New Test Results 06-07
Evidence Intro.
Toxic Toxicology & QHFSS Response
Inquest Transcript
Emergency Services

New Tests from 2006-07 Indicate that Mitch Collins Orally Consumed Heroin.
By Patrick J Collins (Mitch’s father): 26.10.07.

Foreword: On 1.09.06, Magistrate Michael Halliday concluded Mitch Collins’ adjourned inquest after he tabled “Exhibit A”, a major submission from Mitch’s mother (Desley) and me. This drew attention to numerous problems relevant to Mitch’s toxicology tests and to expert opinions based on these. Magistrate Halliday recommended that the Director of QHSS (Queensland Health Scientific Services) consider our concerns in “Exhibit A” and provide a report to the State Coroner Mr Michael Barnes and to us. The requested report was compiled in early September 2007 by Professor Naylor from the John Tonge Centre, after consultation with the Director of QHSS. However, we have not yet (at 26.10.07) received a copy. “Exhibit A” is included nearby on this web site under the menu heading “Toxic Toxicology”. The following addresses developments since then.

Significantly, to us, Queensland Health has recently, though implicitly, agreed with us that at no time since Mitch’s death until now, has this department employed a very senior toxicologist or pharmacologist who was/is competent to authoritatively interpret autopsy toxicology test results. For instance, the Health Minister Stephen Robertson, in a letter to us dated 09.10.07, stated:

            “… the possibility of employing a medical pharmacologist [at QHSS] with additional expertise is under active consideration.”

The Health Minister and Professor Charles Naylor from the John Tonge Centre deserve credit for this, for the need is great. It has already caused legal problems. The following exemplifies this. It draws attention to the faulty notion that opinions from high status interstate “experts” should be accepted as irrefutable by the Offices of the State Coroner and the DPP (Director of Public Prosecutions). This statement is based on the results of recent toxicology tests that were conducted on Mitch’s autopsy specimens on our behalf, by laboratories that were not controlled by Queensland Health. We listed most of the tests and others in “Exhibit A”, for they should have been conducted by QHSS, to enable scientists to determine if Mitch swallowed heroin, or if it entered his body via an injection. It is my opinion that the results of the new tests, which were carried out after the DPP dropped a murder charge against Mardi Mclean, indicate that Mitch unknowingly swallowed the heroin. In other words the DPP’s decision to drop the murder charge was based on flawed toxicological evidence and inadequate “expert” opinions.

As the State Coroner is currently considering the new test results, I have advised Attorney General Shine and Health Minister Robertson of this. Whether or not I provide the DPP with the results will be decided after the State Coroner has considered them. Accordingly, to avoid potential sub judice issues, I have not, in the following summary, named the scientists who conducted the new tests, or the laboratories where the tests were undertaken. It is pertinent to mention that Professor Naylor, during a recent interview with Desley and me, asked us if we had considered applying for a new inquest. However, we will not be doing this, for we believe the new test results confirm Magistrate Halliday’s conclusions at the end of his August 2001 inquest hearing. In other words, there is nothing to be gained from another inquest. On the other hand, there is much to be gained if a murder trial is eventually held. Time will tell.

Disclaimer: Every opinion or conclusion passed by me below, was based on my understanding of highly credible and relevant journal articles and books. If I have misunderstood or wrongly applied the relevant research this was not intentional. However, I will, if asked by appropriate persons, readily provide copies of, or references to, any of the sources upon which I based my opinions and conclusions. Most of these sources have in fact already been provided to the Offices of the State Coroner and the DPP.         

