Landmark Trials Flashcards
ALPS Trial (Antenatal Betamethasone for women at risk of Late Pre-term Delivery) - Study type, publishing details, objective?
Multi-centre RCT, published 2016 in NEJM by Gyamfi-Bannerman et al.
Aim: To determine if betamethasone administered to women at risk of late PTB ↓ risk of neonatal morbidity
ALPS Trial - PICO
Population, Intervention, Control, Outcomes
Primary outcomes x 5, secondary x 4
P - 2831 women, 34+0 – 36+5, singleton pregnancy, at high risk of late preterm delivery *(>3cm dilated/75% effaced OR PPROM) *
I - 2x Betamethasone injections 24/24 apart
C - Placebo administered 24/24 apart
O - Outcomes
Primary: neonatal composite within first 72 hours:
* CPAP or HF nasal cannula for >2hrs
* Supplemental O2 with FiO2 >0.30 >4hrs
* Extracorporeal membrane oxygenation (ECMO)
* Mechanical ventilation
* Stillbirth and NND within 72hrs after delivery
Secondary outcomes:
* RDS, TTN, apnoea, BPD, surfactant
* Hypoglycaemia, feeding difficulty
* Hypothermia, NEC, IVH, NN sepsis, pneumonia
* Maternal: chorio, endometritis, delivery before completion of ACS course, LOS
ALPS Trial - Results
Primary Outcomes
↓ 20% 1° outcomes, RR 0.80, NNT = 35
↓ 35% severe resp Cx, RR 0.66, NNT = 25
Secondary Outcomes
Overall ↓ TTN, BPD, resus at birth, surfactant use
↑ 60% NN hypoglycaemia, no associated ↑ LOS - RR 1.6
No difference SCN/NICU admission
No diffference in maternal outcomes
BPD = bronchopulmonary dysplasia
ALPS Trial overall conclusions
Steroids for if risk of late preterm delivery significantly ↓ neonatal respiratory Cx and need for respiratory support
Steroids significantly ↑ neonatal hypoglycaemia but not rates of other maternal/neonatal complications
Limitations of the ALPS trial?
- Excluded multiple pregnancy and GDM, two groups which have high respiratory morbidity
- Assessed short-term outcomes only, none in impact of neonatal hypoglycaemia
ACTORDS (Neonatal respiratory distress syndrome after repeat exposure to antenatal corticosteroids) - Study type, publishing details, objective?
RCT, published 2006 in the Lancet by Crowther et al.
Aim: To establish whether repeat prenatal corticosteroids given to women at risk of PTB can reduce neonatal morbidity without harm
ACTORDS - PICO?
Population, Intervention, Control, Outcomes
Primary outcomes x 3 (neonatal), secondary x 3 (maternal)
P - 982 women at risk of PTB <32/40 with a single/twin/triplet pregnancy, ≥7 days after receiving 1st course of steroids
I - Repeat single dose of steroids every week until delivery/term
C - saline placebo
O - Primary: RDS, need for O2 or mechanical ventilation, weight/length/HC at birth and at discharge. Secondary: maternal postpartum temp > 38, chorioamnionitis, maternal injection S/Es
ACTORDS - Results
↓ 20% (RR 0.82) RDS + 40% (RR 0.6) severe lung dis + 10% (RR 0.90) O2 + duration of mechanical ventilation, NNT = 14
No significant difference in infant growth measures
↑ 13% CS, *unexplained but likely didn’t affect any results *
No ↑ maternal/infection risk
ACTORDS - Conclusions
- Exposure to **repeat doses of ACS reduces neonatal morbidity **– short-term benefits support use of repeat doses who remain at risk of very PTB ≥7 days after an initial course
- NNTB (number needed to treat to benefit) 14 to reduce RDS risk and severe lung disease
- No evidence of ↑ maternal/fetal infection risk with repeat ACS doses
- No statistically significant difference in infant growth measures
ACS = antenatal corticosteroids
ACTORDS - Limitations?
- Investigated short term effects only, ongoing inconsistencies raised by observational trials re: long term effects of steroids
- Broad range of gestational ages and treatment doses ∴ results not generalisable beyond gestational ages and doses used in the trial
ACHOIS (Effect of treatment of gestational diabetes mellitus on pregnancy outcomes) - Study type, publishing details & objective?
Multicentre (18, Australia and UK) RCT, published 2006 in NEJM by Crowther et al.
Aim: To determine whether treatment of women with GDM reduced the risk of perinatal complications
ACHOIS - PICO?
Population, Intervention, Control, Outcomes
Primary outcomes x 7
P - 1000 women, singleton or twin pregnancy, 16 – 30/40 with:
* ≥1 GDM R/Fs OR pos 50g glucose-challenge test (1hr BSL >7.8mmol/L)
* AND 75g OGTT at 24 – 34/40 with fasting BSL <7.8mmol/L and 2hr BSL 7.8 – 11.0
I - Dietary advice, blood glucose monitoring, insulin therapy as needed
C - Routine care
O - Primary ouctomes:
* Serious perinatal Cx –> death, shoulder dystocia, bone #, nerve palsy
* Admission to neonatal nursery
* Jaundice requiring phototherapy
* IOL
* CS
* Maternal anxiety and depression
* Maternal health status
Secondary outcomes:
* Components of composite primary outcome
* Fetal: GA, BW, other measures of health
* Maternal: # prenatal visits, MOD, pregnancy weight gain, # antenatal admissions, PIH, other Cx
Exclusion: severe glucose impairment, previously Rx GDM, active chronic systemic disease
ACHOIS - Results?
↓ 70% serious perinatal Cx, RR 0.32, NNT = 34
↑ 15% NN admission, RR 1,15
↑ 30% IOL, RR 1.31
No signficiant difference in: rate of CS, neonatal death, shoulder dystocia, or jaundice
No significant difference maternal mental health but trend towards improvement in health-related QOL in intervention group
ACHOIS - Conclusions
Rx of GDM reduces serious perinatal morbidity and may also improve the woman’s health-related QOL
Insulin Rx in 20% intervention vs 3% routine
ACHOIS - Limitations
Dx thresholds used in the study not consistent with current GDM Dx criteria therefore difficult to apply to practice
ASTECS (Antenatal betamethasone and incidence of neonatal respiratory distress after elective Caesarean section) - study type, publishing details, objectives?
Multicentre pragmatic randomised trial, published 2005 in BMJ by Stutchfield et al.
