Title: The Intravenous Immunoglobulins: Current and Future Role in the NICU Could IVIG be neuro-protective?
1The Intravenous Immunoglobulins Current and
Future Role in the NICU Could IVIG be
neuro-protective?
- William Tarnow-Mordi
- Professor of Neonatal Medicine
- University of Sydney
- Westmead Hospital and The Childrens Hospital at
Westmead
2OFF LABEL DISCLOSURE
- William Tarnow-Mordi has documented that his
presentation involves comments or discussion of
unapproved or off-label, experimental or
investigational use of - (a) polyclonal intravenous immunoglobulin (IVIG)
- (b) Veronate (anti-staphylococcal
immunoglobulin)
3DISCLOSURE STATEMENT
-
- Dr. William Tarnow-Mordi has documented that he
has nothing else to disclose.
4MISSION STATEMENT
- The primary aim of clinical studies in the
newborn is to identify interventions that
increase disability-free survival, employing the
most reliable evidence from RCTs.
5 6Current role of IVIG
- (a) Immunomodulatory properties of IVIG
- (b) Thrombocytopenia
- (b) Haemolytic disease
- (c) Infection
7Future potential role of IVIG
-
- Is IVIG neuro-protective in newborns with, or at
risk of, infection?
8Future role of IVIG
- Do neonatal infections remote from the brain,
including CONS, cause brain damage? - Are white matter damage and periventricular
leukomalacia partly immune-mediated?
9- 3. Is multiple sclerosis (and its animal model) a
model for PVL? - 4. Can IVIG promote re-myelination in multiple
sclerosis and other CNS inflammatory states? - Can we do trials big enough to test reliably if
- IVIG is neuro-protective in newborns with, or at
risk of, infection?
10Current role of IVIG
- (a) Immunomodulatory properties of IVIG
- (b) Thrombocytopenia
- (b) Haemolytic disease
- (c) Infection
11(No Transcript)
12(No Transcript)
13Pro-inflammatory
- opsonic activity
- fixation of complement
- antibody dependent cytotoxicity
- neutrophil chemiluminescence
- phagocytosis
- release of stored neutrophils
14The phagocytes wont eat the microbes unless the
microbes are nicely buttered for them. Well, the
patient manufactures the butter himself alright,
but my discovery is that the manufacture of that
butter, which I call OPSONIN, goes in the system
by ups and down.There is at the bottom only one
genuinely scientific treatment for all diseases,
and that is to stimulate phagocytosis.
Dr RidgeonThe Doctors DilemmaG B Shaw
1909
15Anti-inflammatory
- Down-regulation of inflammatory cytokines via
- Fc receptor blockade,
- provision of anti-idiotype antibodies
- interference with activation of
- T-cells
- B-cells
- the cytokine network
- complement
- Immunomodulation of autoimmune and inflammatory
diseases with intravenous immune globulin. - Kazatchkine MD,. et al. N Engl J Med 2001
16(No Transcript)
17Current role of IVIG
- (a) Immunomodulatory properties of IVIG
- (b) Thrombocytopenia
- (b) Haemolytic disease
- (c) Infection
18 Neonatal alloimmune/ autoimmune thrombocytopenia
- Incidence 1 per 1000
- First reports of IVIG in 1980s
- No RCTs
-
19Immune-mediated thrombocytopenia (Idiopathic
thrombocytopenia - ITP)
- Extremely rare in newborns
- First reports of IVIG in 1981
- Subsequently effective in RCTs
- FDA approved
20Current role of IVIG
- (a) Immunomodulatory properties of IVIG
- (b) Thrombocytopenia
- (b) Haemolytic disease
- (c) Infection
21IVIG for Rhesus disease and ABO
incompatibilityTwo systematic reviews of the
same data
- 1. Alcock GS, Liley H. Cochrane Review 2002
- 2. Gottstein R, Cooke RW. Arch Dis Child Fetal
Neonatal 2003 - In 3 RCTs in 199 infants, IVIG led to
- Fewer exchange transfusions
- Relative Risk 0.28 (0.17 0.47)
- Shorter phototherapy and hospital stay
- No data on disability-free survival
22Reduced need for exchange transfusion with IVIG
vs standard treatment for haemolytic jaundice
Number needed to treat 2.7 (95 CI 2.0 to 3.8).
