Transcript The Two Kinds Sudden Cardiac Death

Interpreting Drug Levels
Steven B. Karch, MD
EAPCCT, ATHENS 2007
Absence of Disease
•
•
•
•
Q: Okay Dr X, did you essentially determine the
cause of death based solely on the toxicology
screen?
A: Cause of death is determined after review of
medical records, gross and microscopic findings
and the toxicology report.
Q: Are the blood levels you measured of X and Y,
is that the main reason you draw the conclusion
that X overdose was the cause of death?
A: That and the absence of any significant
disease process that would explain something
else occurring.
What this ME
assumes
• The postmortem blood levels accurately
reflect levels at time of death
• There can be no other explanations for
high postmortem levels besides taking too
much drug?
• Can an autopsy be called negative if
histology was limited or not done at all? Or
if genetic testing was not performed
Redistribution
•
•
•
•
The greater the volume of distribution, the
greater the concentrations increase after death
Postmortem measurements are site dependent,
values in LV>>RV
Blood drawn from femoral vessels is NOT
immune to redistribution
Not study has EVER shown that postmortem
measurements accurately reflect measurement
made in life
Redistribution
•
•
•
Morphine (and cocaine) have high Vss (2-6 for
MS and 2-3 for cocaine)
Metabolites of both drugs have very low Vss.
M3G and M6G on order of .5
As consequence, most of the parent
compound is in tissue and most metabolite in
blood, which means that only a small amount
of M need move back into blood to completely
alter ratio of parent and metabolite
Moveable Feasts
• Glucuronides, particularly of morphine
DO NOT REMAIN STABLE after death
• Free Morphine rises and concentration
may double or triple in first 8 hours
• Initially vitreous levels are HIGHER than
blood levels.
J Anal Toxicol. 2006 Nov-Dec;30(9):651-8.
DNA Diagnoses
•
•
•
•
•
Unable to metabolize drug
Metabolize too much drug
Invisible viruses
Produce abnormal channels and receptors
Don’t produce enough target
CYP Activity
• Activity of P450s is genetically
determined and varies from patient to
patient, and race to race
• Depending on amount of activity,
patients are classified as:
• Ultra-rapid
• Extensive
• Intermediate
• Poor
Morphine Poisoning
•
•
•
•
Normal full term male; mother given codeine 30
mg with acetaminphen 500 mg for episiotomy pain
q 12H
On Day12 poor color and not feeding
Found dead On Day 13 with free MS of 70ng/ml
Typically, infants of mothers taking codeine have
MS levels of 0-2.2 and American Academy of
Pediatrics says codeine users cans safely breast
feed
Not homicide
• Mother had stored Day #10 milk, MS level
was 87 ng/mg (typically 1.9 - 20 ng/ML)
• Infants have impaired ability to excrete M
• Testing showed mother heterozygous for
CYP2D6*2A allele with CYP2D6* 2x2
duplication - an ultrametabolizer. So were
other family members
Lancet 2006,368:704
•
•
•
Was This
Malpractice?
62-year-old male Hx chemotherapy, also
taking valproate for seizure disorder
Admitted with bilateral yeast pneumonia;
patient treated with ceftriaxone,
clarithromycin, and voriconazole for a
concurrent yeast infection, and codeine for
cough.
Suddenly became comatose - blood tests
showed toxic levels of codeine, morphine and
morphine metabolites
NEJM, 351:2827-2831,
It Wasn’t…
• Testing showed multiple duplications of
two P450 enzymes CYP2D6 and
CYP3D4
• Normally 2D6 converts 10% of codeine
to morphine, in this case much more
was converted
• Worse, 2D4, which converts morphine to
inactive metabolites wasn’t functioning
Was this homicide?
• A 9-year-old with multiple developmental
disorders admitted to hospital with
seizures. He was taking
methylphenidate, Clonidine, and Prozac
• He died of uncontrollable seizures, and
other children were removed from the
household
J Child Adolesc Psychopharmacol.
2000 Spring;10(1):27-34.
Absolutely Not
• Samples of the decedent’s liver were
tested for P450 abnormalities
• Child had no functional CYP2D6 and
could not form the main metabolite of
Prozac, causing Proxac concentrations
to reach dangerous levels
• Investigation was dropped
Who Poisoned Her?
• A 6-month pregnant woman diagnosed with
severe rheumatoid arthritis; treated with
methadone and antidepressants
• Found dead New Years Eve by husband
• No anatomic findings at complete forensic
autopsy
• “Heart blood” Methadone, 700 ng/mL,
Amitiptyiline, 1500 ng/mL, Nortriptyline, 2200
ng/ml
J Forensic Sci. 2003 Nov;48(6):1406-15.
She Did!