Aim: To evaluate whether giving 2 doses of betamethasone before delivery reduces incidence of respiratory distress in babies delivered by elective CS at term
ASTECS - PICO
Population, Intervention, Control, Outcomes
Primary outcomes x 3, secondary x 3
P - 998 mothers planned for elective CS at 37/40+
I - 2x IM Betamethasone 12mg doses in the 48/24 before delivery, 24/24 apart
C - Treatment as usual without antenatal steroids (i.e. no placebo used)
O - Primary: Respiratory distress, SCN admission, tachypnoea RR >60 w/ grunting recession or nasal flaring
Secondary: Resp distress severity, arterial gas and oximetry, Level of care needed
Exclusion: severe maternal HTN, Hx peptic ulceration, severe fetal Rh sensitisation, IU infection
ASTECS - Results
↓ >50% RDS/TTN admission, mainly TTN
No signficant difference in respiratory distress severity amongst admitted babies
- RD (RDS + TTN) admission incidence: 5.1% control vs 2.4% treatment, RR 0.46 (CI 0.23 – 0.93)
- RD severity amongst admitted babies similar in both groups
- RDS incidence: 1.1% control vs 0.2% treatment, RR 0.21 (CI 0.03 – 1.32)
- TTN incidence: 4% control vs 2.1% treatment, RR 0.54 (CI 0.26 – 1.12)
ASTECS - Conclusions
Antenatal betamethasone and delaying delivery until 39/40 both reduce admissions to SCN with respiratory distress after ElCS at term by >50%, mainly by ↓ TTN
ASTECS - Limitations
Blinding not practical with placebo, however respiratory distress not susceptible to maternal influence
HAPO (Hyperglycaemia and adverse pregnancy outcomes) - study design, publishing details, objective?
Prospective blinded international multicentre observational study, published 2008 in NEJM by The HAPO Study Cooperative Research Group
Aim: To clarify the risks of adverse outcomes assoc. w/ various degrees of maternal glucose intolerance less severe than that in overt diabetes mellitus
HAPO - PICO?
Population, Intervention, Control, Outcomes
Primary outcomes x 4, secondary x 5
P - 25,505 pregnant women completed an OGTT
Exclusion: < 18yo, uncertain dates, inability to complete OGTT by 32/40, multiple pregnancy, ovulation induction/IVF, glucose testing pre recruitment, Dx diabetes pre preg
I - 75g OGTT at 24 – 32/40
C - no control as observational study
O -
Primary (4):
* BW >90%ile
* Primary CS
* Neonatal hypoglycaemia
* Cord blood serum C-peptide >90%ile, fetal hyperinsulinaemia
Secondary (5):
* Delivery <37/40
* Shoulder dystocia or birth injury
* Need for NICU
* Hyperbilirubinaemia
* Pre-eclampsia
HAPO - Results
Data was analysed in fasting glucose categories, the frequency of each primary outcome was assess in each group
↑ all 1° outcomes - with increasing maternal glucose levels the frequency of each primary outcome increased
* BW >90%ile: 5.3% vs 26.3%, (lowest-highest)
* Primary CS: 13.3% – 27.9%, OR 1.86 in highest category of 1hr BSL
* Neonatal hypoglycaemia: 2.1% – 4.6%
* C-peptide >90%ile: 3.7% – 32.4%,
* Odds ratios for an ↑ BSL by 1 SD highest for BW (1.38 – 1.46) and C-peptide >90%ile (1.37 – 1.55)
Positive associations all 2° outcomes
* Strongest associations with PET, OR for each 1 SD ↑ in each glucose measure 1.21 – 1.28
* Shoulder dystocia/birth injury, OR ~1.2
HAPO - Conclusions
- Strong continuous associations of maternal glucose levels below that diagnostic of diabetes with ↑ BW and ↑ cord-blood serum C-peptide levels.
- Primary outcome frequency ↑ with ↑ maternal glucose levels, less so for neonatal hypoglycaemia
- Positive associations with all 5 secondary outcomes
HAPO - Limitations
- Lack of clear thresholds for risk and 4 primary outcomes not of equal clinical importance make direct translation of results into clinical practice challenging
- No single BSL measure clearly superior in predicting primary outcomes
- No data on variables of nutritional status and gestational weight gain of participants
- Confounders of BMI and PHx macrosomia may have influenced clinical decisions including MOD
- Due to observational nature, cannot conclude that maternal glycaemia is causally related to the AEs
ARRIVE Trial (Labour induction versus expectant management in low-risk nulliparous women), design, publishers, objective?
Multicentre RCT in the USA, published 2018 in NEJM, by Grobman et al.
Aim: To determine perinatal and maternal consequences of IOL at 39/40 among low-risk nulliparous women vs expectant management
ARRIVE Trial - PICO?
Population, Intervention, Control, Outcomes
Primary outcomes x 9 (composite), secondary x 6
P - 6106 low-risk primips, 38+0 – 38+6, with a live singleton fetus, vertex presentation, and absence of maternal/fetal indication for delivery before 40+5
I - IOL at 39+0 – 39+4
C - Expectant management after 40+5 – 42+2
O - Primary = Composite of perinatal death or severe NN Cx
* Perinatal death
* Need for resp support w/i 72hrs after birth
* Apgar at 5 mins ≤3
* HIE, seizure
* Infection, confirmed sepsis/pneumonia
* Meconium aspiration syndrome
* Birth trauma, #, neuro injury, ret haemorrhage
* Intracranial/subgaleal haemorrhage
* Hypotension req vasopressor support
Secondary outcomes (6):
* Caesarean delivery, main 2° outcome
* Operative vaginal delivery
* HTN, chorio, postpartum infection
* 3rd/4th degree tear, PPH, ICU, death
* Labour/BC LOS, postpartum LOS
* Median labour pain score
Exclusion: unreliable GA information, VD C/I, CS planned, PPROM, PVB
ARRIVE Trial - Results
- 1° outcomes: 20% ↓ trend (RR 0.8) in IOL group, however just falling short of statistically significant (p 0.049 and needed to be < 0.046 to be significant). But ∴ IOL not considered to be assoc. with ↑ AEs
- ↓ need for respiratory support, RR 0.71
- ↓ 15% CS, RR 0.84 - 1 x CS avoided every 28 induced
- ↓ 35% HTN, RR 0.64
- IOL ↑ time in labour ward, but ↓ PP LOS, ↓ pain, ↑ perceived control during childbirth - small differences
ARRIVE Trial - Conclusions
- IOL at 39/40 did not result in a significantly lower frequency of composite adverse perinatal outcome, but resulted in a significantly lower frequency of CS
- RR 20% ↓ in IOL suggests IOL probably not as assoc. w/ ↑ adverse perinatal outcomes vs expectant
- No significant difference in perinatal outcome/CS if Bishop unfavourable
ARRIVE Trial - Limitations
- Not powered to detect differences in infrequent outcomes, most perinatal outcomes uncommon
- Unclear if results are broadly generalisable
- Need to evaluate cost-effectiveness of IOL in low-risk nulliparous women at 39/40
Twin Birth Study (Randomised trial of planned CS or VD for twin pregnancy) - design, publishers, objective?
Multicentre, international RCT, published 2013 in NEJM by Barrett et al.
Aim: To determine if planned CS results in a lower risk of adverse outcomes than planned VD in twin pregnancy
Twin Birth Study - PICO?