23Similar results differing conclusions
-
- Alcock GS, Liley H. Cochrane Review 2002
- Well designed studies are needed before routine
use of IVIG can be recommended. - Gottstein R, Cooke RWI. Arch Dis Child Fetal
Neonatal 2003 - IVIG is an effective treatment.
24Mechanisms of action of IVIG in ITP and
haemolytic jaundice
- Similar mechanisms are postulated for effects of
IVIG in both conditions. - They are incompletely understood.
25Possible mechanisms of IVIG in ITP
Theory 1 IVIG (Fab portion fork end) binds
sensitised red cells. IVIG Fc portion (fork
handle) blocks the Fc receptor on macrophages
Macrophage
Fehr et al, 1982
26Possible mechanisms of IVIG in ITP
Theory 2 IVIG upregulates inhibitory Fc?RIIB
receptors on macrophages. This inhibits
phagocytosis.
Fc?RIII or IIA
Plat
Macrophage
?
? Fc?RIIB expression
Samuelsson et al, 2001
27Current role of IVIG
- (a) Immunomodulatory properties of IVIG
- (b) Thrombocytopenia
- (b) Haemolytic disease
- (c) Infection
28(c) Infection
- Therapeutic IVIG
- Prophylactic polyclonal IVIG
- Prophylactic hyper-immune IVIG (Veronate) to
prevent staphylococcal infection
29Therapeutic IVIG
- Two Cochrane reviews
- IVIG for treating sepsis and septic shock
- Alejandria MM, Lansang MA, Dans LF, Mantaring
JBV. - 2. Polyclonal IVIG for suspected or subsequently
proven infection in neonates - Ohlsson A, Lacy JB.
301. IVIG for treating sepsis and septic shock
Alejandria MM, Lansang MA, Dans LF, Mantaring
JBV.
- Polyclonal IVIG reduces mortality in all age
groups combined - n492
- Relative Risk 0.64 (0.64 0.80)
- p 0.00009
31 1. IVIG for treating sepsis and septic shock
Alejandria MM, Lansang MA, Dans LF, Mantaring
JBV.
321. IVIG for treating sepsis and septic shock
Alejandria MM, Lansang MA, Dans LF, Mantaring
JBV.
- No mortality reduction with anti-cytokine or
monoclonal IVIG - Anti-cytokine IVIG (4 RCTs, n 4,318)
- RR0.93 95 CI 0.86 to 1.01) NS
- Monoclonal IVIG (5 RCTs, n 2,826)
- RR0.97 (95 CI 0.88 to 1.07) NS
332. Polyclonal IVIG for suspected or subsequently
proven infection in neonatesOhlsson A, Lacy JB.
Cochrane Review 2004
-
- mortality reduced in suspected infection, but
with borderline statistical significance - n 318 Relative Risk 0.63 (95 CI 0.40 -
1.00) p 0.05 -
- mortality reduced in subsequently proved
infection, but with wide confidence interval - n 262 Relative Risk 0.55 (95 CI 0.31, 0.98)
p 0.04
342. Polyclonal IVIG for suspected or subsequently
proven infection in neonatesOhlsson A, Lacy JB.
Cochrane Review 2004
- Insufficient evidence for routine
IVIG.Well-designed trials to assess long term
disability and cost effectiveness are needed. - One such trial is the International Neonatal
Immunotherapy Study.
35(c) Infection
- Therapeutic IVIG
- Prophylactic polyclonal IVIG
- Prophylactic hyper-immune IVIG (Veronate) to
prevent staphylococcal infection
36Intravenous immunoglobulin for preventing
infection in preterm and/or low-birth-weight
infants
-
- Ohlsson A, Lacy JB. Cochrane Review 2004.
