Transcript 3-JANVIER FINAL - Seattle Children's

Annie Janvier, MD, PhD, FRCPC
Associate Professor of Pediatrics
University of Montreal
Who’s Best Interests?
Resuscitation Decisions for Neonates
Medical developments
• 50 years ago, many neonates / children /
patients did not survive their disease or
accident
• Less treatments to propose
– « Nature decided and took it’s course »
Medical developments
• Technological and medical discoveries 
increased survival
• Now physicians have to take some of these
decisions
– Does this intervention work?
– Should I offer this intervention? Is it worth it
for my patient?
– Should I withdraw this intervention?…
What is in the best interest for my
patient / loved one ?
• Not a new question
• There are some interventions that patients,
physicians and / or families may find
unreasonable
– Medical treatments
– Resuscitation: do / do not
– Surgical treatments
US 1984 amendment:
to treat or not to treat?
Treatment
efficiency
Beneficial
Parents
want Tx
Parents
refuse Tx
Yes
Yes
Grey zone
Yes
No
Futile
yes?
No
Treatment for extreme preterm infants:
beneficial, “grey zone” or futile?
Beneficial, “grey zone” or futile?
• Women 23 week gestation (as per dates),
EFW: 500g, imminent delivery
• Wants to do “everything” for baby
• Do you think it is in this baby’s best interest to
receive intervention?
• Would you offer intervention?
• Would you obtain a court order to withhold
intervention?
Beneficial, “grey zone” or futile?
• How can we determine this?
– Survival
– Outcome
– Quality of Life
– Best Interests
What are the outcomes at 22-23 weeks?
France
• “Parents should receive appropriate
information of survival and risks”, “informed
consent is given in the delivery room”
• “In France, a do not resuscitate order is
(generally made) for newborns weighing less
than 600g and/or with a gestation of less than
24 weeks since the mortality is 100%”
– Desfrere, L; Tsatsaris, V.; Sanchez, L.; Cabrol, D., and Moriette, G. [Extremely preterm
infants: resuscitation criteria in the delivery room and dialogue with parents before birth].
J Gynecol Obstet Biol Reprod. 2004 Feb; 33(1 Suppl):S84-7.
• NICHD neonatal network, the survival of a 23 week
female weighing 600g is 40%
What is the survival at 22-23 weeks?
Germany Herber-Jonas et al, Am J Obs Gyn, 2006Jul, 195(1): 16-22
•
•
•
•
General resuscitation at > 24 weeks,
22-24 weeks: according to parents wishes
In this study, 60% resuscitation at 22-24 weeks
When resuscitated
– Survival at 22 weeks: 37%
– 23 weeks: 75%
What do we tell parents: subjective
(23wk girl, 600g, PNS)
• « 0%: 23wk babies do not survive in our center »
• « 1 in 4 survive without moderate or severe impairment»
• « About 1 out of 2 survive. Among survivors about 1 out of
2 do not have any disability, 1 in 4 have major disability»
• « About 1 out of 2 survive. Among survivors about 1 out of
2 do not have any disability when evaluated at 1824months. If you then evaluate these children at 8y old, 3
out of 4 do not have any moderate-severe disability. When
babies do have disabilities, parents generally adapt to these
difficult challenges »
Morality and ethics are relative:
adaptation to our environment
« It seems to me morality does change. The
evidence is all around us. It used to be that
one could smoke cigarettes in hotel rooms but
had to go out into a dark alley to buy
pornography. Now, we can lie in bed and
choose among four dirty movies on
Spectravision but have to sneak out into the
alley for a smoke. »
– John Lantos, Do we still need doctors?
The evaluation of the outcomes
• The outcomes depend largely on where you
are , the attitudes of healthcare providers and
who is on call
• Pediatricians: generally pessimistic about
survival and disability (Haywood et al, Pediatrics 1998)
• Pessimists would intervene less often
• Nurses are even more pessimistic than
physicians (Streiner et al, Pediatrics, July 2001)
Parents do not share our pessimism
« an attempt should always be made to save all
infants, regardless of outcome »
• Lee et al, Pediatrics, 1991 (Newfoundland)
What if they knew?
Janvier et al, Arch Pediatrics (n=272; resp rate: 90%)
• There is an impending delivery of an infant with a
50% predicted survival and a long term outcome
as follows:
– 50% “within normal limits”
– 35% behavioural / learning disabilities
– 15%-25% major disability
• Do you think such a baby should be resuscitated
in the delivery room?