Population, Intervention, Control, Outcomes
Primary outcomes x 7, secondary x 4
P - 1398 women (2795 fetuses), 32+0 – 38+6 with twin pregnancy
* 1st twin cephalic + both twins live + EFW 1500 – 4000g (on U/S within 7/7 before randomisation)
* Could be DCDA and MCDA (MCMA excluded)
I - Elective delivery at 37+5 – 38+6 via elective CS or IOL
C - Alternate MOD group (i.e. CS group compared to VD group)
O - Fetal: Composite of fetal or neonatal death + serious neonatal morbidity (7)
* Birth trauma - spinal cord injury, skull #, long bone #, periph nerve injury, SDH/ICH
* Apgar 5 mins < 4
* Coma, stupor, ↓ response to pain
* Seizures ≥2 occasions at < 72/24 age
* Septicaemia
* NEC
* BPD, need for assisted ventilation > 24/24
Maternal: Composite of maternal death or serious maternal morbidity < 28/7 postpartum
* PPH >1.5L, blood transfusion or D&C
* Laparotomy
* Hysterectomy, perineal haematoma req evac, broad ligament haematoma
* Intraop damage bladder, ureter, bowel
Exclusion: monoamniotic, fetal reduction at >13/40, lethal fetal anomaly, C/I to labour/VD (fetal compromise, 2nd twin substantially larger, fetal anomaly or condition that may cause mechanical problems at delivery, previous vertical uterine incision or >1 previous CS)
Twin Birth Study - Results
No significant difference composite fetal or maternal 1° outcomes
* Composite primary outcome: 2.2% pCS vs 1.9% pVD, OR 1.16, CI 0.77 – 1.74, p = 0.49, not significant
* Maternal composite outcomes: 7.3% pCS vs 8.5% pVD, p = 0.29, not significant
Twin 2 more likely to have 1° outcome but not related to MOD
Note - Assigned vs actual MOD:
* CS = 90% had CS x 2 / 9% had VD x 2
* VD = 56% had VD x 2 / 40% had CS x2
Assigned CS: 90% had CS for both, 1% combined, 9% VD x2. *59% CS performed before labour *
Assigned VD: 56% VD x2, 4% combined, 40% CS for both. 44% CS performed during labour
Twin Birth Study - Conclusions
- Planned CS did not significantly ↓/↑ the risk of fetal/NN death or serious neonatal morbidity vs VD
- Higher risk of adverse perinatal outcome for T2 but MOD did not affect this risk
Twin Birth Study - Limitations?
- Not powered for subgroup analysis of gestation
- Follow up only until 28/7 after delivery - longer term outcomes not assessed
CLASP Trial (A randomised trial of low-dose aspirin for the prevention and treatment of pre-eclampsia) - design, publishers, objective?
Multicentre RCT, published 1994 in the Lancet by Redman et al.
Aim: To determine the overall safety of low-dose aspirin in pregnancy + whether treatment produces worthwhile effects in pregnancies judged to be at high risk of severe PET or IUGR
CLASP - PICO?
Population, Intervention, Control, Outcomes
Primary outcomes x 5, secondary x 1
P - 9364 women, 12 – 32/40 deemed to be at sufficient risk of PET or IUGR
I - Aspirin 60mg daily until delivery
C - Placebo
O - Development of proteinuric PET,
Estimated duration of pregnancy, BW and BW < 3%ile, Stillbirth/NND ascribed to PET/HTN/IUGR/ maternal/NN bleeding or any cause
Death of baby at any time ascribed to PET/IUGR
Secondary outcome: compliance
CLASP - Results?
- ↓ delivery <37/40, AR 2.5/100, effect larger in PET
- Not significant, 12% ↓ PET, greater effect if started <20/40
- No difference in: BW, stillbirth/NND, abruption, epidural AEs, NN IVH
- ↑ 20% blood transfusion, not assoc. w/ PPH
- Greater ↓ PET the more preterm the delivery ∴ may be justified in women liable to EOPET
- LDA safe for fetus/newborn
CLASP - Conclusions?
- Findings do not support routine prophylactic/therapeutic antiplatelet therapy in pregnancy to all women at ↑ risk PET/IUGR.
- May be justified in women judged to be especially liable to EOPET severe enough to need very preterm delivery.
HYPITAT (Induction of labour versus expectant monitoring for gestational hypertension or mild pre-eclampsia after 36 weeks) - design, publishers, objective?
Multicentre, parallel open-label randomised controlled trial, published 2009 in the Lancet by Koopman’s et al.
Aim: To determine if IOL in women with a singleton pregnancy Cx by gestational HTN or mild PET reduces severe maternal morbidity compared to expectant monitoring
HYPITAT - PICO?
Population, Intervention, Control, Outcomes
Primary outcomes x 3 (composite), secondary x 3
P - 756 women with singleton pregnancy and gestational HTN or mild PET at 36+0 – 41+0
I - IOL within 24h of randomisation
C - Expectant monitoring
O - Primary = composite of:
* Poor maternal outcome, mortality, morbidity (eclampsia, HELLP, pulm oedema, VTE, abruption)
* Progression to severe disease, SBP >170, DBP >110, proteinuria >5g/24h
* Major PPH >1L
Secondary outcomes:
* MOD
* Neonatal mortality
* Neonatal morbidity, 5min Apgar <7, UA pH <7.05, NICU admission
HYPITAT - Results
- ↓ 30% comp mat outcomes, RR 0.71, NNT = 8
- ↓ 40% progression to severe disease (RR 0.64), ↓ 40% requirement for antiHTN (RR 0.63)
- ↓ 55% NN pH <7.05, RR 0.46
- No significant difference PPH rates, trend towards ↓ CS
- No significant difference adverse NN outcomes
HYPITAT - Conclusions
- IOL is assoc. w/ improved maternal outcomes and should be advised for women with mild hypertensive disease >37/40, mainly ascribed to less progression to severe disease.
- 13 per 100 fewer women allocated to IOL had a poor maternal outcome, NNT = 8
HYPITAT - Limitations
- Masking of intervention not possible but unlikely to have influenced outcomes
- Intention to treat analysis, almost half women allocated to expectant underwent IOL
TermPROM (Induction of labour compared with expectant management for prelabour rupture of the membranes at term) - design, publishers, objective?
International multicentre RCT, published 1996 in NEJM by Hannah et al.
Aim: To determine whether inducing labour in women with prelabour ROM is preferable to expectant management if there is no evidence of fetal or maternal compromise
TermPROM Trial - PICO?
Population, Intervention, Control, Outcomes
Primary outcomes x 2, secondary x 2
P - 5041 women with prelabour ROM at ≥37/40 with a single cephalic fetus
I - IOL with IV oxytocin OR vaginal prostaglandin E2 +/- oxytocin
C - Expectant management up to 4 days + labour induced with IV oxytocin OR with PV PGE2 gel
O - Primary Outcomes:
* Definite neonatal infection, clinical + culture+/ CXR or histologic Dx pneumonia
* Probable neonatal infection, clinical + ↑/↓ Neuts, abn CSF with ↑ WCC/protein/glucose
Secondary outcomes:
* Need for CS
* Women’s evaluations of care they received
Term PROM - Results?
↓ 55% chorio
↓ 35% maternal ABx need
↓ 50% PP fever
↓ NN ABx, ↓ NICU >24/24
Better maternal satisfaction with IOL
No difference in NN infection and CS rates
Term PROM - Conclusions
- Rates of neonatal infection and CS were not significantly different among the study groups
- IOL with IV oxytocin results in ↓ risk of maternal infection than expectant management
- Women view IOL more positively than expectant management
Term PROM - Limitations?