-
- 16 RCTs of IVIG versus placebo or no intervention
-
- 5,000 infants lt 37 weeks gestation or lt2500 g
- 3-4 reduction in sepsis or serious infection
-
-
37Cumulative meta-analysis IVIG vs Placebo or No
Treatment
Relative Risk 0.82 (95 CI 0.74-0.92) Risk
Difference 0.04 (95 CI 0.02, 0.06)
38Polyclonal IVIG prophylaxis
- Reduces infection by 3-4
- statistically significant
- marginal clinical significance
- No reductions in mortality, or
- NEC, IVH
- hospital stay
- No short term serious side effects
- Inconclusive on disability-free survival
39Conclusions
- No further RCTs of standard IVIG to prevent
infections - Basic scientists and clinicians should pursue
other avenues
40(c) Infection
- Therapeutic IVIG
- Prophylactic polyclonal IVIG
- Prophylactic hyper-immune IVIG (Veronate) to
prevent staphylococcal infection
41Veronatedata courtesy of Dr Seth Hetherington
- Intravenous Immune globulin
- Donors selected for high titers to
- Clumping factor A S. aureus
- Serine-aspartate repeat G protein S.
epidermidis - Antibodies block attachment of bacteria to
fibrinogen
42Prophylactic hyperimmune anti-staphylococcal IVIG
(Veronate) 2017 infants Bwt 500- 1250 g
- Blood Stream Infection Control
Veronate n 989 n 994 -
- Staph aureus 5
6Probable CONS 16
15Definitive CONS 9
11Any Infection 35 35 Mortality
7 6 -
43Veronate
- No difference was statistically significant
- Including these data may not alter the result of
the meta-analysis of prophylactic trials - 3-4 reduction in sepsis with prophylactic IVIG
- Too early for data on disability-free survival
44Current use of IVIG in the NICU
- Based on promising, but incomplete, evidence of
effectiveness in - Thrombocytopenia
- Haemolytic jaundice
- Neonatal sepsis
- with no evidence of its long-term effects.
45Future role of IVIG
- Do neonatal infections remote from the brain,
including CONS, cause brain damage? - Are white matter damage and periventricular
leukomalacia partly immune-mediated?
46(No Transcript)
47Leviton, Gillies, Neff, Yaney 1976
- White matter damage, or perinatal telencephalic
leuco-encephalopathy (PTL), was more common after
gram negative septicaemia. - It is hypothesized that endotoxin from
bacteria adversely affects developing white
matter
48Stoll et al (JAMA) 2004
49Stoll, Hansen, Adams-Chapman, Fanaroff, Hintz,
Vohr, Higgins, for the NICHD Neonatal Research
Network, JAMA 2004
- 6093 infants lt 1000 g bwt assessed at 18 22
months corrected for gestation
50Neuro-developmental Impairment (NDI) in infected
versus uninfected infants
Category N () NDI (unadjusted) OR for NDI adjusted for 21 risk factors
uninfected 2161 (35) 29 -
Culture neg clinical infection 1538 (25) 43 1.3 (1.1-1.6)
Culture positive sepsis 1922 (32) 48 1.5 (1.2 1.7)
Sepsis and NEC 279 (5) 53 1.8 (1.4 2.5)
meningitis 193 (3) 48 1.6 (1.1 2.3)
51Neuro-developmental impairment in infants with
different pathogens versus uninfected infants
Category N NDI (unadjusted) Adjusted OR for NDI relative to uninfected
uninfected 1976 29 -
CONS 853 44 1.3 (1.1 1.6)
Other Gm pos 256 48 1.7 (1.2 2.3)
Gram negative 185 45 1.8 (1.2 7.6)
Fungal 96 57 1.4 (0.9 2.2)
52After adjustment for antenatal, perinatal and
postnatal factors
- Neonatal infection, including CONS and
culture-negative sepsis, was associated with
increases of 30 80 in the odds of - poor head growth (lt 10th centile) at
- 36 weeks
- 18-22 months
- neuro-developmental impairment at 18-22 months
53- CONS
- although less clinically severe
- may be associated with more disability than any
other pathogen (n853) - Culture-negative clinical infection
- potentially more important than CONS in its
association with disability (n1538)
54Association or causality? RCTs are needed to
prove cause and effect
- Neonatal infection may
- cause neuro-developmental impairment, or
- share a common cause with NDI
- If RCTs (e.g. of IVIG) in neonatal infection
- show a reduction in disability
- this would confirm infection as causal.