Why the Tests Were Conducted:

Professor Drummer’s speculations:

The murder charge against Mardi McLean was dropped essentially because of Professor Olaf Drummer’s speculations in his report of 20.08.04. For instance, although the concentrations of heroin metabolites in Mitch’s stomach were about 500% higher than in his blood, Drummer suggested that gastric bleeding or bile could have been the source of Mitch’s gastric morphine and its metabolites. Drummer also speculated that gastric morphine metabolites might have converted back to morphine between Mitch’s death and the analysis of his stomach contents. Although Drummer could have sought extra tests to confirm or negate these speculations, he did not do so. These tests have now been conducted. The results negated all of Drummer’s above speculations and strongly suggest that Mitch orally consumed the heroin that killed him. As McLean swore that she witnessed Mitch inject himself with heroin twice, just before he died, I believe the new evidence also suggests that the murder charge should not have been dropped. (For further details of Drummer’s speculations see an adjacent paper titled “Toxic Toxicology”.)

The Integrity of the New Tests:

With support from Magistrate Michael Halliday and Professor Charles Naylor from the John Tonge Centre, Desley and I arranged to have extra toxicological tests conducted under the supervision of a senior chemical pathologist from a major private enterprise pathology company. However, Desley and I were at no stage involved with the process. Scientists ensured that Mitch’s autopsy specimens were transported and preserved appropriately: i.e. “continuity” of evidence was ensured. Three laboratories were involved. Two were interstate government laboratories, which are similar to the QHSS laboratory at the John Tonge Centre. The above pathology company conducted a gastric haemoglobin test.   

The Test Results:

Gastric Bile Test:

Professor Graham Cooksley from the QIMR (Queensland Institute of Medical Research) suggested that Desley and I ascertain the concentration of Mitch’s gastric bile salts, for this could be used to determine the volume of bile, if any, that was in Mitch’s stomach contents. A NSW Government laboratory conducted this test and found:

Mitch’s gastric bile salt concentration was: 6umol/litre (or 6micromols/litre).

The above laboratory reported that this is within the normal “reference range”: i.e. the presence of such tiny quantities of gastric bile salts is not unusual.

Gastric bile volume (calculated by Patrick Collins):

To convert the above concentration to a bile volume, it was necessary to allow for the presence of other bile substances, especially cholesterol and phospholipids. When the molecular weights and proportions of these were taken into account, Mitch’s gastric bile quantity was approximately 3.6mg. As the specific gravity of bile has an approximate range of (1.01 – 1.04) this was also taken into account. From all of this:

Mitch’s gastric bile volume was less than 0.004mL, which is about 1,000th of the volume of an eye-drop.

(Note: 1mL means one millilitre and is equivalent to 1cc. Also, 1,000mg = 1gram; and 1,000mL = 1litre)

Mitch’s autopsy gastric TM (total morphine) quantity was 2.6mg in August 2001. If this had been in his above tiny gastric bile volume, the bile TM concentration would have been greater than 700,000mg/L (or 700,000mg/kg). However, such a concentration is hundreds of times higher than any reported in science journals in relation to heroin or morphine deaths. Also, it is improbable that 2.6mg of TM could be dissolved in the above tiny bile volume.

(Note: mg/L means milligrams per litre. Similarly: mg/kg means milligrams per kilogram. In many reports mg/L and mg/kg are used as synonyms.)

Conclusion: Mitch’s gastric bile volume of less than 0.004mL was too miniscule to have transported 2.6mg of TM into his stomach: i.e. bile was not the source of Mitch’s gastric heroin metabolites. Professor Drummer’s bile source speculation was wrong, but he should have determined this prior to his submitting speculations about this in an opinion on Mitch’s death.

(Note: as TM can be found in bile after either oral consumption or an injection of heroin/morphine, the presence of gastric bile does not verify that an injection was the source. It simply introduces doubt about how the heroin/morphine was taken.)

* * * * *

Gastric Blood Test:

Professor Graham Cooksley from the QIMR also suggested that we seek an analysis of Mitch’s stomach contents to determine the quantity of haemoglobin, if any, that was present. This was to verify or negate Professor Drummer’s speculation that gastric bleeding within minutes after a heroin injection might have been the source of Mitch’s gastric heroin/morphine metabolites. Drummer did not however explain why there was no blood in “vomitus” that paramedics found on Mitch’s face near his mouth. This aside, a major Queensland pathology laboratory conducted this test and reported:

“No detectable haemoglobin was present”.