Not powered to assess perinatal mortality, however 4 deaths in expectant vs 0 in IOL
MAGPIE (Do women with pre-eclampsia and their babies benefit from magnesium sulphate) - design, publishers, objective?
International, multicentre RCT
Published in the Lancet, 2002 by MAGPIE Trial collaborative group.
Aim: To evaluate the effects of MgSO4 on women and their babies regardless of whether Rx is started before or after delivery and irrespective of any previous anticonvulsant therapy
MAGPIE - PICO?
Population, Intervention, Control, Outcomes
Primary outcomes x 2 main + composite, secondary x 5
P - 10141 undelivered or < 24/24 postpartum women with PET + clinical uncertainty about use of MgSO4
I - MgSO4
C - Saline placebo
O - Primary: Eclampsia, Death of baby pre-D/C, Composite of serious maternal morbidity
Secondary:
* Serious maternal morbidity
* Mg toxicity + Other AEs
* Labour/del Cx
* Neonatal morbidity (Apgar < 7 at 5mins, intubation, ventilation, abn cerebral U/S, convulsions, NICU/SCN admission)
* Resources (LOS, ICU/HDU, ventilation, dialysis)
MAGPIE - Results?
- ↓ 60% eclampsia, RR 0.42, NNT = 91 per 1000
- NNT severe PET = 63
- ↓ 40% placental abruption, RR 0.63, ↓ 12 per 1000
- No difference: Serious maternal morbidity
- No difference: Other neonatal morbidity
Not powered for maternal death but probably ↓ risk
MAGPIE - Conclusions
MgSO4 halves the risk of eclampsia and probably ↓ risk of maternal death. There do not appear to be substantive harmful effects to mother or baby in the short term.
Results consistent regardless of PET severity or Rx before/after delivery, and across low/middle/high perinatal mortality countries ∴ generalisable
MAGPIE - Limitations
Not yet determined the minimum effective dose of MgSO4
Not powered for maternal death
Magnesium sulphate for women at risk of preterm birth for neuroprotection of the fetus - design, publishers, objective?
Cochrane review (systematic review of 5 RCTs), published 2009, Doyle et al
Aim: To assess the effectiveness and safety using the best available evidence, of MgSO4 as a neuroprotective agent when given to women considered at risk of PTB
MgSO4 for fetal neuroprotection - PICO?
Population, Intervention, Control, Outcomes
Primary outcomes x 3
P - 5 trials (6145 babies) eligible for review
I - MgSO4 given to women at risk of PTB administered IV, IM or PO
C - Placebo or no placebo (trial dependent)
O - Primary
* NN/fetal/later death
* Neurological impairment, CP, substantial gross motor dysf
* Maternal: serious adverse CV/resp Cx, AEs to stop Rx
MgSO4 for fetal neuroprotection - Results?
- ↓ 30% CP risk, RR 0.68, NNT = 63
- ↓ 40% substantial gross motor dysfunction, RR 0.61
- No difference on other neonatalpaed mortality
- Maternal deaths, cardiac/resp arrest, ICU
- ↑ RR 3 cessation of Rx
- ↑ RR 1.5 mat hypotension/tachycardia
MgSO4 for fetal neuroprotection - Conclusions
Neuroprotective role of antenatal MgSO4 in women at risk of PTB is now established
* CP absolute risk: 3.7% MgSO4 vs 5.4% unexposed, absolute risk reduction 1.7%, NNT 63
* ↓ substantial gross motor dysfunction
Beneficial effects on substantial gross motor function in early childhood. Outcomes of long-term neurological effects yet to be evaluated
MgSO4 for fetal neuroprotection - Limitations
- Extensive outcome list ↑ possibility of type 1 errors because multiple outcomes evaluated
- Long-term neurological outcome limitations, in part due to methodological limitations (2 studies assessed long-term effects as primary outcome)
- Dx CP in 1 x trial (Mittendorf) unclear, assessment for CP typically < 2yo when Dx is not always certain
ORACLE I (Broad-spectrum antibiotics for preterm, prelabour rupture of fetal membranes) - design, publishers, objective?
RCT published 2001 in the Lancet, Kenyon et al
Aim: To determine neonatal health benefits of antibiotics for PPROM
ORACLE I - PICO?
Population, Intervention, Control, Outcomes
Primary outcomes x 3 (composite), secondary x 6
P - 4826 women with PPROM < 37/40
I - 3 treatment arms: PO QID for 10/7 or until delivery
* Erythromycin 250mg + placebo
* Co-amoxiclav 325mg (Amoxicillin 250mg + Clavulanic acid 125mg) + placebo
* Both Erythromycin + Co-amoxiclav
C - Placebo
O - Composite of
* Neonatal death
* Chronic lung disease, daily supplementary O2 at age 36/40 post conception
* Major cerebral abnormality on U/S
Secondary: delivery 48/24 or 7/7, MOD, mat ABx, NICU, RDS, blood cultures
ORACLE I - Results
- Erythro: ↓ del 48/24, O2 dependence, surfactant need, +ve BCs.
- ↓ comp outcome - not stat significant
- ↑ 4x NEC (1.9% vs 0.5%) co-amoxiclav
ORACLE I - Conclusions
- Health benefits particularly in singleton pregnancies, of Erythromycin: ↓ delivery at 7/7, ↓ need for surfactant, ↓ +ve BC, ↓ chronic lung disease, ↓ major cerebral abnormality
- Advise against Co-amoxiclav for any preterm delivery due to risk of NEC
ORACLE II (Childhood outcomes after prescription of antibiotics to pregnant women with preterm rupture of the membranes) - design, publishers, objective?
Follow up study of randomised controlled trial (7 yr f/u of ORACLE I), published 2008 in the Lancet, Kenyon et al.
Aim: To determine the long-term effects on children of born following PPROM and receiving Erythromycin and or Co-amoxiclav
ORACLE II - PICO?
Population, Intervention, Control, Outcomes
P - Children at age 7 years, born to the 4148 women who had completed the ORACLE I trial
I - Same groups as ORACLE I (erythro, co-amoxi clav, or both)
C - Placebo group
O - Any level of functional impairment + severity of dysfunction, vision, hearing, speech, ambulation, dexterity, emotion, cognition, pain
ORACLE II - Results?
- Outcome known for 75% children eligible for follow up
- No diff proportion with any functional impairment
- ↓ NN morbidity post Erythro did not translate into long-term benefit
- Overall this group has ↓ national norm educational attainment
ORACLE II - Conclusions
ABx prescription for women with PPROM seems to have little effect on the health of children at 7yo.
ORACLE II - Limitations?
- Use of questionnaire may mean more subtle differences between the Rx groups were missed
- Health conditions only assessed through parental report, potential for under-reporting/inaccuracy
- 75% response rate using postal questionnaires
- –> Disadvantaged groups over-represented in non-responders
- –> Response rate lower than assumed in initial power calculation
PPROMT (Immediate delivery compared with expectant management after preterm pre-labour rupture of membranes close to term) - design, publishers, objective?
