until your retirement, you were
6 engaged in the provision of forensic toxicology services
7 to at least the majority of police forces in Scotland?
8 A. To all of the police forces in Scotland, except Lothian
9 and Borders, which meant that I had the bulk of the
10 investigations in that area.
11 Q. You were asked, Professor, about the effects of alcohol,
12 and asked in particular about the effects of alcohol in
13 circumstances where someone did have a reading of what
14 has been referred to, in shorthand, as 175 millilitres
15 of alcohol in their blood.
16 A. Correct.
17 Q. I think you said, as was already set out in the joint
18 report of experts, that the typical individual would
19 show obvious signs of intoxication if, in fact, that
20 amount of alcohol was in their blood.
21 A. That is correct.
22 Q. You also indicated that a heavy drinker could have
23 a tolerance to alcohol -- is that right?
24 A. That is correct.
25 Q. -- and could therefore exhibit, at least, the appearance

2

1 of sobriety with a measure of alcohol approaching
2 175 millilitres?
3 A. That again is correct.
4 Q. Accordingly, when you were told that the late Henri Paul
5 did have 175 millilitres of alcohol in his blood --
6 LORD JUSTICE SCOTT BAKER: Milligrams.
7 MR KEEN: I apologise -- milligrams of alcohol in his blood,
8 and when you assumed that that was accurate, and when
9 you then saw the CCTV footage of his behaviour in
10 the Ritz in the period immediately preceding the journey
11 to the Alma Tunnel, was it appropriate to at least infer
12 at that time that he must have a high tolerance to
13 alcohol if indeed he did have 175 milligrams of alcohol
14 in his blood?
15 A. That is correct, yes.
16 Q. But presumably we have to take care to ensure that such
17 an argument does not become entirely circular. Can
18 I suggest, Professor, that you cannot say that his
19 behaviour supports the idea that he had 175 milligrams
20 of alcohol in his blood?
21 A. His behaviour -- and again could I quote
22 Professor Forrest -- when he parked his vehicle, it was
23 with elan, which I felt meant professionally,
24 competently, when he parked the vehicle.
25 The behaviour I saw of Henri Paul when he descended

3

1 the staircase in the Ritz Hotel was not of a man
2 struggling to control his actions. He came down there
3 steadily, with a steady gait, with no obvious attempt,
4 it would seem, to ensure that he was actually placing
5 his foot in front of the other.
6 You could draw an analogy from that with the field
7 impairment test carried out by the police, walking
8 a straight line, for example, where you actually have to
9 be very careful when you are placing your feet. There
10 was none of that. There was no hesitancy. That was
11 not, in my view, an intoxicated individual that came
12 down that staircase.
13 I then saw Henri Paul down tying his shoelace, where
14 he is taking down a very narrow base on his feet and
15 then tying his shoelace and then moving, tying the other
16 shoelace, a narrow base. Again this could be akin to
17 the field impairment test of a one-leg stand, where the
18 individual is actually trying to balance while they are
19 counting up to a certain ... He is doing a divided
20 attention test. I find that test very difficult myself,
21 and Henri Paul, there was no sign whatsoever of any
22 movement within that.
23 I have experience with field impairment testing.
24 I have worked with the Glasgow police officers, now
25 superintendents, Superintendent Paul Fleming(?) and

4

1 Superintendent David Stewart(?), when they were both
2 sergeants, who researched field impairment testing,
3 which is based on sobriety testing, and brought this
4 into the United Kingdom. I was also involved in the
5 initial trials in 1998 when five police forces tested
6 this. I was the analyst that was producing the results
7 and giving evidence in court on the basis of it.
8 I was also employed by the Department of Transport
9 to actually monitor the effectiveness of field
10 impairment testing in the United Kingdom, when it was in
11 the hands of the police forces, and produced a report
12 which been published by the Department of Transport. So
13 I am not talking from general experience. I have had
14 the results of all these tests and looked at it and
15 provided the analytical side to it.
16 LORD JUSTICE SCOTT BAKER: Professor, do you go back to the
17 pre-breathalyser days?
18 A. I just about go back to that, my Lord. The
19 pre-breathalyser days based on the walking a line and
20 picking up coins --
21 LORD JUSTICE SCOTT BAKER: What used to happen in those days
22 was that a defendant was charged with driving whilst
23 under the influence of drink --
24 A. Correct.
25 LORD JUSTICE SCOTT BAKER: -- and it all depended on the

5

1 examination of what happened in the police station.
2 A. Correct.
3 LORD JUSTICE SCOTT BAKER: There were things like walking
4 along a straight line, putting your finger on the end of
5 your nose with your eyes shut --
6 A. Indeed.
7 LORD JUSTICE SCOTT BAKER: -- and various other tests. Now,
8 from what you saw of Henri Paul in the video, there
9 wasn't a shred of evidence that would have even got
10 a prima facie case for driving whilst under the
11 influence of drink?
12 A. Precisely, and prosecution of road traffic offence, you
13 need this evidence. The toxicology -- all this going
14 into the newspaper is basically the toxicology results
15 being taken at face value. Toxicology provides
16 supporting evidence for what has been observed, not
17 corroborative. I support the diagnosis.
18 Do my results give a potential to support the
19 findings of the pathologist or of the police surgeon who
20 investigated this? I am not medically qualified.
21 Professor Forrest is. Professor Forrest observed
22 Henri Paul, and, in his statement here, could see no
23 signs of impairment. That is the absolute minimum that
24 the trained forensic medical examiner, on somebody who
25 refuses to be medically examined -- the medical examiner

6

1 can observe the driver and move forward from there.
2 MR KEEN: Can I come in, Professor, at that point? What we
3 seem to have is the primary evidence, the direct
4 evidence, which is recorded on CCTV, of Henri Paul's
5 behaviour in the hour or more before he left for the
6 Alma Tunnel.
7 A. Correct.
8 Q. That evidence is indicative in your mind, and as I think
9 you have indicated, in the mind of Professor Forrest, of
10 someone who is not under the influence of alcohol?
11 A. Correct.
12 Q. That has led, as I understand it, to a crossroads:
13 either he is not under the influence of alcohol, he has
14 not consumed 175 milligrams of alcohol, or the only
15 alternative explanation that was sought and grasped for
16 was he must have a enormously high tolerance to
17 alcohol --
18 A. Correct.
19 Q. -- and therefore he must be someone who is an alcoholic,
20 a chronic alcoholic, as it is termed, in order to have
21 developed that high tolerance to alcohol?
22 A. The only person that could have answered the chronic
23 alcoholic is Professor Lecomte. Had she carried out the
24 histology and looked at the individual's liver --
25 Q. But she did take the liver.

7

1 A. She took it, but she didn't look at it under histology,
2 and Professor Vanezis has complained to the extent that
3 it was not investigated to see if there is any fatty
4 infiltration of the liver which could not be seen by the
5 naked eye. This is effectively the histology, the role
6 of that pathologist, that was not carried through.
7 LORD JUSTICE SCOTT BAKER: I thought we had a definition of
8 an "alcoholic" earlier in the proceedings, and that it
9 wasn't necessarily somebody who just drank a lot, but it
10 was somebody who couldn't control their drinking and
11 couldn't resist the temptation.
12 MR KEEN: I am not sure if I recall the definition. It
13 might render us ... a very worrying definition, if I may
14 say so, sir, but I will say no more lest I incriminate
15 myself.
16 LORD JUSTICE SCOTT BAKER: It does not take us very much
17 further.
18 MR KEEN: Professor, can I just move on a little? We can
19 see where we come to the crossroads because people then
20 have two views. They either say, "Well, this man
21 clearly isn't under the influence of alcohol" or "He has
22 a very high tolerance to alcohol".
23 A. Correct.
24 Q. I just want to move on from there, because, as you say,
25 no histology was carried out, and to touch upon

8

1 a subject that I do not want to take at length, but it's
2 again the carboxyhaemoglobin and the readings in the
3 blood samples that were analysed.
4 If we just have up on the screen again [INQ0035090].
5 I will take it shortly because the jury have seen this
6 before. It's the joint experts' report at 19.3, where
7 it's recorded that:
8 "We are agreed that an individual with
9 a carboxyhaemoglobin concentration of 20.7 per cent
10 would be obviously unwell ..."
11 Not just "unwell", but "obviously unwell".
12 A. You would be unwell effectively -- now, there is
13 a publication -- I think perhaps Professor Johnston can
14 produce that publication to you -- whereby we looked,
15 and there was a survey taken of hospital admission areas
16 in an area of the United States, and they in fact treat
17 a level of 20 per cent of someone who needs treatment
18 and they will treat an individual with that.
19 With 20 per cent, you have diminished your
20 oxygen-carrying capacity of your blood, but not only
21 have you diminished this oxygen-carrying capacity,
22 carboxyhaemoglobin prevents the blood from giving it out
23 to the tissues as well. So, therefore, the
24 individual -- Professor Forrest -- headaches,
25 breathlessness, possibly struggling to actually breathe,

9

1 loss of strength in the limbs. One of the problems in
2 mines, for example -- it's an insidious gas -- you could
3 not, in fact, come out of the mine; you didn't have the
4 energy in your muscles by the time you realised
5 something was going wrong.
6 Q. You say that with over 20 per cent, someone would be
7 obviously unwell, and I think Professor Forrest
8 acknowledged that that would be a toxic level; you would
9 be regarded as suffering from carbon monoxide poisoning.
10 A. Quite.
11 Q. Then you go on to say that:
12 "If alcohol was also present in the blood, the
13 obvious adverse effects of carbon monoxide would be made
14 worse."
15 Again, we have heard Professor Forrest on this --
16 I don't need to take you to it at length -- but what we
17 have heard is evidence that the measurement of more than
18 20 per cent carboxyhaemoglobin in the blood samples is
19 inexplicable. It's been explained that it's
20 inexplicable in the sense that again Henri Paul was
21 clearly not exhibiting the symptoms of carbon monoxide
22 poisoning as he left the Ritz.
23 A. That would be correct, yes.
24 Q. Again, don't we have to avoid the argument becoming
25 circular? The result of the carboxyhaemoglobin analysis

10

1 of the blood samples being over 20 per cent is
2 inexplicable if you assume that the blood came from
3 Henri Paul?
4 A. It's inexplicable if the blood came from Henri Paul,
5 that is correct. We have tried -- Paget team has
6 tried -- to actually find a reason, a source of this
7 carbon monoxide in Henri Paul's sample. I think from
8 our initial work, individuals went and investigated to
9 see if there was a faulty water heater in the actual
10 flat of Henri Paul, to see if he could have been
11 inhaling carbon monoxide. We triggered that back in
12 1997. Nothing. No explanation was ever found for this
13 carbon monoxide.
14 Q. But if you take the blood samples that were produced on
15 31st August, analysis by toxicologists has disclosed, on
16 the basis of three separate tests using two independent
17 machines, carboxyhaemoglobin measurements of 20 to 21.4
18 or 21.7 per cent.
19 A. Say 21 per cent approximately, yes, correct.
20 Q. I don't revisit it, but in your evidence to Mr Hilliard,
21 you explained that given the consistency of those
22 results, you can't attribute that to some defect or
23 deficit in the measuring technique. It's not
24 an artefact. It is the result of, on the face of it,
25 perfectly legitimate analysis.