In other words there was no blood in Mitch’s stomach contents.

As Mitch’s stomach contents specimens were frozen by QHSS, it appears unlikely, from my study of the literature, that every trace of free haemoglobin (if any was ever present) could have disappeared during storage prior to testing. However, as this test should have been conducted prior to Drummer submitting his opinion on Mitch’s death, this raises other issues. For instance, how can any Court determine if a drug has been orally consumed or injected, if the relevant experts do not conduct pertinent gastric tests before the gastric specimens can decay during storage. Drummer could not have done this until around August 2004, but QHSS analysts and pathologists should have done so in early 2001. QHSS lacks the facility to conduct this test, but they could have asked a commercial pathology laboratory to do so. In the face of this, and based solely on appearance, a forensic pathologist and a chemical analyst from QHPSS regarded the red substance in Mitch’s stomach contents as blood-like. This was recorded in Mitch’s Post-mortem Report and stated in evidence during Mitch’s inquest in August 2001.

It astounds me that forensic scientists could have visually decided that a red liquid in Mitch’s stomach was blood, and then referred to it as blood in Court, especially so, given that Magistrate Halliday had requested an analysis of Mitch’s stomach contents. This was reminiscent of an aspect of a Lindy Chamberlain trial. In this instance, a forensic scientist identified a substance in Lindy’s car as blood. However, it was later shown to be a rust proofing compound. In Mitch’s case, the red substance has still not been identified. Could heroin have entered his body in this? Hopefully, this question will one day be answered. 

Conclusion: If storage conditions were appropriate to prevent total free haemoglobin decay, gastric bleeding soon after a heroin injection was not the source of Mitch’s gastric heroin/morphine metabolites. There is no evidence that supports Professor Drummer’s speculation that early bleeding might have occurred.

Note: A leading chemical pathologist told the Coroner’s Court that if the unidentified red substance in Mitch’s stomach was blood or a blood product, it would have been from bleeding that occurred well after death: i.e. after Mitch’s body had started to decay. However, Drummer, whose academic resume of 20.08.04 did not include MB.BS (Bachelor of Medicine and Bachelor of Surgery) qualifications, refuted this chemical pathologist’s opinion but, as previously stated, Drummer did not seek an analysis of Mitch’s stomach contents for the presence of gastric haemoglobin, as evidence of gastric bleeding. As was also previously stated, the unfortunate truth is, no analyst, pathologist or toxicologist bothered to check if the red fluid in Mitch’s stomach was blood, but most assumed that it was. However, the new evidence suggests that this red substance was another fluid, eg. a red drink such as Berocca or a “Bloody Mary”, which could have been used as a solvent for an oral dose of heroin. By contrast, Drummer sought an extra test to determine Mitch’s gastric methamphetamine level. But this was in an attempt to negate an associated query by me, in relation to bile as a source of Mitch’s gastric morphine. So why didn’t Drummer check to see if any gastric blood was present? 

* * * * *

Tests for Gastric 6MAM and Changes to Gastric Morphine:

Monoacetylmorphine (or 6MAM) is a heroin metabolite that has no other source but heroin. If it is found during analysis of autopsy specimens, it is certain that the deceased person had orally consumed, “snorted” or injected heroin. The QHSS laboratory at the John Tonge Centre “detected” 6MAM in Mitch’s stomach and urine but did not quantify either. Surprisingly, Professor Drummer and Professor Duflou, who provided the DPP with opinions about Mitch’s death, also failed to have the gastric 6MAM level quantified. Another unfortunate truth is: not one expert had Mitch’s gastric 6MAM level determined, although this was potentially the most significant evidence of oral heroin consumption: especially so if a bolus of 6MAM had been identified.

As an unknown quantity of 6MAM was in Mitch’s stomach contents, we did have the level determined. A southern government laboratory conducted the analysis. The results follow.