55Association between cerebral palsy and
coagulase-negative staphylococci.Mittendorf et
al. Lancet 1999
- Cultured amnion-chorion space in 107 preterm
infants. - 35 grew no organisms. 28 grew CONS
- CONS isolated in 4/5 (80) infants who later
manifest CP vs 26/102 (25) who did not. (plt0.02)
56Mittendorf et al. Lancet 1999
- Small study, possible chance association
- Consistent with hypotheses that
- CONS may be causal in Cerebral Palsy
- mediated by virulence factors such as
haemolysins, deoxyribonuclease, slime and adhesins
57Future role of IVIG
- Do neonatal infections remote from the brain,
including CONS, cause brain damage? - Are white matter damage and periventricular
leukomalacia partly immune-mediated?
58Zupan et al, 1996 Volpe, 1997,2001
- Focal, cystic PVL
- more severe, becoming less frequent
- progressive loss of all cells in white matter.
- Diffuse PVL
- less severe, increasingly frequent
- mainly affects developing oligodendrocytes
59 Nitrosative and oxidative injury to
premyelinating oligodendrocytes in PVLHaynes RL,
et al. J Neuropathol Exp Neurol 2003
- Autopsied 17 PVL cases and 28 non-PVL controls.
- Diffuse PVL involves injury to premyelinating
oligodendrocytes through - activation of microglia
- release of reactive oxygen and nitrogen species
60Infection remote from the brain, neonatal white
matter damage, and cerebral palsy in the preterm
infant.Damman O, Leviton A. 1998
-
- Remote infection could
-
- lead to an inflammatory response that may be
responsible for initiating and prolonging tissue
injury - induce white matter damage by molecular
mechanisms, especially cytokines. - .
61(No Transcript)
62(No Transcript)
63 Fetal / neonatal cytokinemia and white cell
activation in WMD Duggan et al, Lancet 2001
- 50 infants 23-29 weeks gestation cord blood
- - for cytokines
- - CD45RO T lymphocytes (to assess T cell
activation) - Brain MRI done at median 2 days after birth
- 18 (36) had cerebral lesions associated with
- ? cytokines and activated T cells
-
-
-
-
64(No Transcript)
65(No Transcript)
66Duggan et al, Lancet 2001Immune cells might be
directly involved in injury
-
- Activated CD45RO T lymphocytes express
- activation, homing and adhesion molecules
- migrate into tissues
- induce apoptosis
67Can a 500g baby mount an active inflammatory
response ?
68YES, plasma TNF level gt 1000 pg/ml
69The Toll-like receptor TLR4 is necessary for LPS
induced oligodendrocyte injury in the CNS
Lehnardt et al. J Neuroscience 2002
- To exert its pathogenic effect, LPS requires
TLR4, a receptor present on -
- circulating monocytes/ macrophages
- other systemic immune cells
- microglia in the CNS.
70LPS (endotoxin) kills oligodendrocytes only when
microglia are present in culture
71The Toll-Like Receptor TLR4 Is Necessary for
Lipopolysaccharide- induced Oligodendrocyte
Injury in CNS Lehnardt et al 2002
- Provides a mechanism of action between
-
- (1) LPS (and other microbial antigens)
- ?
- (2) induction of innate, non-specific immunity
- ?
- (2) injury to oligodendrocytes and myelin
- (as in PVL and multiple sclerosis).
72- 3. Is multiple sclerosis (and its animal model) a
model for PVL? - 4. Can IVIG promote re-myelination in multiple
sclerosis and other CNS inflammatory states? - Can we do trials big enough to test reliably if
- IVIG is neuro-protective in newborns with, or at
risk of, infection?
73Histological appearance of Multiple Sclerosis
- Demyelination of nerve axons, mostly in white
matter, with acute focal inflammatory changes. - Lesions include
- axonal loss
- spontaneous re-myelination
- multifocal sclerotic plaques.
74Spontaneous remyelination in Multiple Sclerosis
Perier O, Gregoire A. Brain 1965.
75Premyelinating oligodendrocytes in chronic
lesions of multiple sclerosisChang et al NEJM
2002
- Oligodendrocyte progenitor cells characterized
in - developing brain
- normal adult human brain
- chronic lesions of multiple sclerosis.