11

1 A. It is the result of the fluid that was analysed, be it
2 blood or be it fluid, that the percentage of haemoglobin
3 that was present in that specimen, 21 per cent, was in
4 the form of carboxyhaemoglobin.
5 Q. It got there somehow?
6 A. It got there somehow, and we have no viable reason for
7 this. Now --
8 Q. Can I come in there, Professor? Can we infer that it
9 wasn't in the blood when Henri Paul left the Ritz at
10 about midnight?
11 A. I think it's highly unlikely that it was in the blood
12 because again he was not showing symptoms.
13 Q. Can that, in a sense, bring us back to the chain of
14 custody? We have heard about the chain of custody and
15 I do not want to go over it in great detail, but as
16 Professor Forrest, I think, acknowledged, if instead of
17 simply making the assumption that the blood in the
18 sample on 31st August is entirely Henri Paul's, you
19 instead look for an explanation of the measurements of
20 carboxyhaemoglobin, the obvious explanation is that the
21 blood is not Henri Paul's or is not entirely
22 Henri Paul's?
23 A. That would be an explanation, that's correct.
24 Q. Now, if there was a mixing of samples -- and I think
25 Mr Hilliard used the term "spiked" -- but if there was

12

1 a mixing of blood samples, could you end up with
2 a sample in which you found evidence of drugs which
3 Henri Paul had taken, such as Prozac, and substances
4 such as carboxyhaemoglobin to which he had not been
5 exposed in the form of carbon monoxide?
6 A. Again, if samples are mixed between individuals, yes,
7 that is possible.
8 Q. Looking at the chain of custody, you can just assume
9 that the chain of custody was fine, and then express
10 opinions on the toxicology, but in those circumstances,
11 would the assumption you have made with regard to the
12 chain of custody necessarily, if implicitly, qualify any
13 opinion you actually expressed on the toxicology?
14 A. The assumption that I made on the chain of custody at
15 the beginning was that it would be in accordance with
16 the professionalism of the pathologists in this country
17 that I have dealt with. In this instance, particularly
18 when we went along to have a meeting with Lord Stevens
19 inquiry -- and I have to say I thank the Paget team for
20 the co-operation that they gave us; making documentation
21 readily available to us -- I was appalled. Probably
22 this year will be about the 40th year that I have been
23 giving evidence in courts from the beginning of my
24 career, and I have never, in my entire career, come
25 across such a disgraceful forensic trail that we have

13

1 for this so-called chain of custody.
2 We wouldn't be here if Professor Lecomte had clearly
3 identified her specimens when she took them from the
4 body of Henri Paul. If there had been a clear audit
5 trail, had there been three samples, had there been five
6 samples, had the samples been clearly labelled, were
7 they from the heart, if it's put in, had it been from
8 the haemothorax, we wouldn't be here arguing about this.
9 That has not happened. It's a disgraceful audit trail
10 for which there have been attempts to cover up using the
11 DNA side (inaudible).
12 Q. We will come onto the DNA in just a moment. Looking at
13 the chain of custody, then. Insofar as you, as a team
14 of pathologists and toxicologists, have, from time to
15 time and indeed regularly, expressed opinions on the
16 outcome of the toxicology, do I understand that those
17 opinions have always been implicitly qualified by
18 an assumption that the chain of custody was satisfactory
19 and valid?
20 A. Correct.
21 Q. Once it was clearly established that it was not, do
22 I understand that that necessarily impinges upon any
23 expression of opinion regarding the toxicology itself?
24 A. I think this is exactly the situation, and this is set
25 out in the joint report between ourselves,

14

1 Professor Forrest and Professor Vanezis and
2 Professor Johnston.
3 Q. I think Professor Forrest himself said, "It is a truism
4 that the interpretation of toxicological analysis
5 depends on the provenance of the samples".
6 A. Correct.
7 Q. I do not want to revisit all the problems with the chain
8 of custody, and we have been through the chain, if it
9 can be termed that, in regard to 31st August. We
10 touched yesterday upon the matter of the samples taken
11 on 4th September, and I think you were in court when
12 Professor Vanezis was being cross-examined by Mr Macleod
13 on that.
14 A. Correct.
15 Q. I just want to go to a matter that Mr Macleod raised,
16 which was the suggestion that the sample of left femoral
17 blood which Dr Pepin received might have been taken by
18 Dr Pepin, opened and divided into two samples and then
19 re-sealed with what he termed Dr Pepin's expert's seal.
20 We know that there is no mention of that in the papers
21 so I don't need you to go back to the documents in that
22 respect. I wonder if we could just look at the
23 Metropolitan Police interview with Dr Pepin at
24 [INQ0007345], just to see where this comes from, as it
25 were.

15

1 Now, I think if we look at paragraph 18, first of
2 all, if we can close in on that, Dr Pepin states:
3 "In this particular case, neither of these
4 procedures were necessary ..."
5 He explains why not. If then we go up to
6 paragraph 17, we can see what he means by "these
7 proceedings not being necessary" because what was
8 discussed with Dr Pepin was the matter that Mr Macleod
9 raised yesterday, and he says:
10 "We then discussed what happens when a court
11 requests two sets of analyses on one blood sample, for
12 example, alcohol and CDT. Dr Pepin explained that if
13 a sample had been opened, the police officer could, at
14 the request of the judge, if the laboratory asked, apply
15 a judicial seal after the analysis in the first
16 laboratory had been completed. An example of where this
17 might be done would be when dealing with a very large of
18 seizure of cocaine. There is also something called
19 an 'expert's seal' [which I think is what Mr Macleod
20 referred to] whereby Dr Pepin himself could place the
21 sample in a paper envelope and apply a sealing wax
22 seal."
23 So that's not resealing into a new bottle with
24 an inviolable cap.
25 A. Correct.

16

1 Q. "He provided a sample of such a seal which Mr Beer took
2 possession of. This could be done if he resealed a
3 sample before submitting it to another laboratory on the
4 order of a judge."
5 If we go back to the first sentence in paragraph 18,
6 do we see that what Dr Pepin explained to the
7 Metropolitan Police was that, in this case, neither of
8 these procedures were necessary?
9 A. Correct.
10 Q. If we go on to the next page for a moment to
11 paragraph 23, in order to confirm what it is Dr Pepin
12 claims he did send to Dr Dumestre Toulet, in the first
13 sentence does he say, "The sample sent to
14 Dr Dumestre Toulet was an unused, unopened, sealed
15 vial"?
16 A. Correct.
17 Q. Thank you. Now, if there was no chain of custody, in
18 other words if you started at the beginning and there
19 was no chain of custody through to the toxicology and
20 the toxicology results, do I understand that another way
21 of approaching the matter would be to take the samples
22 which had been the subject of toxicological examination
23 and subject them to DNA analysis in order to establish
24 a genetic link?
25 A. Yes, it is the way that I understand that the situation

17

1 was -- there was an attempt to resolve the situation, to
2 correct the initial errors.
3 Q. I think you were in court when Professor Shepherd gave
4 his evidence.
5 A. I was.
6 Q. Do you recall him explaining that in this case, for
7 reasons that we don't know, none of the samples that
8 were actually the subject of toxicological analysis were
9 made the subject of DNA analysis, and that, in fact, it
10 was other samples that were actually the subject of DNA?
11 A. I think you find that in document D876, document
12 number 2, page 113.
13 Q. You have confirmed it by reference to the documentation
14 as well.
15 A. I have indeed because, effectively, there is a comment
16 that the sample with the empty containers which were
17 used for toxicology analysis was not opened. It
18 remained sealed.
19 Q. So it was present and available?
20 A. It was present and available, and the residue, had it
21 been analysed by DNA, may or may not have tied the
22 specimens that were actually used for toxicology to the
23 body of Henri Paul.
24 Q. Do I understand that the empty containers that were
25 still available would still have been amenable to DNA

18

1 analysis?
2 A. As I understand, that is correct. Perhaps Professor
3 Jamieson is more an expert on DNA than I am, but I do
4 know that you need very small residual samples. You
5 don't get a pristine clean glass vial when you have
6 taken the blood out of it. There is always a blood
7 stain left in the vial. That, to my knowledge, would be
8 more than sufficient for a DNA analysis.
9 Q. I think, in fact, Professor Shepherd acknowledged that
10 there was something -- I do not want to misquote him --
11 that there didn't seem to be an ascertainable point in
12 testing a sample for alcohol and then testing
13 a different sample for DNA. It's perhaps a statement of
14 the obvious.
15 A. I think that is correct, effectively.
16 Q. What was suggested by Mr Macleod was that if you took
17 one sample and tested it for alcohol and then tested
18 another sample for DNA, that would be all right if you
19 assumed the two samples came from the same batch.
20 A. Yes, I think effectively, had these samples been
21 labelled at the time of the autopsy to positively
22 identify them at that point, rather than leaving them
23 around for a later and wrong identification of source,
24 that throws into confusion -- I have no doubts that the
25 samples referred to by Mr Macleod are related to

19

1 Henri Paul, the ones that were tested.
2 Q. That is the DNA test of samples?
3 A. The DNA test. But with such a disgraceful chain of
4 evidence in this particular case, I would not -- they
5 are probably Henri Paul's, but I would not like to make
6 the legal jump between them and say they are actually
7 Henri Paul's.
8 Q. In the absence of any chain of custody?
9 A. The absence of the proper chain of custody.
10 Q. Can I ask you this: there was a DNA analysis, was there
11 not, of the section of liver taken by Professor Lecomte?
12 A. There was, but that was not the section of liver that
13 was used for the analysis on the "body of Henri Paul".
14 Q. I see. So even that wasn't the subject of the DNA
15 analysis?
16 A. That was a different sample. The sample of liver, as
17 I understand, that was analysed by Dr Pepin was taken by
18 Professor Lecomte, not by Dr Campana.
19 Q. Right. Coming to that section of liver, is it in fact
20 the case that no Tiapride was identified in that section
21 of liver?
22 A. Oh gosh, this is where again we have what I would call
23 problems with the presentation of analytical results.
24 We have a list of analytical results in one of the
25 appendices which makes it look as though there is

20

1 a consistency of Tiapride present in practically all the
2 samples, but we have heard from Professor Forrest -- and
3 I myself have questioned this -- what did this mean,
4 "less than limit of detection"?
5 If you then look at all the drugs they looked at, if
6 they are inferring Tiapride is present in these samples,
7 then what about all the other drugs that were tested
8 for? By the same analogy, you can infer that they were
9 all present in these samples. They were negative. They
10 were not found in these particular specimens, and there
11 is a difference between limit of quantitation and limit
12 of detection. Limit of detection is what you can safely
13 detect with your instrumentation.
14 Q. To see that illustrated, could we have [INQ0004471]?
15 A. Indeed.
16 Q. I think what you are referring to is the fact that there
17 is this notation less than 0.005, which
18 Professor Forrest himself, in his examination-in-chief,
19 said was sometimes open to misinterpretation, but was
20 generally to be construed as indicating that the drug
21 was not present.
22 A. Correct.
23 Q. Do we see that that is the case in respect of the sample
24 of liver, kidney, spleen, lungs and pancreas?
25 A. Again that is correct.