Gastric 6MAM Test Result:

Mitch’s gastric 6MAM concentration was 0.25mg/Litre.

As the volume of Mitch’s gastric contents was one litre, the above shows that these contents contained a quantity of 0.25mg of 6MAM.

As the above concentration was determined in late 2006, i.e. nearly six years after Mitch died, it is very possible that the gastric level was much higher when he died. This is because 6MAM metabolises to free morphine (FM) and some 6MAM could have diffused from his stomach to his blood and other organs. No 6MAM was found in his blood but this was not unusual, as 6MAM survives for a very short time in blood. However, it is known to survive for much longer periods in mildly acidic urine and in stomach contents (Moriya 1996).

Comment: Had the relevant scientists known that Mitch’s gastric 6MAM level was at least 0.25mg/kg when he died, they should logically have considered that this gastric 6MAM could have resulted from the oral consumption of heroin, and it was probably higher than 0.25mg/kg at death. 

Also, as Mitch’s stomach contained 2.6mg of TM (total morphine), when this is added to the above 0.25mg of 6MAM, his total quantity of autopsy gastric heroin metabolites was at least 2.85mg, whereas a litre of his peripheral blood probably contained less than his autopsy level of 0.5mg TM at death. 

Note: 6MAM is not found in bile or liver contents. Therefore, as no gastric haemoglobin or significant gastric bile were found, the gastric 6MAM, FM and TM levels indicate that Mitch orally consumed heroin.

* * * * *

6MAM “detected” in Mitch’s Urine:

It is accepted that following oral heroin consumption, only traces of 6MAM (if any) enter circulating (systemic) blood, as 6MAM is metabolised to morphine before it reaches this blood. Therefore, 6MAM is not found in the urine of living persons who swallow heroin unless they also injected. However, 6MAM is a highly lipophilic drug, and, in deceased persons, such drugs are known to diffuse from sites of high concentration to sites where the concentration is low. Also, as paramedics noticed “vomitus” or “aspirate” on Mitch’s face, 6MAM may have entered his blood via his lungs. In other words, it is not valid to conclude that the 6MAM “detected” in Mitch’s urine was from a heroin injection, for it could have derived from oral heroin consumption. However, it is possible that Mitch was injected with heroin after falling asleep from an oral dose of the same drug. For instance, a mark from an injection under his tongue would not have been found.

[Note: The senior chemical pathologist who told the Coroner that Mitch had orally consumed heroin, did not rule out the possibility that some heroin had entered Mitch’s body via an injection. Also, Dr Sinton, who conducted Mitch’s autopsy, told Desley and me that it would not be difficult to inject heroin into an unconscious person.]

* * * * *

Changes to Mitch’s Gastric Morphine During Storage:

As stated previously, Professor Drummer speculated that Mitch’s gastric free morphine level may have increased during frozen storage as a result of morphine metabolites converting back to FM (free morphine) by hydrolysis. Independent tests show that this did not happen.

Mitch’s autopsy was held on 18.12.00 but his gastric contents were not partially analysed until 02.08.01. The gastric tests that we later requested were conducted on or about 7.11.06. The table below compares the results from the separate tests.

Abbreviations in the table below: 6MAM is monoacetylmorphine. FM is free morphine. TM is total morphine (i.e. FM plus its metabolites). FC is free codeine. TC is total codeine.

Gastric Codeine is included as it was an impurity of street heroin.

A Comparison of Gastric Heroin Metabolites From Two Analyses.






FM Metabolites
TM – FM 





None reported








Not reported

Southern Lab 7.11.06

None reported









Morphine changes:

Mitch’s gastric free morphine (FM) decreased from 1.4mg/kg between August 2001 and November 2006. In other words, Drummer’s speculation that Mitch’s gastric FM may have increased post-mortem was wrong.