Multicentre international RCT, published 2016 in The Lancet, by Morris et al
Aim: To establish whether immediate birth in singleton pregnancies with ruptured membranes close to term reduces neonatal infection without increasing other morbidity
Primary outcomes x 1, secondary: neonatal x 6, maternal x 6
PPROMT - PICO?
Population, Intervention, Control, Outcomes
P - 1839 women, singleton pregnancies + PPROM 34 – 36+6/40 with no signs of infection
I - Immediate delivery, ideally within 24/24 of randomisation
C - Expectant management, await SOL or birth planned at term
O - Primary: Neonatal sepsis
Secondary: composite NN morbidity/mortality, RDS, BW, SGA, Apgar5 <7, SCN/NICU LOS. Maternal APH, IP fever, PP ABx, MOD, PPH, chorio
PPROMT - Results?
- ↓ 40% APH (RR 0.6), ↓ LOS, ↓ IP fever
- ↑ 40% CS (RR 1.4), & mech vent (RR 1.4)
- ↑ 60% RDS (RR 1.6)
- ↑ NICU LOS (4 days immediate del. vs 2 days expectant)
- No difference: NN sepsis, NN morbidity/mortality
PPROMT - Conclusions?
In the absence of overt signs of infection or fetal compromise, expectant management with appropriate surveillance of maternal/fetal wellbeing should be followed in PPROM close to term.
PPROMT - Limitations
Expectant PPROM management was as per local guidelines (variation)
* 73% IP Mx
* 90% received ABx
* 40% steroids
Recruitment occured over 10yrs, although no major changes in Mx so results still generalisable
Prophylactic PV progesterone for women at increased risk of PTB - design, publishers, objective?
Double blinded RCT, published 2003 in AJOG by Fonesca et al.
Aim: To evaluate the effect of prophylactic vaginal PG on PTB rates in a high-risk population
Primary outcome x 1
PV progesterone for PTB - PICO?
Population, Intervention, Control, Outcomes
P - 157 asymptomatic singleton pregnancies at high-risk of PTB
* High risk: PHx ≥1 SPTB, cervical cerclage, uterine malformation
I - PV Progesterone 100mg nocte from 24 – 34/40
C - Identical PV placebo
O - Rate of PTB
PV progesterone for PTB - Results?
- ↓ 4x PTB < 34/40, 2.8% PG vs 18.6% placebo
- ↓ 2x PTB < 37/40, 13.8% PG vs 28.5% placebo
- ↓ mean contraction frequency per gestational week with progesterone
- Admissions for PTL, 19.4% PG vs 31.4% placebo, not statistically significant
Average PTB GA 33+5 with PG. vs 32+0 with placebo
PV progesterone for PTB - Conclusions
- Prophylactic PV PG ↓ uterine contraction frequency and PTB rate in women at high prematurity risk
- No significant difference on incidence of PTL admission in both groups
- Given β-mimetic effect, steroids could be used to stimulate type II alveolar cell surfactant synthesis
PV progesterone for PTB - Limitations?
- Fetal outcomes not examined
- Single centre in Brazil, not generalisable
Progesterone and risk of PTB in women with short Cx - design, publishers, objective?
Multicentre RCT published 2007 in NEJM by Fonesca et al.
Aim: To evaluate the effect of PV Progesterone on the incidence of spontaneous early PTB in asymptomatic women found at routine mid-trimester screening to have a short cervix
Primary outcome x 1, secondary x 4
Progesterone for short Cx - PICO?
Population, Intervention, Control, Outcomes
P - 250 women with singleton/twin pregnancies with cervical length ≤15mm on routine anatomy U/S 20 – 25/40
* (out of 24,620 women screened i.e. approx 1% of those screened)
I - PV Progesterone (Utrogestan) 200mg, micronised natural PG
C - Identical-appearing capsules of placebo containing safflower oil
O - Primary: Spontaneous PTB < 34/40
Secondary:
* BW
* Fetal/neonatal death
* Major AEs, IVH, RDS, retinopathy, NEC
* SCN, NICU, ventilation, phototherapy, Rx sepsis, blood transfusion
Progesterone for short Cx - Results?
- ↓ 45% spont PTB < 34/40, RR 0.56
- No difference in neonatal outcomes, no adverse effects either group
- No significant difference in twin pregnancies
- Of the 1.7% with cervical length ≤15mm, 30.9% delivered preterm
- Of the 8.3% with cervical length 16 – 25mm, 5.1% delivered preterm
Progesterone for short Cx - Conclusions
- In women with a short cervix, daily PV PG 200mg from 24 – 34/40 significantly ↓ SPTB
- TVUS at 22/40 identifies a subgroup of ~1.5% of the population at high risk of PTB, incidence 34%
- No significant difference in PTB in twin pregnancies
- ACOG notes that the ideal formulation, optimal route of delivery and long-term safety of PG unknown
Term Breech Trial (Planned Caesarean section versus planned vaginal birth for breech presentation at term) - design, publishers, objective?
International multicentre randomised trial, published in the Lancet in 2000 by Hannah et al.
Aim: To determine whether planned CS was better than planned vaginal birth for selected fetuses in the breech presentation at term - (affects 3 – 4% fetuses)
Primary outcomes x 5
Term Breech Trial - PICO
Population, Intervention, Control, Outcomes
P - 2088 women with a singleton fetus in a frank/complete breech presentation
I - CS at ≥38/40
C - Expectant management until SOL unless indication for IOL, e.g. post-dates (vaginal birth group)
O - Primary outcomes:
* Perinatal/neonatal mortality < 28/7
* Serious NN morbidity
* Birth trauma (SDH, IVH, SCI, basal skull #, nerve injury)
* Seizures / hypotonia / abn. Consciousness
* APGAR5 < 4 / pH < 7.0 / NICU care
Exclusion: fetopelvic disproportion, suspected LGA/EFW >4kg, hyperextension of fetal head, fetal anomaly or condition that might cause mechanical problem at delivery (hydrocephalus), C/I vaginal delivery (praevia), lethal fetal congenital anomaly
Term Breech Trial - Results?
- Randomised to CS - 90.4% delivered by CS
- Randomised to vaginal birth - 56.7% delivered vaginally
- ↓ 33% PN/NN mortality or serious morbidity, NNT = 14
- No difference in serious maternal complications
Term Breech Trial - Conclusions
- Planned CS is better than planned vaginal birth for the term breech fetus – for every additional 14 CS done, 1 baby will avoid death or serious morbidity
- Benefits greater in countries reported to have lower perinatal mortality rates, potentially artefact due to detection bias (less likely to be monitored for evidence of birth trauma, earlier D/C home)
Term Breech Trial - Limitations?
- Potential selection bias - 2088 enrolments across 121 centres over 39mths, average of 5/centre/yr
- Consent could be obtained while intrapartum ∴ little to no time for fetal assessment or counselling
- Intrapartum CTG monitoring optional
- Inclusion of BW < 2.5kg, associated with term FGR
- Variable degrees of experience and training of the accoucheur, self-assessment vs 20yrs
- No blocking of units by delivery
TRUFFLE (Trial of Umbilical and Fetal Flow in Europe - 2 year neurodevelopmental and intermediate perinatal outcomes in infants with very preterm fetal growth restriction) - design, publishers, objective?