21

1 Q. Indeed it is also the case in the stomach contents?
2 A. It's also the case in the stomach -- yes, indeed,
3 correct.
4 Q. Thank you. We have been told that in the sample of
5 liver that was subject to DNA analysis, there was
6 evidence of Prozac or at least the elements of Prozac,
7 Fluoxetine and Norfluoxetine.
8 A. Correct.
9 Q. I think it has already been indicated that that is, in
10 fact, an extremely common drug.
11 A. I think you will obtain statistics from
12 Professor Johnston just showing you how common Prozac is
13 as a drug and prescription, effectively to the extent of
14 even finding it in ground water because of the constant
15 passing through of individuals. It is quite
16 a commonly -- a very commonly prescribed drug.
17 Q. Finally, Professor, there was reference by
18 Professor Forrest and then by yourself to a paper by
19 Winek.
20 A. Yes.
21 Q. I think Professor Forrest said "Winek of Luxembourg",
22 but it's Winek of Pennsylvania?
23 A. He was thinking of Professor Wennig of Luxembourg, an
24 ex-president of the International Association of
25 Forensic Toxicologists.

22

1 Q. If we have the right Winek and the right paper --
2 A. We have the right Winek, yes, indeed.
3 Q. This was just about carboxyhaemoglobin readings.
4 I think you made the point that, if you actually read
5 the paper, what it indicates is that while there can be
6 a 20 to 30 per cent variation in carboxyhaemoglobin
7 readings, they are lower, not higher.
8 A. That is effectively the findings of this particular
9 paper and this is on decomposition. What we are being
10 asked to accept, to justify the high level, is that
11 there has been interference with breakdown products,
12 causing interference with the measurement of
13 carboxyhaemoglobin. That is on one side.
14 On the other side, we are looking at the alcohol
15 level in the blood specimen, wherever it came from, and
16 saying "There is no interference in that". I am sorry,
17 you either have interference or you don't have
18 interference. I see nothing in this that would actually
19 increase the level of carboxyhaemoglobin.
20 Q. There is a reference, as I understand it, in that paper,
21 Professor, to putrefaction. Is that what you mean by
22 the "interference"?
23 A. I mean the putrefaction.
24 Q. The effects of putrefaction.
25 A. Indeed.

23

1 Q. Do I understand that if there was putrefaction -- and
2 I think in this context Winek's analysis was dealing
3 with blood that was at least 30 days old.
4 A. 30 to 150 days old.
5 Q. So if there was putrefaction, are you saying that would
6 not only affect the carboxyhaemoglobin, it would also
7 affect the alcohol?
8 A. Yes. If the blood sample had putrified, you have the
9 potential to increase alcohol levels. It is
10 a side-product of breakdown. Again, it would have shown
11 up in the traces of Dr Pepin. I have looked at these
12 traces, and -- not so much Ricordel's, he has used a
13 packed column GC(?), but Pepin has a better definition,
14 and there is no evidence of volatiles in that that could
15 come from putrefaction.
16 Q. If we can confirm that. If we go to [INQ0032033].
17 Again these are questions that I believe were being
18 posed by the Metropolitan Police to Dr Pepin.
19 A. Exactly.
20 Q. If we go to the foot of the page, at question 48(b),
21 does he say:
22 "The organoleptic characteristics of the blood ..."
23 Then he says:
24 "Colour, texture and smell indicated non-putrified
25 blood in a very good state of conservation."

24

1 A. Indeed.
2 Q. Is that distinct from the form of blood samples that
3 were the subject of reporting by Winek in the paper that
4 Professor Forrest brought to the attention of the
5 inquiry?
6 A. Exactly. They were putrifying and breaking down.
7 MR KEEN: Thank you, Professor. No further questions.
8 LORD JUSTICE SCOTT BAKER: Thank you.
9 MR CROXFORD: No, thank you, sir.
10 LORD JUSTICE SCOTT BAKER: Mr Macleod?
11 Questions from MR MACLEOD
12 MR MACLEOD: Professor, as you know, I ask questions on
13 behalf of the Commissioner of Police for the Metropolis.
14 A. Indeed.
15 Q. My name is Duncan Macleod. Professor, what I would like
16 to concentrate on, please, is the way in which the
17 toxicological analyses conducted in this particular case
18 can help the jury in tackling the critical question
19 which is relevant to this topic, namely whether
20 Henri Paul's ability to drive on the occasion was likely
21 to have been impaired by the consumption of alcohol.
22 A. Yes.
23 Q. I think you would agree, would you not, that this case
24 is not a question of whether Henri Paul had been
25 drinking, it's a question of how much he had been

25

1 drinking that the jury have to grapple with?
2 A. Provided you can directly link the blood specimen to
3 Henri Paul. That is where the evidence in this case,
4 I feel, is suspect because of incorrect labelling at the
5 time.
6 LORD JUSTICE SCOTT BAKER: I think you misunderstood
7 Mr Macleod's question.
8 A. Yes.
9 MR MACLEOD: I can clarify it in this way: the obvious
10 starting point, Professor, is not the haemoglobin
11 reading or the audit trail; the obvious starting point
12 is that it is a known fact that Henri Paul had consumed
13 10 centilitres of Ricard, the equivalent of more than
14 four measures of whiskey, in two 20-minute periods,
15 ending at 10 past 11 in the evening, approximately
16 an hour before he left the Ritz to drive the car.
17 A. I have seen the bar receipts for that and I have no
18 argument with that whatsoever.
19 Q. That is the obvious starting point, isn't it?
20 A. That is the starting point, that's correct.
21 Q. It's in the context of considerations like that that all
22 these toxicological data need to be interpreted. That's
23 just one fact.
24 A. That's one fact.
25 Q. But it has to be looked at in the round.

26

1 A. We have a specific piece of information on the
2 consumption of two 50-millilitre measures of Ricards
3 which, according to Professor Forrest, was 45 per cent
4 alcohol by volume. According to the owners of the Ritz,
5 when I made the inquiry, it was 40 per cent, but I am
6 willing to accept Professor Forrest's calculation of
7 45 per cent. It doesn't make a great deal of
8 difference.
9 Q. I am sorry, Professor, but that is public knowledge, the
10 percentage proof of Ricard. You can look on the
11 website.
12 A. Yes, but at the time I asked specifically --
13 Q. It's not an issue in this case, is it?
14 A. It's not an issue. I am happy to --
15 Q. Ricard is stronger than whiskey by a factor of about
16 12.5 per cent.
17 A. No, 5 per cent.
18 Q. 5 per cent in alcohol, but in terms that it is stronger
19 than whiskey --
20 A. Yes, it is.
21 Q. -- it is 12.5 per cent.
22 A. Yes, it sounds worse. Indeed.
23 Q. Now, we know that Henri Paul was observed having drunk
24 that quantity of alcohol in that short space of time
25 both by witnesses and by CCTV cameras, and he apparently

27

1 observed no signs of intoxication at all?
2 A. Correct. I wouldn't expect him to.
3 Q. We also know, or do you know -- have you read the
4 witness statement provided by his closest friend,
5 Claude Garrec, concerning the tolerance of alcohol which
6 Henri Paul apparently displayed?
7 A. I may have done. I am not -- I can't quite remember
8 that.
9 Q. Mr Garrec is giving evidence this afternoon.
10 A. Fine.
11 Q. He was the best man at his wedding, he has known this
12 man since high school, he was his closest friend, and he
13 gives examples such as: Henri Paul could drink four
14 Ricards and a few beers and show no signs of
15 intoxication either in the way he looked or the way he
16 stood or in any way at all.
17 A. In other words, he had a high tolerance to alcohol.
18 Q. Exactly. That's another fact which should be taken into
19 account when looking at the toxicological analysis and
20 in the assistance that can be given to the jury in the
21 interpretation?
22 A. I am willing to interpret the two Ricards.
23 Q. Mr Claude Garrec also indicated, on the same topic of
24 alcohol tolerance, that Henri Paul would drink during
25 festivals or fetes, and he would, on occasion, drive

28

1 home having consumed considerable amounts of alcohol,
2 but that on those occasions, he displayed no apparent
3 inability to control the car and he always got home
4 without mishap.
5 A. It is a high level of tolerance.
6 Q. Now, in your report, when you first dealt with this
7 issue with other experts -- and it's the report of
8 5th December 1997, paragraphs 20 and 21 -- you said this
9 in relation to Henri Paul and his tolerance to alcohol,
10 and this was a view shared by Professor Krompecher --
11 A. Correct.
12 Q. -- Professor Mangin --
13 A. Correct.
14 Q. -- yourself and Professor Vanezis.
15 A. Correct.
16 Q. "Looking at the overall picture [which is the word that
17 is used], it may be fairly observed that M Paul had
18 an alcohol problem and he drank high levels of alcohol
19 regularly."
20 A. Yes.
21 Q. Then, in brackets, there is a dossier reference to
22 a document in the French investigation. It's
23 a typographical error, but it's clearly a reference to
24 document D1519.
25 A. Mm.

29

1 Q. That is obviously the source of the information that led
2 to that conclusion.
3 A. Indeed.
4 Q. That conclusion was the CDT report of
5 Dr Dumestre Toulet, wasn't it?
6 A. It was, but I am not particularly qualified to comment
7 on that because I am not a clinical biochemist. I think
8 Professor Johnston will comment on the CDT for you.
9 Q. But on that occasion, a professor of forensic medicine,
10 a professor of forensic medicine specialising in
11 toxicology, a toxicologist and Professor Vanezis did not
12 regard, at that time, the methodology or the reliability
13 of a CDT test as in any way questionable.
14 A. I don't think -- well, I wasn't there. I wasn't
15 involved in CDT testing and never have been, so
16 I couldn't really comment on that, but they --
17 Q. But none of those four experts questioned --
18 A. The didn't question the CDT --
19 Q. -- the validity of a CDT test --
20 A. Not at all.
21 Q. -- to indicate that someone may have been drinking high
22 levels of alcohol for a week preceding death.
23 A. Exactly, yes.
24 Q. The evidence goes on in this way:
25 "He may have built up a tolerance of the levels of

30

1 alcohol found. There is no doubt that the average man's
2 faculties would have been markedly impaired. A regular
3 heavy drinker, like M Paul, is likely to have been
4 impaired less. M Paul's faculties would have suffered
5 some impairment, but what may be debated is the degree
6 of impairment."
7 And that's fair?
8 A. This is a fair assessment.
9 Q. At 21:
10 "It is noted that the video from the Ritz Hotel
11 showed M Paul seemingly walking normally. This is not
12 surprising as a habitual heavy drinker would have built
13 up a degree of tolerance."
14 A. Correct.
15 Q. "A normal man would have been noticeably affected.
16 Further, for M Paul to be able to behave apparently
17 normally does not mean that he was fit to drive,
18 especially if something extraordinary happened whilst he
19 was driving. In such circumstances he may have had
20 difficulties in reacting appropriately."
21 Now, the first thing I want to ask you about this,
22 Professor, is: that opinion, which is talking about the
23 average person who has a high tolerance to alcohol,
24 wasn't an opinion that was qualified in any way by
25 subsequent toxicological results or papers or anything