A corresponding increase in the gastric total morphine (TM) was also determined. This indicates that gastric metabolism of FM to morphine metabolites had occurred during the same period. It also strongly suggests that this metabolism had been occurring during five years of storage after Mitch’s autopsy, and also in his body between death and autopsy. In other words, Mitch’s gastric FM was probably higher than 1.4mg/kg when he died: not lower as speculated by Drummer.

Note: The increase in gastric TM was greater than the decrease in FM. This was because the major metabolites of FM have higher molecular weights than FM.

Gastric Codeine: As heroin is a derivative of the poppy flower, from which FM and Codeine are extracted, Codeine (and sometimes 6MAM) are found in heroin produced in backyard laboratories. Accordingly, the 0.7mg of gastric total codeine that was determined in November 2006 was part of the heroin dose that killed Mitch.

* * *

Total Gastric Remnant of the Fatal Heroin Dose:

The total weight of the heroin metabolites plus impurities found in Mitch’s stomach was the sum of the 6MAM, TM and TC above.

These were at least:

In August 2001 = 3.55mg.

In November 2006 = 3.75mg.

Although 3.75mg is a relatively small amount, the new tests show there is no evidence that it came from a source other than orally consumed heroin.

Note: As stated earlier, 6MAM is a highly lipophilic drug that can diffuse throughout the body between death and autopsy. As Mitch’s gastric 6MAM concentration was 0.25mg/kg, six years after he died, it was almost certainly higher when he died. This also indicates that when he died the quantity of the combined quantity of gastric heroin metabolites and impurities was greater than the above amounts.

* * * * *

Test for Liver FM and TM:

The same southern laboratory that determined the levels of Mitch’s gastric heroin compounds also evaluated the levels of FM and TM in a specimen of his liver. The concentrations were:

FM @ 0.5mg/kg   and   TM @ 2.5mg/kg.

As Mitch’s gastric FM concentration was 1.4mg/kg in August 2001 and although it had fallen to 1.1mg/kg by November 2006, it was never as low as it was in his liver. It has also been reported (by Moriya 1996) that FM is very stable even in putrefying liver, whereas morphine metabolites can totally convert to FM at room temperatures within 10 days. In other words, Mitch’s post-mortem liver FM of 0.5mg/kg was never higher than this.

It can therefore be concluded that Mitch’s August 2001 post-mortem gastric FM concentration of 1.4mg/kg did not derive from his liver, eg by diffusion before death.

* * * * *

Some conclusions from the new (2006-07) toxicological tests conducted on Mitch’s stored autopsy specimens.

Assuming that Mitch’s autopsy specimens were stored appropriately to prevent the complete breakdown of any free haemoglobin, he orally consumed all, or at least some, of the heroin that killed him. The evidence that supports this includes:

(i): His gastric autopsy FM and TM levels were both approximately 500% of his peripheral blood FM and TM levels of 0.3mg/kg and 0.5mg/kg. However, it is very possible that his peripheral blood FM concentration had increased post-mortem, eg from two sessions of post-mortem CPR. If this happened, his gastric FM level was even greater than 500% of his peripheral blood FM level. Also, post-mortem blood FM levels can increase from the post-mortem metabolism of 6MAM to FM.

Research has verified that ante-mortem blood FM levels cannot be determined from post-mortem blood levels. Blood FM increases of up to 850% have been noted between when bodies were admitted to a mortuary and the subsequent autopsies.

(ii): His stored gastric contents contained a minimum of 2.85mg of heroin metabolites, including 0.25mg of the heroin marker 6MAM.

(iii): It is very likely, perhaps highly probable, that his gastric 6MAM level was higher at death than it was at autopsy. This is because 6MAM could have diffused from his stomach to his blood and to other organs.

(iv): 6MAM from Mitch’s stomach could have been the source of the urine 6MAM that was “detected” in 2001. As there is research that lends support to this conclusion, it is not speculative.

(v): As Mitch gastric FM level fell between August 2001 and November 2006, it is very likely that his gastric FM = 1.4mg/kg had also fallen during storage between his December 2000 autopsy and August 2001.