Prospective multicentre, unblinded randomised study
Published 2015 in The Lancet by Lees et al.
Aim: To assess whether changes in DV waveform can be used as indications for delivery instead of CTG in severe early onset FGR
Primary outcomes x 3
TRUFFLE - PICO
Population, Intervention, Control, Outcomes
P - 511 women with singleton fetuses 26 – 32/40 with very preterm FGR
I - 3 timing of delivery plans:
1. cCTG short-term variation: <3.5ms <29/40 or <4ms ≥29/40
1. Early DV changes: DV PI >95%ile
1. Late DV changes: ‘a’ wave absent or reversed
C - Monitoring for all groups: UA Doppler and CTG ≥1x/week
* Safety net delivery criteria: CTG STV <2.6ms <29/40 or <3ms ≥29/40 OR persistent unprovoked decels
O - Primary:
* Survival w/o neurodev impairment at 2yo
* Perinatal/infant death < 2yo
* Impairments at 2yo - CP, neurosensory impairment
Secondary: composite adverse NN outcome (fetal/postnatal death, ≥1 severe morbidities – BPD, IVH, PVL, NN sepsis, NEC)
cCTG = computerised CTG
STV/Short term variation = variability
Inclusion: FGR = AC < 10%ile) + UAPI > 95%ile, EFW > 500g and normal DV waveform with PI < 95%ile
Exclusion: delivery known/planned/impending, major fetal structural abn, fetal karyotype abn, <18yo
TRUFFLE - Results?
92% infants alive at 2yrs
Median gestation at delivery 30.7 weeks, mean BW 1019g
Improved outcome in survivors for DV absent/rev ‘a’ vs CTG, p = 0.005
Mortality 14% enrolled < 29/40 vs 4% enrolled ≥ 29/40, however rates of abn neurodevelopment similar –> not sig
TRUFFLE - Conclusions
- Difference in proportion of infants surviving without neuroimpairment non-significant, but trend towards least impairment in DV no ‘a’ vs CTG STV
- Delivery timing based on late DV waveform changes might produce an improvement in 2yo dev outcomes
TRUFFLE - Limitations
- 13% lost to follow up - but analysis did not show significant changes on imputation sets
- Individual components of composite primary outcome did not differ - but these events were low frequency
- Difficulty in reliance on one marker to trigger delivery
WOMAN (Effects of early tranexamic acid administration on mortality, hysterectomy and other morbidities in women with postpartum haemorrhage) - design, publishers, objective?
International, multicentre, randomised, double-blind, placebo-controlled
Published 2017 in The Lancet, by Shakur et al.
World Maternal Antifibrinolytic Trial
Aim: To assess the effects of early administration of TXA on death, hysterectomy and other relevant outcomes in women with PPH
Primary outcomes x 2, secondary x 5
WOMAN - PICO?
Population, Intervention, Control, Outcomes
P - 20 060 women ≥16yo with clinical PPH post vaginal birth
I - IV TXA 1g (100mg/ml) at 1ml/min
C - Matching NaCl placebo
O - Composite of:
* Death from all causes, also assessed separately
* Hysterectomy within 42/7 randomisation
Secondary: death due to bleeding, thromboembolic events, surgical intervention, Cx, QOL
WOMAN - Results?
- ↓ 30% death from bleeding - if < 3/24, RR 0.69 (if > 3/24 then RR 1.07)
- ↓ 35% laparotomy, RR 0.64
- No difference: Death from all causes or need for hysterectomy
- No difference: Thromboembolic events, organ failure, sepsis, QOL
WOMAN - Conclusions
- TXA ↓ death due to bleeding in women with PPH with no adverse effects or complications
- When given soon after delivery, TXA ↓ death from bleeding by nearly 1/3
- Death from all causes or need for hysterectomy not reduced with TXA
- No significant difference in thromboembolic events, organ failure, sepsis, QOL
WOMAN - Limitations?
- During trial, apparent that hysterectomy decision often made at same time as decision to enrol ∴ likely dilutes the effect of TXA on the risk of hysterectomy.
- Further assessment needed into bioavailability of TXA via non-IV routes as many PPH deaths occur at home or settings where IV may not be feasible
A comparison of medical management with misoprostol and surgical management for early pregnancy failure - design, publishers, objective?
Multi-centre randomised, non-inferiority trial. Published in 2005 in NEJM by Zhang et al.
Aim: To assess the efficacy, safety and acceptability of Misoprostol for treatment of failed pregnancy
Medical vs Surgical Mx for Miscarriage - PICO?
P - 652 women with T1 pregnancy failure (anembryonic, embryonic or fetal death, or incomplete spontaneous abortion)
I - PV Miso 800microg, in a 3:1 ratio with vacuum aspiration
* Treatment on D1
* +/- 2nd dose D3 if expulsion incomplete
* +/- vacuum aspiration D8 if expulsion still incomplete
C - Vacuum aspiration (standard of care)
O - Treatment success or failure (primary)
Secondary: Haemorrhage, Pelvic infection, Hospitalisation, Side effects, Acceptability to patient
Note: slight difference in criteria compared to today
* CRL 5 – 40mm
* Anembryonic gestational sac, MSD 16 – 45mm
* Growth of GS <2mm over 5/7 or <3mm over 7/7
* ↑ hCG levels <15% over 2/7
* Incomplete/inevitable: residual AP endometrial lining >30mm, uterus <13/40 size
Medical vs Surgical Mx for Miscarriage - Results?
Miso:
* 71% complete expulsion by D3
* 84% complete expulsion by D8
* 16% Rx failure (esp anembryonic)
* 78% would have Miso again
Surg:
* 97% success rate (absolute diff 12%)
* 1% haemorrhage
* < 1% endometritis req hospitalisation
Medical vs Surgical Mx for Miscarriage - Conclusions
- Treatment of early pregnancy failure with PV Miso 800 microg, +/- repeat dose after 48/24 if required, is efficacious, safe and acceptable, success ~84%
- Similar risks of haemorrhage and pelvic infection compared with vacuum aspiration
- Side effects tolerable
Medical vs Surgical Mx for Miscarriage - Limitations?
- Only vaginal administration of misoprostol was studied
- The lowest effective dose is not clear (? possibly less than 800 microg)
ALIFE Trial (Aspirin plus heparin or aspirin alone in women with recurrent miscarriage) - design, publishers, objective?
Multicentre randomised placebo-controlled trial, published 2009 in NEJM by Kaandorp et al.
Aim: To determine if aspirin +/- LMWH improves the chance of a livebirth for women with unexplained recurrent miscarriage
ALIFE Trial - PICO?
P - 364 women with a Hx of unexplained recurrent miscarriage (≥2 MC, not specified consecutive) and attempting to conceive or < 6/40
I - Aspirin 80mg + Nadroparin 2850 IU vs Aspirin alone
C - Placebo
O - Live birth rate (primary)
Secondary:
* Miscarriage
* FDIU
* Obs Cx, PET, HELLP, abruption, PTB, SGA
* Maternal and fetal adverse events
* Thrombocytopaenia
ALIFE Trial - Results?