31

1 else that you have received since?
2 A. It was our opinion at the time. There are subsequent
3 papers. I think I mentioned one when I was in here
4 yesterday.
5 Q. Indeed. I was going to ask you about that in due
6 course.
7 A. Indeed.
8 Q. If I can shortcircuit, I hope fairly -- and correct me
9 if I am wrong -- but it is a truism, is it not, that for
10 a person who has developed a high degree of tolerance to
11 alcohol, that person may not develop or show signs of
12 intoxication to the naked eye or to the film at all?
13 A. Correct.
14 Q. When you look at a CCTV footage of a person, that can be
15 a very unreliable guide as to the amount of alcohol in
16 that person's system?
17 A. Exactly, yes.
18 Q. Dealing with the audit trail, I think there is a large
19 measure of agreement between you and Professor Forrest
20 concerning the inconsistencies and deficiencies of the
21 audit trail; is that right?
22 A. That is correct, yes.
23 Q. But nonetheless, taking into account all of those
24 matters, Professor Forrest has indicated, looking at the
25 matter in the round, at the balance of the evidence,

32

1 that he can be comfortably satisfied that Henri Paul's
2 blood/alcohol level was likely to be in the region of
3 twice the legal limit when he died.
4 A. That is the evidence in the round, and if I think -- if
5 I recollect correctly, you were linking the DNA testing
6 on specimens to the specimens that were used for the
7 analytical purposes.
8 Q. Excuse me, Professor, it wasn't me. This is in
9 Professor Forrest's reports to Operation Paget --
10 A. Indeed, yes.
11 Q. -- long before I was involved in this case.
12 A. Yes.
13 Q. Now you have heard also Professor Vanezis' evidence
14 yesterday --
15 A. I have.
16 Q. -- on that same topic.
17 A. I have.
18 Q. Now, if you were to look at all the evidence in the
19 round, would you not agree that the samples that were
20 tested were probably those of Henri Paul?
21 A. I would agree they were probably those of Henri Paul,
22 but because of the, again, last-minute introduction of
23 new specimens -- the photographs by
24 Professor Shepherd -- where did these samples come from?
25 Were they indeed samples of blood from Henri Paul? We

33

1 do not know.
2 The only person that can clear this up to you,
3 Mr Macleod, is Professor Lecomte, who knows exactly what
4 specimens she took and what happened. They were not
5 properly labelled. We have no idea where these blood
6 samples came from. I am sorry, I would have difficulty
7 ascribing the analytical samples if I was to stand up
8 and give evidence in a criminal court because of the --
9 LORD JUSTICE SCOTT BAKER: I can understand that. Your
10 position is probably they were his samples --
11 A. They were probably --
12 LORD JUSTICE SCOTT BAKER: -- but you certainly couldn't say
13 that they definitely were?
14 A. I am sorry, I have never come across such a disgraceful
15 audit trail in my entire career.
16 LORD JUSTICE SCOTT BAKER: One of the problems here is that
17 you look at the audit trail and it's really an affront
18 to any scientist when you see it.
19 A. Absolutely, my Lord.
20 LORD JUSTICE SCOTT BAKER: If it had happened in one of your
21 laboratories, I dare say the perpetrator would not have
22 been there very much longer.
23 A. I even refuse to change misspellings of names,
24 and I would put it into my report, "Received labelled
25 as..." with the misspelt names. That is -- every part

34

1 of that audit trail is consistent. In this instance,
2 I am sorry, it is not.
3 MR MACLEOD: Professor, what I am trying to do is lift our
4 eyes up from the tarmac of the audit trail and look at
5 it in a rather broader evidential context, because the
6 jury are not sitting in a criminal trial. They have to
7 weigh the question of whether Henri Paul's blood/alcohol
8 level was such that it may have impaired his driving
9 ability on the night in question.
10 A. If the sample has come from Henri Paul -- if the sample
11 has come from Henri Paul -- then it is probable that it
12 could have impaired his driving -- not necessarily,
13 because of tolerance -- but it is probable it could
14 have.
15 My difficulty is the concrete link between sample
16 and Henri Paul, and again you are giving Henri Paul, if
17 it's wrong, a raw deal. If it's correct, then he is at
18 fault.
19 LORD JUSTICE SCOTT BAKER: You are not suggesting that
20 somebody with a blood/alcohol of twice the legal limit
21 wasn't impaired, are you?
22 A. I am not suggesting it. It is coming from the
23 literature, my Lord. I would like to put this in,
24 "Tolerance to high blood/alcohol concentrations", where
25 effectively up to a level of 200 --

35

1 LORD JUSTICE SCOTT BAKER: It's your evidence we want.
2 A. It is. I am just evaluating the findings in the
3 literature.
4 MR MACLEOD: I think the point that you were making
5 yesterday, Professor, is that research shows quite
6 conclusively that persons with high levels of tolerance
7 to alcohol can have a reading as high as 200 and display
8 no obvious manifestations.
9 A. Quite.
10 LORD JUSTICE SCOTT BAKER: But it's a different matter when
11 they may be faced with an emergency on the road.
12 A. It's probably a different matter, yes.
13 LORD JUSTICE SCOTT BAKER: That's why there is a limit that
14 there is.
15 A. There is a limit, and again that is based on the
16 likelihood of being involved in an accident at
17 particular levels.
18 MR MACLEOD: I do not want to go through the same ground
19 again with you, Professor, but if we just look at the
20 samples that were taken on 31st August, and accepting
21 all the problems with the audit trail that you have
22 identified, Professor Forrest has identified and others
23 have identified, taking that as read, first of all we
24 know, don't we, that there was a batch of samples -- and
25 I am dealing first taken on 31st August 1997 --

36

1 A. Correct.
2 Q. -- a batch of samples, blood and other tissue samples,
3 taken from Henri Paul on that day?
4 A. Yes, I agree.
5 Q. Just so that there is no doubt about this, all the
6 people concerned in this case have seen the post-mortem
7 photographs. The jury for obvious reasons won't do.
8 But there has never been an issue, so far as you are
9 concerned, that the body shown in the photographs was
10 that of Henri Paul?
11 A. I have never had an issue with that. I have taken it on
12 trust that it was the body of Henri Paul.
13 Q. I hope we do not have to go into that any further.
14 A. Indeed.
15 Q. Now, in relation to the samples taken on 31st August,
16 recorded to be that of Henri Paul, taken from the naked
17 body of Henri Paul marked "2147", one sample of blood
18 was sent to Professor Ricordel, and he analysed it and
19 recorded a reading of 1.87.
20 A. Correct.
21 Q. Another two samples were tested by Dr Pepin.
22 A. The other part of that sample -- I think, if you are
23 reading the literature, Dr Pepin divided one of the
24 specimens and it went to Dr Ricordel.
25 Q. And achieved a consistent result within a margin of

37

1 tolerance?
2 A. Correct, yes.
3 Q. Both of those indicating that the level of alcohol in
4 the blood that was taken from that corpse was around
5 twice the legal limit in the UK?
6 A. If we accept that this audit trail is sufficient, yes,
7 that is correct.
8 Q. We also know that when the drugs were tested for, from
9 the blood that was taken from the same batch of samples,
10 from that same corpse, on 31st August, they showed, one,
11 evidence of recent consumption of alcohol.
12 A. Correct.
13 Q. Two, that they contained a combination of drugs,
14 prescribed drugs, indicative of a person who had been
15 taking Prozac and Tiapride.
16 A. Indicative, yes, that's correct; a very low level of
17 Tiapride, but indicative, yes.
18 Q. We know that Henri Paul was prescribed and was taking
19 that combination of drugs.
20 A. Among others, yes.
21 Q. We also know that a liver sample recorded as taken in
22 that same batch --
23 A. Correct.
24 Q. -- from the same corpse was sent for DNA analysis and
25 came up with an identical matching profile to

38

1 Henri Paul.
2 A. From the same batch?
3 Q. From the same batch of samples that were taken on
4 31st August.
5 A. I thought the sample that was sent for DNA analysis came
6 from Dr Campana.
7 Q. Well, I believe that this is the sample of liver that
8 was taken on 31st August. I think it's right, isn't it,
9 that Professor Forrest was the person who alerted you
10 and the remainder of the experts who were instructed by
11 Mohamed Al Fayed that there was in fact a liver DNA
12 sample undertaken?
13 A. I knew there was a liver DNA sample undertaken.
14 Q. But you didn't know until Professor Forrest -- I have
15 seen the e-mail exchange and that was as late as
16 July 2007.
17 A. Correct.
18 Q. That point hadn't previously been brought to any of the
19 experts' attention --
20 A. Again, that's correct.
21 Q. -- instructed by Mr Al Fayed. That liver sample
22 provided an identical DNA profile matching that corpse's
23 liver to the DNA profile of Henri Paul's mother?
24 A. At this point I have to take your word on this, that
25 this liver sample was taken by Professor Lecomte at that

39

1 stage. If it is correct, then that is correct.
2 Q. Furthermore, a blood sample also collected on
3 31st August 1997 by Professor Lecomte, from the same
4 batch of tissue and blood samples, was sent by the
5 French investigators for DNA analysis and it produced
6 an identical matching profile to that of Henri Paul's
7 mother.
8 A. Yes, that is correct.
9 Q. Now, I am going to ask you the same question, Professor,
10 that I asked Professor Vanezis: when you look fairly and
11 objectively on that balance of evidence, that is
12 compelling evidence to suggest that the samples which
13 were taken from the corpse by Professor Lecomte on
14 31st August were samples that were taken from the body
15 of Henri Paul.
16 A. The samples that were proved to actually be taken from
17 Henri Paul were the ones taken for DNA. You have to
18 then hope that the labelling is consistent between the
19 blood samples and has been honestly carried out. If
20 that is the case, then yes, I would accept it.
21 Q. Professor, we also know, don't we, that in relation to
22 other samples taken on that same occasion, 31st August,
23 such as urine --
24 A. Correct.
25 Q. -- that also showed recent consumption of alcohol?