76Spinal cord axons in the healthy mouse
-
- Normally myelinated axons.
-
-
77Spinal cord axons after infection with Theilers
murine encephalo-myelitis virus (TMEV)
- Extensive demyelination
-
- Some spontaneous remyelination.
-
- Macrophages/ microglia with ingested myelin
debris (arrows). -
-
78Recruitment of T lymphocytes across Blood Brain
Barrier endothelium in MS and TME Virus
infectionEngelhardt J Neural Transmission 2006
79- 3. Is multiple sclerosis (and its animal model) a
model for PVL? - 4. Can IVIG promote re-myelination in multiple
sclerosis and other CNS inflammatory states? - Can we do trials big enough to test reliably if
- IVIG is neuro-protective in newborns with, or at
risk of, infection?
80Remyelination by oligodendrocytes stimulated by
antiserum to spinal cord. Rodriguez, M. J
Neuropathol Exp Neurol. 1987
- Produced anti-serum from mice immunized with
homogenized spinal cord (SCH). - Anti-serum increased remyelination 10 fold in
spinal cords after infection by TME virus.
81Human monoclonal antibodies reactive
tooligodendrocytes promote remyelination in a
model of multiple sclerosisWarrington et al PNAS
2000
- Experiments with human remyelination-promoting
mAbs in mice using normal human IVIG (IgG) as
control -
82Spinal cord axons in the healthy mouse
-
- Normally myelinated axons not infected with
TMEV. -
-
83Spinal cord axons after chronic infection with
Theilers murine encephalo-myelitis virus
- Extensive demyelination
-
- Some spontaneous remyelination.
-
- Macrophages/ microglia with ingested myelin
debris (arrows). -
-
84Partial remyelination after infection with TME
virus treatment with monoclonal IgM
-
- Significantly greater remyelination after
treatment with monoclonal IgM - Same magnification
85Effect of IVIG on remyelination in mouse spinal
cord after TME Virus infection Warrington et al
PNAS 2000
Treatment Group Oligodendrocyte remyelination P value
Controls 6.7 -
After human IVIG IgG 14.2 lt0.05
After Human IgM mAB 23.2 lt0.001
86Human monoclonal antibodies reactive
tooligodendrocytes promote remyelination in a
model of multiple sclerosisWarrington et al PNAS
2000
-
- OL remyelination in mice was increased
-
- twofold with human IgG
- fourfold with human IgM.
87Human monoclonal antibodies reactive
tooligodendrocytes promote remyelination in a
model of multiple sclerosisWarrington et al PNAS
2000
- Human remyelination-promoting mAbs
-
- may be a simple, effective therapy.
- can be produced free of infectious agents
- may alleviate the high cost of IVIg.
- may simplify investigation of mechanisms of
immunomodulatory therapies.
88Anti-apoptotic signaling by a remyelination-promot
ing human antimyelin antibodyHowe CL Neurobiol
Dis 2004
-
- In mice with TMEV, a recombinant hman monoclonal
anti-myelin antibody - induced anti-apoptotic signaling in
premyelinating oligodendrocytes and - reduced caspase-3 activation and caspase gene
expression -
89Anti-apoptotic signaling by a remyelination-promot
ing human antimyelin antibody Howe CL,
Neurobiology of Disease 2004
- Polyreactive autoantibodies that trigger repair
within demyelinated lesions marked an important
shift in our understanding of the role of
antibodies in the CNS. - Recombinant monoclonal human IgM autoantibody
rescued a pre-myelinating oligodendrocyte cell
line from death induced by either hydrogen
peroxide or TNF-a.
90Anti-apoptotic signaling by a remyelination-promot
ing human antimyelin antibody Howe CL,
Neurobiology of Disease 2004
- One mechanism by which IVIG may promote
remyelination in CNS inflammatory conditions is
by -
- binding oligodendrocyte progenitors and
- preventing them from undergoing death induced by
H2O2 of TNFa.
91CD54 lymphocytes expressing ICAM-1 in blood from
patients before and after IVIG Créange et al. J
Neuroimmunol 2003
92Cytokine expression by CSF monocytes in patients
with Post Polio Syndrome after IVIG.Gonzalez J
Neuroimmunol 2004
-
- ? TNF-a (p lt 0.05)
-
- ? IFN-? (p lt 0.001)
93- Is clinical evidence consistent with in vitro and
in vivo laboratory evidence?