- No differences in livebirth rates
- No significant diff secondary outcomes except combination group delivered 1/52 earlier than placebo
- Increase AEs in Combination group: 50% bruising, 40% swelling/itch at injection site
Trial was discontinued early as deemed futile
ALIFE Trial - Conclusions
Neither aspirin + nadroparin nor aspirin alone improved the live birth rate vs placebo in women with a Hx of unexplained recurrent miscarriage
ALIFE Trial - Limitations?
- Trial discontinued when 22 women were still in follow up as continuation was deemed futile
- Study-drug adherence was 85%, may increase statistical uncertainty around observed absence of effect
- Use of nadroparin was not blinded
- Use of broad definition of RM (≥2 MC) may have diluted results compared with ≥3 MC - however no significant benefit of combination vs aspirin alone that were stratified according to no. of MCs
SPIN Study (LMWH and LDA in women with recurrent miscarriage) - design, publishers, objective?
Multicentre randomised controlled trial, published 2010 by Clark et al in ‘Blood’ - Journal of Haematology.
Aim: To assess whether treatment with enoxaparin and LDA + intensive pregnancy surveillance reduces rate of pregnancy loss c/w intensive pregnancy surveillance alone in women with a Hx of ≥2 consecutive previous pregnancy losses
SPIN Study - PICO?
P - 294 women with a Hx of ≥ 2 consecutive previous pregnancy losses at ≤ 24/40 with a positive pregnancy test at < 7/40
I - Subcut Enoxaparin 40mg daily + PO Aspirin 75mg daily until 36/40 + intense pregnancy surveillance
C - Intensive pregnancy surveillance alone
O - Pregnancy loss rate (primary)
Secondary:
* Serious adverse event, APH, PPH, ↓ Hb
* Non-serious adverse event, nosebleed, injection site pain/itch/rash, ↓ Plts
SPIN Study - Results?
22% pregnancy loss in pharmacologic intervention vs 20% intensive surveillance alone, OR 0.91, 95% CI 0.52 – 1.59
No serious AEs recorded
SPIN Study - Conclusions
No measurable benefit of LDA + LMWH in preventing further pregnancy loss c/w intensive fetal surveillance
SPIN Study - Limitations?
57% women had 2 previous miscarriages so it is possible that a beneficial effect of pharmacologic intervention may be more apparent in women with ≥3 RMs
Ovarian conservation at time of hysterectomy for benign disease - design, publishers, objective?
Retrospective meta-analysis (PubMed/Cochrane) of the English papers from 1990
Published 2005, in AJOG by Parker et al.
Aim: To determine the optimal age for prophylactic oophorectomy at time of hysterectomy for benign disease
Ovarian conservation at time of hysterectomy - PICO?
P - Healthy women aged 40-80yo who had a hysterectomy for benign disease
I - 4 circumstances:
* Ovarian conservation + oestrogen therapy
* Ovarian conservation – oestrogen therapy
* Oophorectomy + oestrogen therapy
* Oophorectomy – oestrogen therapy
C - Background population risk in those not undergoing interventions
O - Mortality from:
* Ovarian cancer
* Coronary heart disease
* Hip fracture
* Breast cancer
* Stroke
Note: Mortality only (due to lack of data on morbidity/QOL post ovarian conservation vs oophorectomy)
Ovarian conservation at time of hysterectomy - Results?
- No statistically sig diff in survival >64yo
- Age < 65yo clearly benefit from ovarian conservation
- No clear benefit for prophylactic oophorectomy at any age – RR CVD/# death > ovarian Ca death
- Ooph and no E replacement: ↑ hip # and CHD risk
Ovarian conservation at time of hysterectomy - Conclusions
- Women < 65yo clearly benefit from ovarian conservation
- > 65yo = definite conclusions more difficult to reach
- At no age is there a clear benefit from prophylactic oophorectomy
RR of dying from ovarian ca is overshadowed by the risks from CVD + hip fracture
Ovarian conservation at time of hysterectomy - Limitations
Retrospective study
Reliability of statistics used for mortality estimates and assumptions made
* Probability estimates derived mostly from case-control studies, selection/reporting bias, chance
* No data for CHD risk in women >65yo
* No direct mortality rate for osteoporotic hip fracture found
Model assumes conditions are mutually exclusive, death attributed to 1 of the 5 conditions or to all other causes
LACE Trial (Effect of TLH vs TAH on disease free survival among women with stage 1 endometrial ca) - design, publishers, objective?
Multinational randomised equivalence trial, published in 2017 in the JAMA by Janda et al.
Aim: To investigate whether TLH is equivalent to TAH in women with treatment-naïve endometrial cancer
LACE Trial - PICO?
P - 760 women with histologically confirmed stage 1 endometrioid endometrial ca
I - TLH +/- LN dissection
C - TAH +/- LN dissection
O - Disease free survival - at 4.5 years (Primary)
Secondary:
* Recurrence of endometrial ca
* Overall survival
* Morbidity, pain, analgesic use, QOL, cost-effectiveness
LACE Trial - Results?
- Disease-free survival 81.6% TLH vs 81.3% TAH at 4.5yrs (i.e. no difference)
- TLH ↓ post-op AEs and lower cost
- Survival: BMI < 30 TAH, ≥30 TLH more favourable
- No sig diff for age, FIGO stage, Hx malignancy
LACE Trial - Conclusions
Equivalent disease-free survival at 4.5yrs and no overall difference in overall survival in TAH c/w TLH ∴ support the use of laparoscopy hysterectomy for women with stage 1 endometrial ca
LACE Trial - Limitations?
- Blinding not possible however unlikely to affect disease-free survival outcomes
- Pelvic/aortic retroperitoneal node dissection left to surgeon’s discretion ∴ inconsistent application
PLCO (Effect of screening on ovarian ca mortality - The Prostate, Lung, Colorectal, and Ovarian cancer screening RCT) - design, publishers, objective?
Randomised control trial published in the JAMA in 2011 by Buys et al.
Aim: To evaluate the effect of screening for ovarian Ca on mortality in the Prostate, Lung, Colorectal and Ovarian (PLCO) Cancer Screening Trial
PLCO - PICO?
P - 78, 216 women aged 55 – 74 with no previous Dx of lung, colorectal or ovarian Ca
I - Annual screening (n = 39, 105)
CA-125 for 6 years + TVUS for 4 years
C - Routine care
O - Mortality from ovarian Ca, including primary peritoneal and fallopian tube Ca
Secondary:
* Ovarian Ca incidence
* Cancer stage
* Survival
* Potential harms of screening and Dx procedures (Bleeding, fainting, bruising, Infection, CV events, Bowel injury)
* All-cause mortality
PLCO - Results?