(vi): The recent tests negate speculations that Mitch’s gastric morphine compounds could have derived from blood or bile that entered his stomach.

(vii): There was no blood in vomitus on Mitch’s face, but there would have been if he had experienced gastric bleeding within minutes of a heroin injection.

(vii): His liver FM concentration of 0.5mg/kg was too low to have been the source of his autopsy gastric FM of 1.4mg/kg.

(viii): His gastric free and total Codeine levels were consistent with their deriving from Codeine impurities in street heroin. When added to his gastric heroin metabolites, the total quantity of autopsy gastric heroin compounds and impurities was greater than 3.5mg. This is a significant proportion of a potentially fatal heroin dose in a non-user who lacked heroin/morphine tolerance. However, this quantity was possibly even greater at death, for 6MAM could have diffused from his stomach between death and autopsy.

(ix): Although it is evident that Mitch orally consumed heroin, it is nevertheless possible that he was injected with more heroin after he was unconscious. This could have been under his tongue or in some other non-obvious body location.

(x): Paramedics, who found no needle marks, confirmed that marks found at autopsy were consistent with their attempts to inject life-saving drugs into Mitch. Hypotheses that contradicted this did not take into account McLean’s claim that she saw a large lump from a botched attempt by Mitch to inject himself. If this had happened, the skin damage should have been obvious (see below). 

* * * * *

Additional support for oral heroin consumption from needle marks:

Paramedics and police, who initially suspected that Mitch had died from a heroin overdose, searched for but could not find any needle marks on Mitch’s arms. Their searches also failed to discover any needles or heroin paraphernalia. When needle marks were found on Mitch’s arms during his autopsy, a senior paramedic gave evidence that these were consistent with his attempts to inject potentially life-saving drugs into Mitch.

Professor Duflou, who, in 2006, provided an opinion on Mitch’s death, suggested that some needle marks observed during autopsy could have been from heroin injections, for the marks might have become more obvious during the two days between death and autopsy. However, Duflou had not read Mardi McLean’s claim to the Coroner, that she saw a large lump on Mitch’s arm as a result of his missing a vein while injecting heroin about five hours before he died. It follows from Duflou’s evidence, that associated needle damage should have been more visible with time, but it was not seen by any person.

A needle that Mitch had probably used to self-inject Ribavirin, a legal antiviral pharmaceutical, contained traces of blood. QHSS analysed this but found no heroin in the blood. Similarly, some of Mitch’s blood specimens from a then recent medical treatment programme were also analysed but again, no heroin was found.

The only witness to have stated that Mitch injected heroin was Mardi McLean. However she did not state this until her own heroin needles were found hidden in her bedroom two days after Mitch died. Until then she had claimed that Mitch had had a heart attack. She also told Emergency Services and paramedics that she did not know if Mitch had used heroin. Professor Drummer’s speculative opinion supported McLean’s claim, but the new test results should legally nullify this support.

Note: Unfortunately, the issue of needle marks could have been quickly resolved, if QHPSS pathologists had preserved flesh from around the needle marks and had these analysed. It is known that heroin concentrations are high in such specimens, if a relevant heroin injection had occurred.

From all of the above it is my opinion that Mitch died from a heroin overdose that he orally consumed, at least in part. If this is correct, it verifies that McLean had “guilty knowledge”, for she belatedly said that Mitch had injected heroin twice but she did not mention any oral consumption of heroin. It is now up to the Offices of the State Coroner and hopefully the DPP to decide what action to take as a result of the new evidence. However, it can be assumed that the matter will not be swept under the proverbial carpet without very loud protestations.

Regardless of what action, if any is taken by the above legal personnel, nothing can take away the gains made from the new test results. When read in conjunction with evidence from Mitch’s inquest and telephone records accessed since then, they show that Mitch did not kill himself by an accidental fatal heroin overdose. None of Mitch’s loved ones or close friends ever believed that he did, but it is comforting to have this confirmed scientifically.

* * * * *

Media Releases