Corresponds to a 24 wk infant
What if they knew?
Hypothetical case (24 wk)
vs actual 24 wk
70
60
50
40
pourcentage
Hyp oui
24 oui
30
20
10
0
Inf
Children
Inf néo
RVH
Inf obs
RVH
Tous rés
Rés ped
Rés obs
What is their future quality of Life?
Dr Saigal
• Quality of life: Dr Saigal, JAMA 2007
• QOL of < 1000g at adolescence and adulthood
• QOL of ex-preterm and controls difficult to
differentiate
• In all her QOL research, QOL is always superior
when judged by children and their parents
compared to heathcar providers.
Design: McMaster Longitudinal* Cohort,
Dr Saigal, JAMA
Prospective Longitudinal Population-Based Study
 90% follow-up, evaluated at 23 years of life
 ELBW: 179/ 397 (45%) survivors, 501 to 1000g BW
(1977-82 births), born in c-w Ontario
 NBW: 145 socio-demographically matched reference
group recruited at age 8 years from
a random list of public school children
Chronic Physical Health Conditions
copyright Dr Saigal, JAMA
ELBW
n = 149
%
No Problems
1-2
3
Problems # 1
Mean (SD)
1
13
56
32
2(2)
Mean # for those with  1 problem
NBW
n = 133
%
35
45
20
2(1)
P
<.001
NS
QOL (23y old)
Dr Saigal
•
•
•
•
•
•
•
Permanent job = 35%
Completed school = 40%
Having children= 10% same as control
Living independantly= 30% same as control
Being in school education= same as control
Being sexually active = less (60% vs 78%)
Problems with the law = less frequent
Quality of Life conclusions
Overall, despite their health problems,at young
adulthood, ELBW adults are functioning at about
the same level as the NBW participants in their
educational attainments, employment status,
independent living, marital status/ cohabiting,
and becoming parents
Life is not a matter
of holding good cards,
but of playing
a poor hand well
Robert Louis Stevenson (1850-1894)
Decision making for incompetent patients
• Guiding principle = patient’s best interest:
– Legal decisions
– Parents as surrogate decision-makers
– Physicians advocating for their patients
• If an intervention is in an incompetent patient’s
best interest -and serious harm would follow
withholding intervention- refusal of care is
generally not acceptable ethically and legally
Estimations of best interests for
incompetent patients
• Anonymous questionnaire to physicians in 8
culturally different countries
• “You are a physician working in an ER in a
tertiary care center…”
• 8 Critically ill patients of different ages with
outcomes explicitly described; all arrive and
– need immediate intubation + resuscitation
– Have potential serious neurological
consequences
– Are currently incompetent, no known
preferences for adults
8 patients
•
•
•
•
4 patients have a 50% survival. If they survive,
50% will be “normal” and 25% will have severe
disability:
24 week preterm
Term baby with an AV brain malformation
2 month with meningitis
50 y old trauma victim
2 pts have a 5% survival:
• 14 y with leukemia with a 20% risk of severe
disability if survives
• 35 y with brain cancer; needs surgery, radiotherapy
and chemotherapy to survive with 100% risk of
disability if operated
2 pts have preexisting disabilities and a 50%
survival. If survival: 50% risk of further impairments
• 7 y old: cerebral palsy, deafness, hyperactivity,
learning disability with a new head trauma; (“bad
outcome” ex-preterm)
• Demented 80 y old with a new stroke
Results:
Physician responses, n = 2230
70% neonatologists
Response rate = 66%
Do you think resuscitation is in the
patient’s best interest?
100
90
80
70
60
50
40
30
20
10
0
24wk
term
2mth
7y
13y
35y
50y
80y
Would you accept to not resuscitate at the
family’s demand and give comfort care?
90
80
70
60
50
Accept WH ICU
40
30
20
10
0
24wk
term
2mth
7y
13y
35y
50y
80y
Best interest vs accepting family refusal of care
What one expects from the ethical-legal
theories
percentages
100
80
60
best int
40
accept WH care
20
0
good outcome
100% death
outcomes
Best interest vs accepting comfort care
100
90
80
70
60
Best interest
50
Accept WH ICU
40
30
20
10
0
24wk
term
2mth
7y
13y
35y
50y
80y
Best interest vs accepting comfort care
100
90
80
70
60
Best interest
50
Accept WH ICU
40
30
20
10
0
24wk
term
2mth
7y
13y
35y
50y
80y
In what order would you intubate
patients if all needed intervention at
the same time (average ranking)?