40

1 A. Yes, it supports the alcohol figure.
2 Q. It also showed Fluoxetine, Norfluoxetine and Tiapride?
3 A. Yes, it did.
4 Q. The self-same combination of prescribed drugs that were
5 given to Henri Paul?
6 A. Correct.
7 Q. We know that the spinal cord sample taken from the same
8 corpse on 9th September, subjected to toxicological
9 analysis, had Fluoxetine, Norfluoxetine and Tiapride?
10 A. This was taken on 9th September?
11 Q. Yes.
12 A. Yes.
13 Q. But from the same 2147 corpse.
14 A. Right.
15 Q. And likewise the hair sample gave identical
16 toxicological results?
17 A. Plus Albendazole.
18 Q. Yes, well, I can go to Albendazole if we think it's
19 really important. But what I am trying to establish
20 with you, Professor, is that combination of findings is
21 strongly indicative that that batch of samples taken
22 from Professor Lecomte from a corpse on 31st August
23 during the only autopsy conducted in the IML on that day
24 came from Henri Paul.
25 A. It's strongly indicative. Yes, I would use your words,

41

1 "strongly indicative".
2 Q. Because if it were otherwise and there was a confusion
3 in blood with another corpse, a sample taken from
4 another corpse, leaving aside that there was only one
5 autopsy that day at IML in a discrete suite, it would
6 have to be samples coming from a person who had been
7 drinking alcohol recently before they died.
8 A. And exposed to carbon monoxide.
9 Q. You would have the self-same problem -- but bear with
10 me -- and who coincidentally had in their bloodstream
11 the exact self-same drugs in all the samples tested that
12 were prescribed to Henri Paul?
13 A. Not all the samples, because we don't have Tiapride in
14 all the samples.
15 Q. Well, we have no detectable limit on some samples, but
16 the ones I have just referred to, Professor.
17 A. Correct.
18 Q. I think you follow my argument.
19 A. I do, indeed.
20 Q. It's right, isn't it, that if a person is prescribed
21 Prozac, they are told by their doctor and by the warning
22 on the packet not to drink alcohol?
23 A. As I understand, that is correct.
24 Q. It's counter-indicated, the consumption of alcohol with
25 the consumption of Prozac.

42

1 A. It is contra-indicated, that's correct.
2 Q. So although it's a commonly prescribed drug, it would
3 also have to be a person who is ignoring that medical
4 advice --
5 A. Correct.
6 Q. -- as we know Henri Paul did.
7 A. Correct. If specimens came from him, correct.
8 Q. If blood was mixed between samples, or spiked, then
9 there would be a detection in the DNA profile of that,
10 wouldn't there?
11 A. I am not a DNA expert, but the DNA profile of Henri Paul
12 would still come through.
13 Q. But it wouldn't be an identical match, would it?
14 A. Again, I am not a DNA expert.
15 Q. All right. We will be hearing from DNA experts in due
16 course, but you are a professor of forensic medicine.
17 A. No, I am a professor of forensic toxicology.
18 Q. My mistake. I stand corrected. But from your
19 understanding, the position would be that there would be
20 no direct correlation -- if there were two people's
21 blood mixed together, the DNA profile would be
22 contaminated, wouldn't it?
23 A. The DNA profile would be contaminated, but there would
24 still be a possibility that it came from Henri Paul.
25 Q. That just deals with the samples taken on 31st August.

43

1 A. Correct.
2 Q. Can I deal briefly with the samples taken on
3 4th September by Dr Campana? First of all, dealing with
4 the pre-analytical issues, we know that there is no
5 problem concerning the methodology of the audit trail in
6 this case when the samples were taken.
7 A. I had no problem with that until I read the description
8 from the Metropolitan Police that Dr Pepin stated that
9 only one femoral blood sample was taken that day.
10 Q. Can I deal with that in a moment?
11 A. Sure.
12 Q. Before I deal with that, what I didn't deal with you in
13 relation to 31st August -- if we have established that
14 all these samples, the batch of samples, many of which
15 have been subjected to the corresponding tests and
16 consistent tests which I have referred to -- there was
17 also the question of the vitreous humour, wasn't there?
18 A. Correct.
19 Q. That's a very reliable source for alcohol/blood
20 analysis, isn't it?
21 A. It is a source for alcohol/blood analysis which
22 I receive in times of compromise, when the samples
23 were -- when the body was putrified. It's very often
24 this area here, the eye, because it's protected, you can
25 only get that to indicate the presence of alcohol. Now,

44

1 we have a situation -- there is an equilibration between
2 the alcohol in the eye fluid and in the blood. Again
3 I suspect it is like urine in that it probably tracks
4 the blood/alcohol levels --
5 Q. Professor, sorry to interrupt. I think this is
6 virtually common ground.
7 A. It is.
8 Q. Vitreous humour is one of the ideal sources to test true
9 alcohol level in a post-mortem analysis.
10 A. The difficulty is -- no, I am sorry. I think there is
11 a variation, and I think again we have had papers
12 showing --
13 Q. Professor Vanezis --
14 MR KEEN: I wonder if the witness could be allowed to answer
15 the question, sir.
16 MR MACLEOD: I am conscious of the time.
17 LORD JUSTICE SCOTT BAKER: So am I. We have a lot of
18 evidence today and we are really going over territory
19 that really I thought, until now, was common ground.
20 MR KEEN: If my learned friend asks the question, the
21 witness is entitled to answer it. If he wants to save
22 time, he doesn't have to ask the question.
23 LORD JUSTICE SCOTT BAKER: That's the point.
24 A. In an ideal situation, yes, it would be a direct
25 reflection, but again, because it is historic, you are

45

1 going to have lower in the absorptive phase and you are
2 going to probably have a higher level in the declining
3 phase.
4 MR MACLEOD: We are not worried about whether it's
5 10 per cent, 20 per cent higher of lower. We are just
6 wondering whether this is at a level that is going to
7 impair driving.
8 A. It's indicative.
9 Q. We know that two samples taken from that batch on
10 31st August were DNA-matched to Henri Paul. A sample of
11 vitreous humour taken at the same time from the same
12 corpse was the vitreous humour, and it gave a reading of
13 1.74 in alcohol.
14 A. Exactly.
15 Q. If that came from Henri Paul, that is probably as near
16 conclusive evidence as to the level of alcohol at the
17 time of death as you can get?
18 A. It is a remarkable coincidence, and it would indicate
19 a support, and I think I used that as a support in my
20 initial reports.
21 Q. So far as 4th September is concerned, the samples taken
22 there, under supervision of the judge from the femoral
23 vein --
24 A. Correct.
25 Q. -- two experts present, and taken by Dr Pepin

46

1 immediately for forensic analysis at ToxLab that same
2 day, those alcohol readings came out at 1.80, reported
3 at 1.75.
4 A. Recalculated to 175, not "reported".
5 Q. It was from the left femoral?
6 A. That's correct.
7 Q. And the drugs revealed were the same drugs, Fluoxetine,
8 Norfluoxetine, Tiapride?
9 A. Indeed.
10 Q. Again consistent with 31st August?
11 A. With a better audit trail, yes, or apparent audit trail.
12 Q. Later the blood that was taken on that very self-same
13 occasion by Operation Paget for DNA analysis, sent to
14 LGC Forensic for DNA profiling, again gave an identical
15 match.
16 A. Provided that audit trail can be substantiated, yes,
17 I think that's been a matter of debate.
18 Q. I will put the same question to you as I put to
19 Professor Vanezis: that combination of circumstances, as
20 a matter of common sense, if nothing else, is powerful
21 evidence that those blood samples were taken from the
22 body of Henri Paul?
23 A. It is indicative that it is from the body of Henri Paul.
24 It is evidence, yes, I would agree with you.
25 Q. If I can deal, please, with -- I should put this point

47

1 to you. We have heard about the issue of the seal. The
2 remarks made by Dr Pepin when he was interviewed by
3 Professor Forrest for Operation Paget were, of course,
4 made several years after the event, weren't they?
5 A. I assume so, yes, correct. Yes, they would be.
6 Q. So Dr Pepin was relying on his recollection as to what
7 would have occurred seven years prior?
8 A. He should have been relying on his laboratory notes.
9 Q. I know, but there were no notes to which he could refer.
10 A. This is his problem.
11 Q. That is the problem, but as I have said, we are looking
12 above the audit trail. But that is the fact, Dr Pepin
13 was trying to recall what occurred some seven years
14 prior?
15 A. Yes.
16 Q. Can I just deal with carboxyhaemoglobin as briefly as
17 I can, please? First of all, in relation to studies in
18 relation to carboxyhaemoglobin, where they refer to the
19 typical population of smokers and the level of
20 carboxyhaemoglobin found in that typical sample across
21 the range --
22 A. Yes.
23 Q. -- it's right, isn't it, that most smokers in that range
24 would be smoking first of all cigarettes?
25 A. Correct.

48

1 Q. And they would be filtered?
2 A. Yes.
3 Q. We know, from M Claude Garrec, that so far as Henri Paul
4 was concerned, he smoked unfiltered cigarillos.
5 A. Yes.
6 Q. His evidence, given to Operation Paget, was that he
7 smoked at least 20 a day.
8 A. Yes.
9 Q. We have heard evidence from people like Mr Wingfield and
10 Mr Trevor Rees-Jones that his breath reeked of
11 cigarillos on the day of the crash.
12 A. Yes.
13 LORD JUSTICE SCOTT BAKER: All this explains the reading of
14 12, but not the reading of 20.
15 A. Precisely, my Lord.
16 LORD JUSTICE SCOTT BAKER: Can we move on? We have been
17 round and round and round this. If you have something
18 brand new, well, put it, but otherwise what's happening
19 is that this cross-examination is jeopardising the
20 prospects of hearing other witnesses.
21 MR MACLEOD: Sir, this is very important evidence.
22 LORD JUSTICE SCOTT BAKER: If it's something new, you put
23 it.
24 MR MACLEOD: We have spent a lot of time on things like
25 boxes and letters. This is critical evidence.

49

1 LORD JUSTICE SCOTT BAKER: If there is something new ...
2 MR MACLEOD: So you would accept that the 12.8 per cent
3 reading, agreeing with Professor Forrest, that was
4 a typical reading that could be recovered from a person,
5 a heavy smoker of unfiltered cigarillos, at the top
6 range of the bracket?
7 A. It is, yes. I would agree with that.
8 Q. Dealing with the 20.8 per cent, I will put to you
9 a series of propositions that Professor Forrest said
10 because he agrees with you that there is difficulty over
11 this. First of all, it can't be excluded that this is
12 simply a rogue reading.
13 A. On three occasions on two instruments?
14 Q. On the same test twice.
15 A. Yes, but that's repetition.
16 Q. And a different test each time. But it can't be
17 excluded; these things happen?
18 LORD JUSTICE SCOTT BAKER: He said there was no biological
19 explanation, it was the most troubling feature in the
20 case, and it would have been very helpful to have had
21 a chance to see what the French had to say about it.
22 A. I certainly agree with this.
23 MR MACLEOD: I am just going to press you. It cannot be
24 discounted; that was Professor Forrest's evidence --
25 A. Discounted from what?