94Cochrane Reviews of IVIG in neurological
inflammatory conditions
Condition No. of patients Effect
Multiple sclerosis 168 ? relapses (by 38-74)
Guillain Barre syndrome 75 speeds recovery
Multifocal neuropathy 34 ? strength
Myasthenia Gravis 60 inconclusive
Chronic demyelinating inflammatory polyneuropathy 170 ? disability
95IVIG for the treatment of relapsing-remitting
multiple sclerosis a meta analysis.Sorensen PS,
Fazekas F, Lee M. Eur J Neurol 2002
96Achiron et al Arch Neurol 2004
97Achiron et al Arch Neurol 2004
- Placebo controlled RCT of IVIG vs saline
- within 6 weeks of first signs of MS in 91
patients - 36 lower probability of developing clinically
definite multiple sclerosis -
- Risk ratio, 0.36
- 95 CI 0.15- 0.88
- p 0.03.
-
98Achiron et al Arch Neurol 2004
- The IVIG group had fewer MRI lesions at 12
months - reduced volume T2-weighted (P 0.01)
- fewer T2 weighted (P 0.01) of
- reduced volume of gadolinium enhancing lesions
(P0.03) .
99Achiron et al Arch Neurol 2004
-
- Future studies evaluating the combined effects
of - IVIG
- interferons
- glatiramer acetate
- may offer additional benefit on the occurrence
of second attack.
100- These studies suggest that, in animals and adult
humans, IVIG can reduce cerebral inflammation and
ameliorate pre-existing cerebral lesions.
101- 3. Is multiple sclerosis a model for PVL?
- 4. Can IVIG promote re-myelination in multiple
sclerosis and other CNS inflammatory states? - Can we do trials big enough to test reliably if
- IVIG is neuro-protective in newborns with, or at
risk of, infection?
102An ideal RCT to test if IVIG increases
disability-free survival should
- randomise neonates to prophylactic, repeated IVIG
or placebo. - be large enough to detect realistically moderate
differences with adequate power - include long term follow up of
- neuro-developmental and cognitive outcomes.
103International Neonatal Immunotherapy Study, INIS
- Google npeu inis
- www.npeu.ox.ac.uk/INIS.htm
- (downloadable copy of this talk will be
available)
104(No Transcript)
105International Neonatal Immunotherapy Study, INIS
-
- A randomised placebo controlled trial of IVIG in
proven or suspected neonatal sepsis - Revised sample size 4,000
- to yield 90 power to detect a 5 risk difference
in death or major disability -
106International Neonatal Immunotherapy Study, INIS
-
- 2 doses of 500 mg/ kg (total dose of 1 g/kg )
- as an adjunct to antibiotic treatment
-
- It does not therefore fulfil all the criteria
for an ideal test of these hypotheses. -
107However,
- Two multi-centre trials that provided more
sustained doses of IVIG, for 2 weeks of more,
have already been done, in a total of over 4,000
patients.
108(No Transcript)
109- 2416 infants lt 72 hr after birth
- stratified according to birth weight
- (501-1000 g and 1001-1500 g)
- randomly assigned to fortnightly
- prophylactic IVIG (n 1204) or
- control (n 1212).
110- Prophylactic IVIG failed to reduce the incidence
of hospital-acquired infections. - These were 208 (17.3) in the IVIG group and 231
(19.1) in the control group - (Relative Risk, 0.91 95 percent confidence
interval, 0.77 to 1.08). -
111(No Transcript)
112(No Transcript)
113Prophylactic hyperimmune anti-staphylococcal IVIG
(Veronate) 2017 infants Bwt 500- 1250 g
- Blood Stream Infection Control
Veronate n 989 n 994 -
- Staph aureus 5
6Probable CONS 16
15Definitive CONS 9
11Any Infection 35 35 Mortality
7 6 -
114Speculation
- It would be valuable to determine the rates of
neuro-developmental impairment in these
randomised cohort, or subsets within them - To test the hypothesis that sustained IVIG
therapy is (a) safe, (b) neuro-protective in
neonates at risk of sepsis. -
-
115Conclusions
- Modulation of inflammation is a relatively new
strategy for preventing or treating neurological
injury. -
- New evidence is needed to evaluate IVIG as a
non-steroidal anti-inflammatory agent that may
reduce neonatal mortality and morbidity.