TRUE IN EVERY COUNTRY
1st: 2 months (meningitis)
2nd: 7 y (multiple severe disabilities)
8th: 80y (dementia and a new stroke)
First and second positions
1600
1400
1200
1000
800
600
400
200
0
24
term
2m
7y
13y
35y
50y
80y
It seems…
• The majority think resuscitation is in the best
interest of sick neonates
• YET, the majority would accept not
intervening for neonates than for patients
with similar or much worse outcomes
• The best interest principle is not used for
neonatal resuscitation in many culturally
different countries.
Beneficial, “grey zone” or futile?
• Women 23 week gestation (as per dates),
EFW: 500g, imminent delivery
• Wants to do “everything” for baby
• Do you think it is in this baby’s best interest to
receive intervention?
• Would you offer intervention?
• Would you obtain a court order to withhold
intervention?
Beneficial, “grey zone” or futile?
• 40y old women, married, works with disabled
children. Infertility treatment x 7 years. 5 trials
of IVF, Second mortgage on her house
• IVF GA = 22 5/7 week gestation, EFW: 500g
• Wants to do “everything” for baby
• Do you think it is in this baby’s best interest to
receive intervention?
• Would you offer intervention?
• Would you obtain a court order to withhold
intervention?
Do parental demographics influence our
attitudes towards intervention?
Marcello, Janvier, in press Pediatrics
• 850 surveys, 78% resp rate
– Montreal, Delaware, Philadelphia
• Described outcomes at 22-23wks, 24wks, and
27-28wks
Do parental demographics influence our
attitudes towards intervention?
Marcello, Janvier, in press Pediatrics
3 scenarios
• 16 yo female with two previous abortions,
accidental pregnancy. (TEEN)
• 30 yo couple who are trial lawyers, no fertility
problems. (LAW)
• 40 yo couple with their 5 IVF attempt. Mother
works as a caregiver of handicapped children.
(IVF)
Do parental demographics influence our
attitudes towards intervention?
Marcello, Janvier, in press Pediatrics
• More would comply with parental wishes at
22-23 weeks and 24 weeks for the IVF > law>
teen mother
• More would obtain a court order to give
comfort care for the 22-23 week preterm of
the teen mother than for the other parents
Following the preterm’s best
interests?
• 22weeks: best interests estimates are lowest (23%)
– Yet 68% would intervene at parental demand… if mother
is old and had IVF
• 24 weeks: best interest estimates are higher (60%)
– But 70% would accept comfort care at parental demand.
• 27 weeks: best interest estimates are highest (95%)
– Nonetheless, 30% of respondents would accept to give
comfort care at the family’s demand
Cost consideration
Stolz et al, Pediatrics, 1998
Nonsurvivors cost little
• Denying care to <500g would save 0.8% of
the NICU budget
• <600g = 3.2%, <700 = 10.3%
Why are NICUs “relatively cheap”?
• Babies die quickly: 70% deaths in 1st 5 days
(Meadow et al; Lucey et al)
• approximately 9 of every 10 NICU beds are
devoted to babies who survive (adults 4/10
ICU beds)
• Most survivors will be productive citizens:
– costs per quality adjusted life year saved being
about 1/100 of the cost of acute adult coronary
care (Doyle et al)
Beneficial, “grey zone” or futile?
• Women 23 week gestation (as per dates),
EFW: 500g, imminent delivery
• Wants to do “everything” for baby
• Do you think it is in this baby’s best interest to
receive intervention?
• Would you offer intervention?
• Would you obtain a court order to withhold
intervention?
Conclusions
• Physicians, nurses and residents are not
well informed about outcomes of
premature infants: they underestimate
survival and overestimate disability
• If they knew the outcomes, physicians
might intervene more
• What is described as being « beneficial »,
« futile », « grey zone » is relative and
subjective.
Conclusion
• QOL of ex-preterm infants is hard to
differenciate from controls.