50

1 Q. Rogue error; that it's just a rogue reading.
2 A. I find that difficult when he moves on to a different
3 technique and gets exactly ... I am sorry.
4 Q. Okay, a difference of opinion. Analytical error cannot
5 be discounted?
6 A. You had to make it three times.
7 Q. Professor Forrest made this point, that the
8 spectrophotometric method was used --
9 A. Yes.
10 Q. -- as opposed to the far more reliable gasometric
11 method --
12 A. He did.
13 Q. -- and he was not prepared to discount analytical error.
14 A. I am not ... very difficult because I use CO-oximeter or
15 used a CO-oximeter on fire-related deaths, carbon
16 monoxide poisoning. The only time I did not have to use
17 this equipment was effectively as I came up on an air
18 crash which I was involved in.
19 LORD JUSTICE SCOTT BAKER: Mr Macleod, we are not going to
20 get to the bottom of this. The evidence at the moment
21 is that there is no biological explanation. There may
22 be an explanation but we don't know what it is, full
23 stop.
24 MR MACLEOD: I have two more questions on this. In my
25 submission, sir, this is very important evidence. The

51

1 jury need to know what they can safely discount and what
2 they cannot safely discount.
3 The third matter which Professor Forrest put forward
4 as a possible matter was that this result was a rogue
5 result caused by some unusual contaminant, for example
6 a small particle of fat.
7 A. A particle of fat would block the aspiration into the
8 instrument and it would have to be a consistent error
9 from a small particle of fat on three analytical
10 occasions.
11 Q. But if the blood was contaminated in that way --
12 A. With fat?
13 Q. Yes.
14 A. But it's the haemoglobin the instrument is looking at,
15 not fat.
16 Q. So that's a difference of opinion between you and
17 Professor Forrest?
18 A. Yes, very much so.
19 Q. A calibration error on the machine is the final matter.
20 That cannot be excluded, can it?
21 A. If he calibrated his instrument. I hope he has
22 calibrated his instrument because he has used the same
23 technique for the 12.8.
24 MR MACLEOD: Thank you very much.
25 Further questions from MR HILLIARD

52

1 MR HILLIARD: Just this, because I do not want there to be
2 any misunderstanding, Professor: you accepted in your
3 1997 report that if a person had been drinking and could
4 apparently behave normally, that that didn't mean that
5 the person in question was fit to drive because you gave
6 the example that it could affect somebody's ability to
7 react appropriately. That is your view, isn't it?
8 A. That was the view at that stage.
9 MR HILLIARD: Thank you very much indeed.
10 A. Could I leave perhaps with the court this "Tolerance to
11 high blood/alcohol concentrations"? It will actually
12 show that even reaction times -- any clinical
13 manifestation looked at effectively, it does not
14 necessarily mean that at that level the person -- I was
15 surprised when I came across this. I really was
16 surprised. It's this paper that has changed my view on
17 it.
18 MR HILLIARD: All right. Thank you.
19 LORD JUSTICE SCOTT BAKER: Thank you very much, Professor.
20 (The witness withdrew)
21 LORD JUSTICE SCOTT BAKER: Then Professor Johnston.
22 PROFESSOR ATHOLL JOHNSTON (sworn)
23 LORD JUSTICE SCOTT BAKER: Do you prefer to sit?
24 A. I will sit, please. Thank you.
25 Questions from MR HILLIARD

53

1 MR HILLIARD: Can you give us your full name, please?
2 A. Atholl Johnson.
3 Q. And your occupation?
4 A. I am a professor of clinical pharmacology at Barts Inner
5 London University of London.
6 Q. Can you give us some idea, please, of your experience?
7 A. I trained originally with -- and my first degree was in
8 biochemistry and toxicology. Then my first job was at
9 Charing Cross Hospital in the department of forensic
10 medicine and toxicology, where I spent two years as
11 a forensic toxicologist.
12 I then moved to St Bartholomew's Hospital to do a
13 PhD in clinical pharmacology and I have been there since
14 with that. I also hold the visiting chair in forensic
15 pharmacology and forensic toxicology at Saint George's
16 Hospital Medical School, which is also University of
17 London. I have degrees in biochemistry and toxicology,
18 a PhD in pharmacology, a Master of Science in
19 statistics. I am a fellow of the Royal College of
20 Pathologists with a specialism in toxicology.
21 Q. Professor, you have been in court, I think, and heard
22 the evidence that's been given on this topic from
23 Professor Forrest onwards; is that right?
24 A. Yes.
25 Q. You very kindly prepared a document called "Toxicology

54

1 briefing document".
2 A. Correct.
3 Q. Do you have that there?
4 A. I have.
5 Q. Professor, you will understand that what I am going to
6 do is look, if I may, with you at one or two matters in
7 that document because, as you have heard, whatever the
8 answers to it, it's pretty well-trodden ground now, most
9 of it. I am sure you understand that.
10 A. Yes.
11 Q. First of all, the CDT measurement. Do you remember that
12 the suggestion is that blood from Henri Paul had gone to
13 Dr Dumestre Toulet and a reading of 32 -- do you
14 remember -- on this test had come back?
15 A. Yes.
16 Q. We have got it in a little bundle of documents, the test
17 results, but do you remember that you heard
18 Professor Forrest say that he thought it needed to be
19 looked at with caution, particularly when the sample had
20 been taken after death? He thought it shouldn't be
21 looked at in isolation, but said, do you remember, that
22 you could -- for example, if somebody had gone to see
23 their doctor and had said that they had had a problem
24 drinking over time, that if the report of this test was
25 to the same effect, well, you could look at the evidence

55

1 in conjunction; do you remember that?
2 A. You really need to have more than one measurement of CDT
3 because it's something that will change with time. Just
4 taking a snapshot of a single sample actually tells you
5 nothing because there is such an overlap between what
6 you see in moderate drinkers and people who actually
7 drink a lot.
8 Q. So do you agree with Professor Forrest, not something to
9 be looked at in isolation?
10 A. You can't look at it in isolation. It does not --
11 a level of 32 does not mean that you are a heavy drinker
12 or have been drinking heavily for the week before.
13 Q. Right. Then next, please, we have heard evidence about
14 results that came from samples all said to come from
15 Henri Paul, but, as it were, from different places, so
16 blood, urine, stomach contents. What I want to ask you
17 about in particular is vitreous humour. I am looking at
18 page 3 of 9 of your report. If we have open page 75.
19 We have a bundle of documents. It may still be there.
20 A. It is still here.
21 Q. We can see the vitreous humour result, our page 75,
22 level of 1.73.
23 A. Yes.
24 Q. These are Pepin results. So 31st August blood is 1.74.
25 We know that Professor Ricordel is said to have had

56

1 a result of 1.87. But anything that you want to say
2 about the vitreous humour figure compared with the
3 others?
4 A. From the first meeting I had with the Paget team and
5 actually saw all the data together, the one thing that
6 has disturbed me probably the most in terms of the
7 alcohol is this close agreement of 173, 174, 175. We
8 have two blood samples, taken four days apart, from
9 different parts of the body. We have a haemothorax in
10 one, taken from a chest cavity, an open chest cavity,
11 another taken from a vein or is it an artery -- that's
12 not clear from the evidence -- and then we have
13 a vitreous humour sample, and yet we manage to get 173,
14 174, 175. Any analyst that I have suggested that that's
15 probable to would just go "What?" They just can't
16 believe that you could possibly get those three results
17 that closely.
18 In fact it's possible to look at that statistically,
19 and the probability of, at random, getting those
20 three -- if the value actually was 174 -- if the mean
21 value actually was 174 and you took three identical
22 samples, what's the probability of getting 173, 174,
23 175? That works out to be something like 1 in 10,000.
24 Q. Can you just help us? Is this one of these areas where
25 it's possible for experts to differ, there can be more

57

1 than one view, because you heard Professor Forrest and
2 Professor Oliver and I don't think were as troubled
3 about it as you.
4 A. I don't think they have actually looked at it. In my
5 briefing document I have given you an analogy. I have
6 said that we know that the average height of a male in
7 this country is 175 centimetres. If you walked out into
8 The Strand or into Fleet Street today and you selected
9 three people at random, three males, what's the
10 likelihood of getting 174, 175, 176? Well, it's about
11 300 to 1.
12 The variability in height compared to the
13 variability in alcohol measurements between, for
14 example, vitreous humour and blood is much narrower.
15 The variability in height is probably about 4 per cent.
16 The variability has been measured in various studies.
17 There is an example by one from Sylvester which shows
18 that on average the variability is about 15 per cent
19 between vitreous humour, femoral blood and cardiac
20 blood. So it's very variable. Yet here we have 173,
21 174, 175. I just find it incredible.
22 Q. Then carboxyhaemoglobin. I suspect the answer is that
23 you can't shed any light on it either?
24 A. No, I mean, it's -- the most likely explanation is that
25 isn't Henri Paul's blood; it's someone else's. We have

58

1 gone through all the other ones.
2 Q. We have. Then you have a comment in your report just
3 about how common or not Prozac is; yes?
4 A. I did ask Dista, the manufacturers in this country, if
5 they could tell me how many people took Prozac in 1997,
6 and they said they couldn't even tell me this year how
7 many people took Prozac, but they said 54 million
8 prescriptions were written for Prozac last year. In
9 fact there is a report in the literature showing that
10 Prozac is so common that actually, if you take drinking
11 water, you can detect Prozac in it, so it's a widely
12 used drug.
13 Q. Can you help at all about 1997 and France?
14 A. It was a widely used drug in 1997. It's out of patent
15 now. The patents last 20 years, so 10 years ago it
16 would have been in patent, widely used.
17 Q. Tiapride, can you help about that?
18 A. In this country it's not used at all.
19 Q. In France?
20 A. I don't know what the --
21 Q. Have you tried to find that one out as well so we can --
22 A. Yes, I could find very little out.
23 Q. Is actually the position that that's pretty rarely
24 prescribed? Is that your understanding?
25 A. I would think that would be the case.

59

1 Q. So that one is quite rare.
2 A. But then it doesn't appear on most of the samples.
3 LORD JUSTICE SCOTT BAKER: Professor, you said you found it
4 astonishing the similarity, 1.73, 1.74, 1.75. What are
5 you suggesting there, that that suggests to you that the
6 results have been cooked or what?
7 A. That would be my interpretation. We have already seen
8 that, as Professor Forrest says, Professor Pepin has
9 used the facility to alter the results, and in at least
10 one sample we know this, where he has recalculated to
11 get it closer to what he got originally.
12 LORD JUSTICE SCOTT BAKER: Having got the original result
13 and it being put into the public domain, he has then
14 justifying it; is that --
15 A. I can't say. You will have to ask him that. But it's
16 just such an improbable result.
17 LORD JUSTICE SCOTT BAKER: Yes.
18 Mr Mansfield?
19 MR MANSFIELD: No, thank you, sir.
20 LORD JUSTICE SCOTT BAKER: Mr Keen?
21 Questions from MR KEEN
22 MR KEEN: Professor, just taking up the point raised by the
23 Coroner on the matter of coincidence. What we have here
24 is a statement by the French Public Prosecutor's office,
25 official or unofficial, on 1st September 1997 --

60

1 A. Yes.
2 Q. -- saying that the late Henri Paul had been a drunken
3 driver with a blood/alcohol measurement of
4 175 milligrams, to use the shorthand.
5 A. Yes.
6 Q. Then a series of tests on different blood samples, and
7 then, on the vitreous humour, giving results 1.74, 1.73,
8 1.75.
9 A. Correct.
10 Q. Before we go on to the plausibility or the
11 implausibility of that, can I ask first of all this: if
12 you measure alcohol in blood, assuming that you know the
13 body is in front of you, you take the blood out, you
14 measure it, and you measure it in the vitreous humour --
15 that is the liquid within the eye -- do you expect to
16 get a 1 to 1 comparison of the alcohol measurement as
17 between the vitreous humour and the blood in the same
18 body?
19 A. I think actually Professor Forrest dealt with this in
20 his statement, and I think what he said was: looking at
21 the vitreous humour alcohol concentration, normally you
22 would expect a vitreous humour alcohol concentration to
23 be higher than the blood/alcohol concentration, and, in
24 fact, if you look in scientific literature, there are
25 various reports, but on average somewhere between 20 and