116Sir Richard Doll and Sir Richard Peto
117Sir Richard Doll Controlled trials the
1948 watershed BMJ 1998 317 1217-20
- Early randomised trials (1940s) can properly be
criticised on the grounds that they were often
too small to have any chance of detecting
moderate effects. - Small trials can be successful when the effect
is large but this seldom occurs.
118- Only in the 1980s did it become possible to
organise the groundbreaking international study
of infarct survival (ISIS) trials - in tens of thousands of patients, which showed
the value of moderate improvements in the
treatment of common diseases.
119The ISIS 1 Trial
- Atenolol infusion vs placebo after heart attack
- 16,000 patients recruited
- Showed a 2 mortality reduction- from 13 to 11
120(No Transcript)
121The ISIS 2 Trial
- Aspirin vs placebo after heart attack
- 17,000 patients recruited
- Showed a 23 mortality reduction- from c. 11 to
8.
122(No Transcript)
123- The Goddess of Chance is a capricious mistress.
Yet, with large enough numbers, she becomes our
obedient servant yielding up the truth. - Anon.
124Sample sizes to show moderate or large
differences in disability-free survival with 90
power at 2p 0.05
Adverse outcome in controls C Adverse outcome in treated group T Absolute risk difference C-T Total sample needed (both groups combined) Number needed to treat (NNT) to prevent one case
42 40 2 20, 712 50
42 39 3 9, 168 33
42 38 4 5,134 25
42 37 5 3,270 20
42 32 10 800 10
125- If newborns are to benefit from evidence as
reliable as in cardiology, - much larger numbers are needed
- i.e. thousands, not hundreds
126A challenge
- Judged by whether sufficient evidence exists to
exclude moderate benefits or harms, most current
perinatal treatments remain inadequately
evaluated.
127A solution
- Samples of thousands, not hundreds, where the
research question merits it - Multicentre, international collaboration
- Prospective meta-analysis
- Simple, inexpensive modes of follow up
128Prospective meta-analysis
- Multiple trials with same design hypotheses
- Pledged to combine their results in a single,
individual patient based meta-analysis - New trials and hypotheses can be added before any
results are known - Avoids publication bias, data driven hypotheses
- Combines the methodological advantages of a
single trial with pragmatic benefits of flexible
funding arrangements and recognition needs of
different national agencies
129International neonatal oxygen trials NeoProm
prospective meta-analysisco-ordinator Lisa Askie
- Australia 1200 NHMRC
- NZ - 300 HRC
- UK - 1200 MRC
- Canada - 1200 - CHIR
- US - 1500 NICHD US POST
- US - 1300 - NICHD SUPPORT
- Already funded. Possible total gt 6,000
130Acknowledgments
- PanelistsCarol J. Baker, Baylor College of
Medicine, Houston, TXBarbara J. Stoll, Emory
University School of Medicine, Atlanta,
GALeonard E. Weisman, Baylor College of
Medicine, Houston, TX
131Further acknowledgments
- Alan Leviton, Arne Ohlsson, Olaf Damman, Peter
Brocklehurst, Khalid Haque, David Isaacs, Wendy
Hague, John Simes, Barbara Farrell, Priya
Duggal-Beri, Ann Cust, David Henderson-Smart, Ben
Stenson, David Edwards, Ryszard Lauterbach, Seth
Hetherington, Amy Morris, Iain Chalmers, The
Cochrane Collaboration and the hundreds of
parents, babies, nursing and medical staff in
over 100 hospitals worldwide who are contributing
to the International Neonatal Immunotherapy Study
(INIS) of polyclonal IVIG.
132(No Transcript)
133William Silverman (Fumer) 1917-2004
-
- Taught his students and friends
- to consider the long term consequences of
neonatal care, for patients and families - to cultivate a habit of lifelong (un) learning
- semper plangere (always complain) - and look
for better evidence -
-