– Outcomes also improve over time
• NICUs are cost effective
• Despite having similar or better outcomes,
neonates are often ranked after older
patients with similar or worse outcomes
Conclusion: Best interests
• Respondents seem to follow the family’s best
interest more than the child’s
– Unless the mother is a young poor single mother
– Specially if the baby is « precious »
• The best interest principle is not applied when
considering life and death decisions for neonates,
particularly for the extreme preterm
Devaluation the the newborn, why might
this be so? See Dr Dupont’s poster
• Lack of personhood / experience / attachment ? Still
considered a reproductive choice?
– Abortions may be performed at similar gestational ages
– Do we react differently to their death?
• Decreased sense of duty than older individuals?
• Media: focus on neonatal disasters and miracles
• Easier to discriminate against possibility of handicap
than confirmed handicap?
• Cultural, anthropological, sociological, evolutionary?
– Were we “built” this way?
“You're lovely, but you're empty," says the little Prince to
roses on the earth. “You don’t mean anything. Nobody
has tamed you yet. At first, my rose was like all of you.
But now my rose, all on her own, is more important than
you altogether, since she's the one I've watered. Since
she's the one I put under glass. Since she's the one I
sheltered behind a screen. Since she's the one for whom I
killed the caterpillars (except for two or three for
butterflies). Since she’s the one I listened to when she
complained, or when she boasted, or even sometimes
when she said nothing at all.
She is important because of all the time I spent with her."
Antoine de Saint-Exupéry, Le Petit Prince
Des questions?
Ethical confrontation:
During your practice, were you confronted
with resuscitating / treating with important
support / caring for extreme preterms
against your moral beliefs and values
(having « no choice » to do it as resident /
nurse)?
– Choices: Always, generally, exceptionally, never
Ethical confrontation:
Frequent ethical confrontation
• Ethical confrontation is frequently experienced by
residents and nurses in the NICU
– Nurses = 35%
– Residents = 19%
Not associated with ethical confrontation:
• Level of training, years of practice
• sex, age
• Children +/-
Ethical confrontation and center (p<0.05)
Center
frequ conf
Nurse CH (%)
56%
Nurse MAT (%)
24
Nurse DR (%)
22
PED A
28
PED B
22
PED C
0
PED D
36
OB A
6
OB B
6
OB C
0
OB D
25
Ethical confrontation, CP rates and centers
C e nt e r
F re que nt
e t hic a l
c o nf ro nt
Wro ng
e s t im a t e o f
C P >2 5 %
N urs e C H ( %)
56%
93%
N urs e M A T ( %)
N urs e D R ( %)
P ED A
P ED B
P ED C
P ED D
OB A
OB B
OB C
OB D
24
22
28
22
0
36
6
6
0
25
36
54
48
83
79
41
50
61
36
31
• Residents training in centres where resuscitation is
uncommon at 23 and 24 weeks will have higher
thresholds for resuc and less ethical confrontation
• Residents who overestimated outcomes were less
likely to have ethical confrontations (p<0.05)
• Nurses who incorrectly thought CP rates were high
were more likely to have ethical confrontations
(p<0.05)
• Nurses working in a unit with a high incidence of
complications have profoundly different views on the
ethics of resuscitation
• WE GET USED TO WHAT WE SEE
• Our experience shapes our conception of
what is normal or not
•  WHAT GOES ON IN OUR ENVIRONMENT
IS NORMAL (ie ethical?)
How do we take clinical decisions?
• What is legal?
• What others around recommend
– local expert, statements from associations, evidence…
• What is done around us: local experts, what is done
where we are (ethical microcosms)
• What we learned through formal process
• Previous experience: Instinct, intuitions, emotions
• Personality
• What is best for the community: cost consideration
Birth Demographics
ELBW (149)
NBW (133)
Gestation (wks), Mean (SD)
27 ( 2.3)
Full-term
Birthweight (g), Mean (SD)
841 (124)
3384 (487)
BW <750, n (%)
27
--
SGA <10th percentile, n (%)
24
3
Neurosensory Imp, n (%)
27
2
Gender: Male, n (%)
45
45
Caucasian, n %
94
97
Would you comply with parents request for
intervention?
224/7 weeks
P<0.0001
% response affirmative
70
P<0.0001
P<0.0004
60
50
40
30
*
20
10
0
TEEN
LAW
IVF
Would you comply with parents request for
intervention?
24 weeks
% response affimative
P<0.001
90
88
86
84
82
80
78
76
74
72
70
P<0.04
P<0.0001
*
TEEN
LAW
IVF
If you disagree with parent’s decision –
Would you obtain a court order to withhold treatment?