61

1 40 per cent higher.
2 Q. So before we get to the statistical implausibility, just
3 comparing a blood analysis and a vitreous humour
4 analysis from the same body, the first thing you would
5 notice is that the vitreous humour is actually
6 20 per cent to 40 per cent higher in its measurement of
7 alcohol?
8 A. The vitreous humour has more water in it than blood and
9 that's why it's higher.
10 Q. I see. That's the first stage. Going on from there,
11 I think you have said that, on a statistical analysis --
12 and do I understand that statistical analysis in this
13 clinical field is one of your areas of expertise?
14 A. Yes, it is.
15 Q. You bring out a result that you can expect the results
16 reported, if they were genuine, to occur once in 10,000
17 occasions.
18 A. Yes.
19 Q. Yet Professor Forrest seems to pray in aid the
20 consistency of the results as somehow justifying the
21 reliability of what was done. Do you have any comment
22 to make on that?
23 A. That's in fact what he said when he was speaking to me
24 at the meeting with the Paget experts, and I said "To me
25 it means the opposite, actually something is very wrong

62

1 with those results".
2 Q. I think, in fact, ultimately in the joint experts'
3 report, following the meeting on 14th July 2007,
4 Professor Forrest actually came to agree that the
5 closeness of the results of the analysis for alcohol in
6 the two blood samples and the vitreous humour was
7 unusual.
8 A. Yes.
9 Q. Of course, all of that can be dispelled if you did have
10 a suitable chain of custody or a DNA analysis of the
11 material that was subject to toxicological analysis?
12 A. I am not sure whether it would dispel it or not because
13 you are still left with a fairly improbable result
14 which, even if you had a chain of custody which we
15 haven't got, I would have difficulty explaining.
16 Q. So even with a chain of custody there would be question
17 marks over the coincidence of these results?
18 A. Yes, and then when you start to see that he has already
19 altered one result to try and make it fit better to the
20 previous --
21 Q. This was the so-called recalibration calculation?
22 A. Yes.
23 Q. But in addition to that, would you make any observation
24 with regard to the evidence that Dr Pepin on at least
25 one occasion has, to put it politely, transposed the

63

1 result from one analysis to another report?
2 A. Yes. He has just assumed it would be the same result
3 and put it on a different report.
4 Q. But in the different report he purports to report that
5 he has done a further analysis, does he not?
6 A. Yes. There doesn't seem to be any evidence that he has.
7 Q. So that could go some way to explaining how you get
8 consistency of results?
9 A. Yes, it does. Just write them down the same.
10 Q. Could that ever be a legitimate method to employ?
11 A. No, definitely not.
12 LORD JUSTICE SCOTT BAKER: Mr Keen, I see we have been going
13 an hour and a half, which really is the limit,
14 I think --
15 MR KEEN: I was forgetting that we commenced at 9.30, sir.
16 I will not be much longer, but I am perfectly content to
17 break there.
18 LORD JUSTICE SCOTT BAKER: I think we might break now then.
19 (11.00 am)
20 (A short break)
21 (11.15 am)
22 (Jury present)
23 LORD JUSTICE SCOTT BAKER: Yes, Mr Keen.
24 MR KEEN: I am obliged, sir.
25 Professor Johnston, I think I in fact touched upon

64

1 this just before the break. You indicated that one of
2 your areas of expertise is statistical analysis in the
3 context of clinical pharmacology --
4 A. Yes, it is.
5 Q. -- and therefore the issue of what might be termed
6 "possibility, probability and reliability and
7 unreliability" in that context?
8 A. Yes.
9 Q. You were asked briefly by my learned friend,
10 Mr Hilliard, first of all about the matter of CDT.
11 I think we know -- and I will just quote it to you --
12 that "The joint experts agreed that a false positive
13 result can be obtained in post-mortem samples, and
14 consequently, taken in isolation, the CDT results in
15 this case cannot be said to be probative of recent heavy
16 alcohol use by Henri Paul". That touches upon the idea
17 of a false positive result from the toxicological
18 analysis.
19 Putting that to one side, I think in your most
20 recent briefing document you observe, under reference to
21 at least one published paper, that a value of 32 units,
22 which is the value given in the French reports, could in
23 fact come from a group or individuals who are only
24 moderate users of alcohol; is that right?
25 A. That's correct.

65

1 Q. I think you are referring to a paper by Athol(?) and
2 others published in 1998, which compared CDT
3 measurements as between light to moderate alcohol users
4 and, on the other hand, heavy alcohol users.
5 A. That's correct.
6 Q. Is it the case that a measurement of 32 falls within the
7 band for both types of user?
8 A. Yes, it does.
9 Q. Is it the conclusion of that particular report or of the
10 authors of that particular report that CDT for this
11 reason alone is of limited value in screening for
12 alcohol use?
13 A. That's correct.
14 Q. Your conclusion then is stated, if I can summarise it,
15 that even if you take the CDT measurement of 32, that
16 cannot be used to infer heavy drinking or heavy use of
17 alcohol.
18 A. That's correct.
19 Q. You have already touched upon the matter of the
20 coincidence, if you like, between the blood/alcohol
21 reading of 175, the vitreous humour of 173 or 174 --
22 A. Yes.
23 Q. -- and the fact that that in itself is highly unusual.
24 Does that factor have to be seen against the background
25 of evidence that Dr Pepin had engaged in the

66

1 transposition of results in his reports, even so far as
2 you have been able to discern from reading his papers?
3 A. I think it has to be seen against that and against the
4 fact that he altered one of the other results to make it
5 match.
6 Q. Thank you. Can I move on briefly to the question of
7 carboxyhaemoglobin again? We have heard repeatedly that
8 no-one approaching this from the perspective of
9 toxicological examination can advance a biological
10 explanation for the reading of 20 per cent or thereby
11 carboxyhaemoglobin in the blood samples.
12 A. Yes.
13 Q. I think you were present in court when we heard evidence
14 from Professor Oliver, saying that what we have are
15 three distinct measurements, all of which are
16 consistent --
17 A. Yes.
18 Q. -- the two measurements on one machine and one on
19 another.
20 A. Yes.
21 Q. Could I briefly ask you your view as to the possibility
22 or probability of these three results being due to error
23 in measurement or instrumentation?
24 A. These are quite close together, but then we are dealing
25 with an instrument or with instruments that are fully

67

1 automated and the measurement -- they are in general
2 very precise and were measuring just one sample.
3 If you had particulate matter, which I think has
4 been suggested by Professor Forrest, in the sample, as
5 Professor Oliver said, you might get blocking of the
6 machine, but even if it didn't block the machine, it
7 would cause light to scatter because this is
8 a spectrophotometric method, so it relies on shining a
9 light through the sample and getting a result from that.
10 Now, if there was a bit of particulate material, it
11 would alter the way the light -- it would scatter the
12 light to give you a more variable reading and these are
13 very consistent readings. We also have two slightly
14 different techniques used to measure the
15 carboxyhaemoglobin, so you have got one test and then
16 you have confirmation with a slightly different test.
17 So I am happy to accept these as being a reliable
18 result, and there is good evidence in the literature
19 that this particular technique, the oximeter, is
20 perfectly good for post-mortem measurements.
21 Q. I think you mentioned that one explanation is that the
22 blood samples didn't in fact come from Henri Paul.
23 A. That's the most obvious explanation, that it isn't
24 actually Henri Paul's blood we are looking at.
25 Q. Can you think of any other explanation?

68

1 A. We have sat round tables -- in Paget we had endless
2 discussions about cracked manifolds, about carbon
3 monoxide coming from the airbag, the amount of smoking
4 that Henri Paul did, a faulty gas fire in his flat,
5 which has been checked and found not to be faulty. No
6 explanation can be found for this level.
7 Q. Just on the matter of smoking, there is a reference to
8 the second set of blood samples revealing
9 a carboxyhaemoglobin level of about 12.7 per cent.
10 A. Yes.
11 Q. Professor Forrest suggested that that was a result you
12 might expect to find -- something of more than 1 in 20
13 heavy smokers. Do you have any observation to make on
14 that?
15 A. 1 in 20 is more frequent than actually it occurs. At
16 St Bartholomew's Hospital, we have one of the world's
17 experts on smoking and health, Professor Nicholas
18 Wall(?), who has actually done a study with 8,500 people
19 who are smokers, and the 99th percentile -- that is the
20 level at below which 99 people would be -- is 11.3. So
21 12.7 -- the very maximum it could be is 1 in 100, but
22 there are other studies in literature showing it's
23 probably 1 in 200 or 1 in 300.
24 Q. So even at 12.7 per cent, is it your view that the
25 carboxyhaemoglobin is exceptionally high even for

69

1 a heavy smoker?
2 A. Yes.
3 Q. In the final 1 per cent --
4 A. Yes.
5 Q. -- of any study that's been done?
6 A. Yes.
7 Q. Thank you. We have touched upon the question of genetic
8 testing, and I think you are aware that a portion of
9 liver taken on 31st August 1997 we understand was the
10 subject of DNA analysis.
11 A. Yes.
12 Q. We have heard that in fact such liver could have been
13 the subject of histological examination --
14 A. Yes.
15 Q. -- for signs of alcohol use?
16 A. Yes, that's correct.
17 Q. As regards that liver sample, we know from Dr Pepin's
18 records that there was, in fact, no finding of Tiapride.
19 A. That's correct.
20 Q. We have looked at the sheet already and I think you have
21 seen it yourself, have you?
22 A. Yes, I have it here, in fact.
23 Q. There are signs of Fluoxetine and Norfluoxetine?
24 A. That's correct.
25 Q. Just to be clear, do I understand that these are both

70

1 products of the use of the one drug, namely Prozac?
2 A. Yes.
3 Q. You indicated that, in the past year, there had been
4 something of the order of 54 million prescriptions for
5 Prozac --
6 A. Yes.
7 Q. -- or generic drugs of that type --
8 A. Yes.
9 Q. -- and, indeed, that the use is so wide that indicators
10 of the drug can even be found in ground water.
11 A. That's correct.
12 Q. Do I understand from your supplementary briefing
13 document that that information is taken from a report
14 from the Drinking Water Inspectorate?
15 A. Yes, that's correct.
16 Q. So what weight would you place upon the fact that two
17 samples both have traces of Fluoxetine and
18 Norfluoxetine? Would that be taken by you as
19 an indicator of a connection between samples?
20 A. It's an indicator of connection, but you can't put any
21 weight on that because Fluoxetine is so common in the
22 population.
23 Q. Can I come back for a moment to a joint report that you
24 prepared last year on the instruction of the Coroner,
25 and in particular to certain observations that you made