• CYP2D6*4 homozygous = poor
metabolizer
• Therefore unable to hydroxylate
methadone or amytriptyline or nortriptyline
• She couldn’t clear the drug from her
system and, essentially poisoned herself
Patchwork
• Young woman, polydrug user, hurt knee
and took one of room mates Duragesic
patches
• Found dead next AM with blood fentanyl
of 19 ng/mL plus, norfentanyl 7.6 +DPHN
(80 ng/ml, tramadol (.060 ng/mL),
citalopram 220 ng/mL
• Decreased CYP3A4*1B +
CYP3A5*3,
could not metabolize fentanyl
J Anal Toxicol. 2005 Oct;29(7):590-8.
PCR Arrays
• Use of PCR
arrays may
provide many
surprising
answers but
they are costly
Myocarditis
• At autopsy, incidence
is 0.1% to 5.5%, but
may be much higher
in selected series
• True incidence not
known because
infection is not always
clinically evident
Is this Myocarditis?
Yes, Actually
• 24 consecutive patients admitted within 24
hours of ACS, including ST elevation and/or
T-wave inversion, but negative angiography
• Each had RV biopsy - half used for routine
histological grading (Dallas Criteria), and half
analyzed with nested polymerase for all
viruses known to cause myocarditis
Histologic Dx of
Myocarditis
• 14 hearts with histologically proven
myocarditis
• Biopsied at autopsy with Stanford
Bioptome
• Only 11 could be diagnosed and that
required an average of 17 samples per
patient
JACC 1989;14;915-920
Some Virused Can
Only be Detected with
PCR
• In all 24 cases CPKMB, Troponin-T, and
C-reactive protein were elevated at
admission and increased while in
hospital, even after optimal therapy
• One patient with EB was positive by
Dallas Criteria, none of the others were
even borderline positive by histological
criteria
Editorial
Channelopathies
Na-K exchange
Implications
•
•
What if a viral myocarditis goes undiagnosed (as would
happen most of the time) and drug X is present “in
therapeutic range.” Would you agree with the doctor in
his deposition?
Or what if there were multiple duplications of an
important P450 enzyme. Without doing DNA testing, a
pathologist would not make the diagnosis, no
“significant abnormalities” would be reported, and death
would be blamed on a drug, no matter how much of the
drug was present (post hoc ergo propter hoc)
Free Morphine
•
•
•
Regina v David Moore: Alleged Homicide,
cancer patient on morphine
Very high free MS found in heart blood and
the accused was seen giving injection 20
minutes before death
Crown argued that high ratio of free to total
MS proved injection just before death otherwise it would all be bound
Uridine diphosphateglucuronosyltranferases
•
•
•
Morphine is converted to glucuronides (bound)
by UGT2B7
This enzyme is polymorphic (79 SNPs at last
count)
There are many reasons for increased free MS
postmortem, and polymorphism could be one depending on how many copies of the
abnormal gene are present, either too much or
too little morphine could be conjugated
Autopsy in SCD
Genetic Causes of
SCD
• Hypertrophic Cardiomyopathy
• Long QT syndromes (LQT1-LQT8)
• Drug HERG (KCN2) interactions
• Brugada Syndrome (SCN5A)
• Catecholaminergic (RyR2)
Drug-HERG
•
•
•
Major problem is acquired LQTS syndrome
caused by drugs that block human hERG
Result is delay cardiac repolarization and
increase risk of torsades de pointes arrhythmia
(TdP).
Not all hERG channel blockers induce TdP
because they can also modulate other channels
that counteract the hERG channel-mediated
effect (like aresnic) However, hERG channel
blockade is an important indicator of potential
pro-arrhythmic liability.
Symptoms
• One-third of symptoms are
indistinguishable from those of acute
heart attack
• One-third simply die suddenly
• Of those dying suddenly, as many as
one-half have had no previous
symptoms of any illness AND THEY
VERY WELL MAY HAVE A NEGATIVE
AUTOPSY
Warfarin
• FDA about to change Warfarin labeling
to recommend genetic screening
• CYP2C9, 3-20% of USA is deficient,
which means they cannot metabolize
• VKORC1, 50% of USA is deficient
(“Vitamin K epoxide reductase complex
subunit”) warfarin’s target
Warfarin Metabolism
Methadone
Compliance
• Methadone is substrate for P-
glycoprotein transporter encoted by
ABCB1
• Individuals with two copies of wildhaplotype need higher doses than
patients with only one copy
• Conversely, carriers of the AGCTT
haplotype need lower dose
Conclusions
• Anatomically normal heart is not necessarily
a normal heart
• Mere detection of a drug does not imply
causation, especially if autopsy is
incomplete
• High drug levels pre- or postmortem do not
prove excessive ingestion.
Things to Think About
• How can a pathologists testify with any
certainty that a heart is normal if the
heart has not been examined
completely?
• If there is even a 10% chance of being
mistaken, then when is it possible to
testify “beyond all reasonable doubt?”
Moral & Forensic
Issues
• If 10-15 % of SIDS are due to
channelopathy, shouldn’t testing be
included for family members if no one
else