224/7 weeks
P<0.00
2
P<0.00
1
% response affirmative
35
P<0.02
30
25
20
*
15
*
10
5
0
TEEN
LAW
IVF
How do babies die compared to older
children?
• Most pediatric deaths occur in ICUs
• NICU: neonatal intensive care
• PICU: pediatric intensive care
• Many of them occur after life sustaining
treatments (LST) are withdrawn (WD) or
withheld (WH)
There are 4 ways to die in ICUs
• Patients who die «no matter what »: unstable
physiology
– With CPR
– On the respirator, without CPR
– After WH/WD Life sustaining therapies
« because patient is dying »
• Patients who died and might have survived
– LST are WH/WD because of quality of life
(QOL) considerations
Survival: what kind of death?
• Most studies do not make the distinction between
• WD/WH interventions from dying children and
• WD/WH interventions from physiologically
stable children for quality-of-life (QOL)
considerations.
Survival in ICU: can depend on the
attitude of the physician
• This is an important distinction, as children who
were stable might have lived if interventions has
not been withdrawn, or withheld.
• Ethically: There is a wide variation / comfort
(physicians, hospitals, countries) when it comes to
WH/WD interventions for children who are not
actively dying
• Important when comparing outcomes between
ICUs and describing survival
Review of all patient deaths over 2y
Janvier et al in review
214 deaths
–NICU = 77
–PICU = 68
–Delivery room = 22
–Outside of DR-NICU-PICU = 47
• Floors> home> hospice > ER
Modes of death: PICU vs NICU
PICU
vs
NICU
PICU
NICU
6%
7%
NS
Died on a respirator
(no WD, WH CPR)
51%
5%
p<0.05
WH / WD because dying
27%
35%
NS
WH / WD for QOL (stable)
16%
53%
p<0.05
Mode of death
Died with CPR
(no WH + no WD)
n = 68
n = 77
Modes of death: PICU vs NICU
PICU
vs
NICU
PICU
NICU
6%
7%
NS
Died on a respirator
(no WD, WH CPR)
51%
5%
p<0.05
WH / WD because dying
27%
35%
NS
WH / WD for QOL (stable)
16%
53%
p<0.05
Mode of death
Died with CPR
(no WH + no WD)
n = 68
n = 77
Modes of death: PICU vs NICU
PICU
vs
NICU
PICU
NICU
6%
7%
NS
Died on a respirator
(no WD, WH CPR)
51%
5%
p<0.05
WH / WD because dying
27%
35%
NS
WH / WD for QOL (stable)
16%
53%
p<0.05
Mode of death
Died with CPR
(no WH + no WD)
n = 68
n = 77
Pts who died despite maximal therapy
• Much more common in the PICU (82%) than the
NICU (47%)
• Many of these PICU deaths were not
unexpected: the majority of these deaths
occurred in children with serious preexisting
medical conditions
WD/WH interventions for QOL reasons (pts
who might have lived if ICU had been
continued)
• The most common mode of death in the NICU
• In the PICU “stable QOL” category, all patients
had serious confirmed disability (could not
walk, talk, eat independently).
• In the NICU, 60% of “QOL stable babies” had
only a RISK of long term disability: (grade 3 or 4
IVH in preterm, pall care for HLHS)
Delivery room deaths, n= 22
(all babies > 22wks and > 400g)
• Median GA = 24wks; BW = 685g
• 23% had confirmed severe disabilities
• 60% of babies had a calculated survival > 50%
with a RISK of disability
• Using Tyson’s “Preemie calculator”, avg.
survival = 61%, and if survival ~ 50% normal
outcomes
Modes of death for neonates and older
pediatric patients are strikingly different
• For neonates, there is a greater proportion of:
– Deaths following withdrawal of life sustaining
therapies with stable physiology for a risk of
disability (patients who might have lived with CP)
– Withholding ICU admission with good chance of
survival (> 50%) and a risk of disability,
which suggests that...
1. Neonatologists & parents (and
obstetricians?) are more inclined to
WD/WH life sustaining treatments
-before babies become unstable?
-and / or when there is only a risk of
disability?
2. PICU physicians & parents are more
inclined to WD/WH life sustaining treatments
when older children become unstable, but
not earlier?
3. Parents & physicians are more willing to
WD/WH life sustaining treatments for
neonates for QOL considerations in babies
who might live as compared with older
children?