71

1 with regard to alcohol use.
2 Could we have [INQ0051951]? I think this is where
3 a series of assumptions are made about the quantities of
4 alcohol that could have been consumed by the late
5 Henri Paul; is that right?
6 A. That's correct.
7 Q. You begin by noting that:
8 "On the assumption that Henri Paul commenced
9 drinking at 7.30 pm and that his peak alcohol
10 concentration was achieved approximately one hour prior
11 to the crash at, say, 11.30 pm, then for a man of his
12 weight ... he would on average have had to consume eight
13 50-millilitre measures of Ricards ...", and you refer
14 there to 40 per cent alcohol by volume.
15 A. Correct.
16 Q. It has been suggested that in fact Ricard may be
17 45 per cent alcohol by volume. Does that materially
18 alter --
19 A. No, it won't make a very big difference. One of the
20 things we have not factored into that is the variability
21 in just the way you handle alcohol and that would just
22 be lost in the wash.
23 Q. But you are of the order of eight double measures of
24 Ricard -- that would be 16 single measures of Ricard --
25 in order to achieve the level of alcohol that is

72

1 purported to be measured in the blood samples?
2 A. Yes.
3 Q. You then go on to say that:
4 "This again would assume that he metabolised alcohol
5 at a rate of approximately 15 to 17 milligrams of
6 alcohol per 100 millilitres of blood per hour."
7 A. That's a generally accepted fact with alcohol.
8 Q. So we have to remember that as well as taking alcohol in
9 and absorbing, we are also metabolising it and ridding
10 ourselves of it?
11 A. That's correct. It's being removed the whole time.
12 Q. I think you go on to say that:
13 "Had he commenced drinking at 10 pm, then the
14 consumption of six double measures of Ricard would have
15 been sufficient to explain the concentrations in the
16 post-mortem samples that have been produced and reported
17 upon."
18 A. Yes.
19 Q. You go on to say that:
20 "On the assumption that Henri Paul commenced
21 drinking at, say, 10 pm ...", and this is where we have
22 him arriving at the Ritz, you will recall.
23 A. (Witness nods).
24 Q. "... two 50-millilitre measures of Ricard would result
25 in a concentration of approximately 30 milligrams of

73

1 alcohol per 100 millilitres of blood at the time of the
2 crash."
3 A. Yes.
4 Q. You go on to say that:
5 "The above calculations are based on the
6 Widmark-Factor-based equation and on the average of
7 a man of Henri Paul's weight. In all cases it is
8 assumed that all of the consumed alcohol has been
9 absorbed into his system."
10 A. Yes.
11 Q. Can you explain what the "Widmark Factor" is?
12 A. It's a way of calculating effectively total body water.
13 When you drink alcohol, it distributes in your total
14 body water and that's the concentration that you are
15 actually measuring when -- you have to adjust the blood
16 concentration slightly to get that because blood isn't
17 100 per cent water, but it has gone into your total body
18 water and that's what you are measuring.
19 Q. If we go on, you say that:
20 "With respect to the two drinks of Ricard, which are
21 sometimes transmogrified into more than four whiskeys,
22 the average man of Henri Paul's weight would at all
23 times have been below the legal limit for driving in the
24 United Kingdom."
25 A. Yes, if you calculate it, it works out between about

74

1 70 milligrams per cent, instead of 170 -- 175 milligrams
2 per cent is what we have been talking about. This would
3 calculate out to 70 milligrams per cent. The legal
4 limit for driving is 80 milligrams per cent.
5 Q. So if you assume that Henri Paul did arrive at the Ritz
6 at 10 o'clock, and did consume two 5-centilitre measures
7 of Ricard between, say, about 10 o'clock and 11 o'clock,
8 and if you allow for metabolism of alcohol up until the
9 time when he leaves the Ritz just after midnight --
10 I think it's about 20 past 12, we are told -- do
11 I understand that, according to these calculations and
12 assuming that there had not been previous consumption of
13 alcohol, he would have been within the legal limit for
14 driving?
15 A. Both in France and in Britain.
16 Q. Both --
17 A. In France it's 50 milligrams per cent and he would be at
18 a level of about 30 milligrammes per cent.
19 Q. So if those were the drinks he had consumed, he would be
20 within the French legal limit for driving and within the
21 UK legal limit for driving?
22 A. That's correct.
23 Q. Do I assume that these legal limits for driving, whether
24 they be lower in France than the UK or otherwise, are
25 intended to be indicative of the levels at which it's

75

1 feared someone could be impaired due to the consumption
2 of alcohol?
3 A. They are really set on a limit of what's considered to
4 be safe and not to increase your chance of having
5 an accident greatly.
6 Q. So do I understand that if somebody was below the French
7 legal limit for driving, let alone below the UK limit
8 for driving, with regard to alcohol, they would, on the
9 basis of the literature and research in this area, be
10 considered safe to drive from the perspective of
11 alcohol?
12 A. Yes. The Americans have done quite a large study on
13 this, looking at blood/alcohol versus the relative risk
14 of driving, and at a level of -- the French level of 50,
15 you increase your chance of having an accident by a very
16 small amount. Just slightly over would be -- than you
17 would get if you were sober. If you go up to the
18 British limit, it increases your relative risk by about
19 50 per cent. You have to talk about relative risk, and
20 understand that that means if you have an accident, one
21 accident in 10 years, on average, that means that you
22 might have one and a half accidents in 10 years.
23 Q. So it is a half an accident every 10 years?
24 A. 3 in 20.
25 Q. Personal experience might differ, of course,

76

1 Professor --
2 A. Yes.
3 Q. -- even without the influence of alcohol.
4 A. That's all other things being equal.
5 Q. I appreciate that. Do I understand that what you are
6 telling us is that if Henri Paul had consumed two
7 50-millilitre measures of Ricard in the time indicated
8 and then had driven at about 20 past midnight toward the
9 Alma Tunnel, he would have, in fact, absent earlier
10 consumption of alcohol, have been within the French
11 legal limit for driving so far as alcohol is concerned?
12 A. Yes, and it wouldn't have increased his risk of having
13 an accident.
14 Q. Would you just allow me one moment, sir?
15 LORD JUSTICE SCOTT BAKER: Certainly. (Pause)
16 MR KEEN: Thank you, sir. Thank you, Professor.
17 LORD JUSTICE SCOTT BAKER: Professor, you gave one answer
18 that Fluoxetine was very common in the population.
19 A. Yes.
20 LORD JUSTICE SCOTT BAKER: If you tested 100 people for
21 Fluoxetine, how many would you --
22 A. I really couldn't answer that. But it's so common that
23 there has been books published, for example, with titles
24 like "The Prozac nation". It has been such
25 a widespread, widely used drug.

77

1 LORD JUSTICE SCOTT BAKER: I appreciate it's common, but it
2 may be relevant as to --
3 A. I don't think anyone has done that research where they
4 have actually sampled what you would get if we were to
5 just --
6 LORD JUSTICE SCOTT BAKER: This may be important from the
7 point of view of the conclusions that the jury can reach
8 from the finding of --
9 A. I don't think anyone can tell you that, but I think the
10 fact that you find the widespread use of it can be
11 demonstrated by the fact you get it in ground water; you
12 know, there is so much of it about.
13 LORD JUSTICE SCOTT BAKER: Being in ground water is one
14 thing; being a sample in the human body is another.
15 A. Well, we all drink.
16 LORD JUSTICE SCOTT BAKER: Mr Croxford?
17 MR CROXFORD: No, thank you, sir.
18 LORD JUSTICE SCOTT BAKER: Mr Macleod?
19 Questions from MR MACLEOD
20 MR MACLEOD: Professor, picking up on that last point about
21 how common it would be if you tested an individual to
22 see if he had Prozac in his system, if you tested
23 an individual who had a combination of Prozac and
24 Tiapride, that would be a rare event, wouldn't it?
25 A. Yes. That's if you accept the Tiapride result.

78

1 Q. If somebody had Prozac and alcohol and Tiapride in their
2 blood, that person would be using the medication
3 contrary to medical instruction which would appear on
4 the packet and from the prescribing doctor?
5 A. Well, I haven't looked at the packet insert for Prozac,
6 but I have looked in the British National Formulary and
7 it is a small note that there is an interaction of
8 alcohol and Fluoxetine. It's not there in big letters
9 with a warning. It's rather like you get a garment that
10 says "Dry clean only" because they are just taking no
11 chances that normal washing will do it.
12 Q. Picking up the Coroner's question, if you were to ask
13 yourself the likelihood of somebody in this room, for
14 example, having a combination of both Prozac and
15 Tiapride in their blood system, if we were all analysed
16 now, what would the prospects be?
17 A. Tiapride in this room I would say is zero because it's
18 not available in this country. France I can't say.
19 LORD JUSTICE SCOTT BAKER: I notice you have not asked about
20 alcohol!
21 MR MACLEOD: I am not unsteady on my feet, I hope.
22 Dealing with the blood range and the levels of
23 alcohol found in the various analyses that were
24 conducted --
25 A. Yes.

79

1 Q. -- in fact the range is somewhat broader than you have
2 indicated, isn't it?
3 A. In what way?
4 Q. Professor Ricordel recorded 1.87; isn't that right?
5 A. That's correct.
6 Q. That wasn't mentioned in the correlation between the
7 blood analyses when you gave your evidence.
8 A. No, it certainly wasn't.
9 Q. Then there was Dr Pepin, 1.74.
10 A. Yes.
11 Q. None of the experts at the time, in the report of
12 December 1997, nor Professor Vanezis, nor
13 Professor Forrest, nor anyone else, has considered those
14 two results in any way surprising.
15 A. If those were the only two results, I would be happy to
16 accept them. That's more the sort of variation you
17 would expect to see.
18 Q. The third result, which was the result you recorded as
19 1.75, was actually recorded as 1.80, but Dr Pepin
20 subjected it to another calculation to reduce it to 1.75
21 and we don't know precisely why he did that; is that
22 right?
23 A. That's correct.
24 Q. So 1.80 was the analysis, not 1.75 --
25 A. Yes.

80

1 Q. -- and the vitreous humour was 1.73.
2 A. Yes.
3 Q. This matter has been considered by all the relevant
4 experts in this case, hasn't it?
5 A. Yes.
6 Q. In a joint meeting?
7 A. Yes.
8 Q. Could I please ask for the experts' joint report of the
9 25th July to be put up? It is [INQ0035089]. We see
10 this heading, "Meeting of experts, 14th July 2007".
11 A. Yes.
12 Q. If we could go, please, to the last page, which is
13 [INQ0035093], you see the tail piece of that towards --
14 just above "End of report", the words, if we could
15 highlight that. The first sentence:
16 "We believe that this report fairly characterises
17 the extent to which the signatories of this report can
18 agree over the issues identified as matters of concern
19 by those properly interested persons who have expressed
20 opinions."
21 At the bottom, we can see that Professor Forrest has
22 signed up, yourself, Professor Oliver and
23 Professor Vanezis.
24 A. That's correct.
25 Q. If we could turn, please, to [