No sig diff in ovarian Ca Dx or Death
* Diagnosis = 5.7 per 10,000 in screening group vs 4.7 in routine
* Ovarian Ca death = 3.1 per 10,000 (screening), vs 2.6 per 10,000 in control
No stage shift
5% false +ve each round of screening (60% from TVUS)
* increased surgical F/Up –> 15% (n = 163) had ≥1 serious Cx
PLCO - Conclusions?
- Simultaneous screening with CA-125 and TVUS vs usual care did not reduce ovarian Ca mortality
- No stage shift, suggesting screening tests not effective in detecting ovarian Ca early enough
- Diagnostic evaluation following a false +ve screening test result was associated with complications
PLCO - Limitations
Not blinded due to invasive nature of screening procedures
Trial powered for 35% mortality reduction, however from a public health perspective, smaller effect sizes are still potentially worthwhile to detect
Provision of no-cost long-acting contraception and teenage pregnancy (as part of the Contraceptive ‘CHOICE’ Project) - design, publishers, objective?
Prospective cohort study, published in NEJM in 2014 by Secura et al.
Aim: To determine pregnancy, birth and abortion rates in a cohort of teens among whom the barriers to highly effective reversible contraception were removed, compared to the rates observed nationally among all teens in the US
Provision of no-cost long-acting contraception and teenage pregnancy - PICO?
P - 1404 aged 14 – 19yo
Overall project enrolled 9256 girls and women aged 14 – 45yo
I - Standardised contraceptive counselling to participants re: commonly used RCMs
Presented in order from most to least effective, side effects, risks, benefits
Participants provided with their chosen method on the day of enrolment
C - Compared primary outcomes with most recent available rates among all US teens 15 – 10yo (2010)
O - Pregnancy, Live birth, Induced Abortion
RCMs = reversible contracetive methods
Provision of no-cost long-acting contraception and teenage pregnancy - Results?
- Pregnancy 34 vs 159 per 1000 teens
- Teen births 19 vs 94 per 1000 teens
- Abortion 10 vs 42 per 1000 teens
- Use/failure: IUD < DMPA < OCP < ring < patch
- Methods at time of conception: IUD (2), DMPA (1), OCP (13), ring (4), patch (2), condoms (9), none (25)
- No pregnancies with copper IUD or Implanon
Provision of no-cost long-acting contraception and teenage pregnancy - Conclusion
Much lower rates of pregnancy, birth and abortion in teen girls provided with contraception at no cost and educated about RCMs and benefits c/w national rates for sexually experienced teens
Provision of no-cost long-acting contraception and teenage pregnancy - Limitations?
- Information about pregnancy self-reported by participants, may be underestimated
- Being surveyed on a regular basis may have influenced adherence
- Not randomised, pts allowed to allocate themselves to a contraceptive
- Generalisability of results uncertain, ?different counselling approach to the community
- Parental consent required ∴ enrolled cohort may represent teens at lower risk for contraceptive non-use and pregnancy
Compared mean annual rate for 2008 – 2013 with 2010 national rates, pregnancy rates ↓ by 15% 2008 – 2010, however reductions observed in CHOICE are substantial and of public health importance
Breast cancer and hormone replacement therapy in the Million Women study - design, publishers, objective?
Observational study, published in 2003 in The Lancet by Banks et al.
Aim: To investigate the relation between various patterns of use of HRT and breast Ca incidence and mortality
Million Women Study - PICO?
P - 1,084,110 UK women aged 50 – 64yo, who had not previously been registered as having cancer
I - Use of HRT – oestrogen-only, oestrogen-progesterone combination, tibolone, progesterone-only, vaginal Rx, combination of the above
C - No HRT use
O - Risk of breast Ca amongst ever (current/past) and never users, and different preparations used
Million Women Study - Results?
Ever vs never used HRT - RR 1.43
* Ever - Current use RR 1.66
* Ever - Past use = RR 1.14 if only ceased within 1yr, no diff after that (same as never users)
Preparations: E only RR 1.3, E+P RR 2 (4x E only), Tibolone 1.5
* No sig diff oral, transdermal or implanted
* Effect ↑ with duration of use
RR of breast Ca
Million Women Study - Conclusions
- Current use of HRT is associated with an increased risk of incident and fatal breast Ca
- Effect substantially greater for oestrogen-progesterone combinations than other types of HRT
- RR in current users increases with increasing duration of use of HRT
- No overall increase in the RR of breast ca in past users of HRT
Million Women Study - Limitations?
Misclassification of the women’s use of HRT but should not affect the main conclusions, as was used to define current, past and never users
WHI - Effects of conjugated equine estrogen in postmenopausal women with hysterectomy - design, publishers, objective?
Randomised, double-blind, placebo-controlled disease prevention trial, published in 2004 in the JAMA as part of the ‘Women’s Health Initiative’
Aim: To assess the effects on major disease incidence rates of the most commonly used postmenopausal hormone treatment preparation in the US on chronic disease incidence
Oestrogen in PM women with hysterectomy - PICO?
P - 10,739 postmenopausal women aged 50 – 79yo who had undergone hysterectomy
I - Conjugated equine estrogen, 0.625mg/d
C - Placebo
O - Primary:
* Coronary heart diseaseincidence, non-fatal MI or CHD death
* Invasive breast ca incidence
Secondary:
* Stroke
* Pulmonary embolism
* Colorectal ca
* Hip fracture
* Death from other causes
Oestrogen in PM women with hysterectomy - Results?
Hazard Ratios: (< 1 is good, > 1 is bad)
* CHD: 0.91 (0.75 – 1.12) - not significant
* Breast ca: 0.77 (0.59 – 1.01)
* Stroke: 1.39 (1.10 – 1.77)
* PE: 1.34 (0.87 – 2.06)
* CRC: 1.08 (0.75 – 1.55)
* Hip #: 0.61 (0.41 – 0.91)
Equates to roughly:
* ↑ 40% stroke risk
* ↑ 30% DVT
* ↓ 40% hip #
* ↓ 20% invasive breast Ca
Oestrogen in PM women with hysterectomy - Conclusions
No sig diff CHD, CRC, total Ca
No overall benefit to mortality ∴ CEE should not be recommended for chronic disease prevention
Oestrogen in PM women with hysterectomy - Limitations?
- Only tested one unopposed estrogen preparation at a single oral dose ∴ cannot extrapolate to other formulations, doses or routes
- High rates of discontinuation of study medications, 53.8% at time of study termination
- Higher than expected crossover from placebo to active hormone use, 9.1% in placebo group
- Trial ceased early, reducing precision of estimated effects. Longer intervention period may have provided stronger statistical evidence of CEE effects on CHD and breast ca
WHI - Risks and benefits of estrogen plus progestin in healthy menopausal women - design, publishers, objective?
Randomised controlled trial, published in 2002 in the JAMA by Rossouw et al. (Part of Women’s Health Initiative)
Aim: To assess the major health benefits and risks of the most commonly used combination hormone preparation in the US
WHI Risks + Benefits of HRT - PICO?
P - 16,608 postmenopausal women aged 50 – 70yo with an intact uterus
I - Conjugated equine oestrogens 0.625mg/d + Medroxyprogesterone acetate 2.5mg/d
C - Placebo
O - Primary:
* Coronary heart diseaseincidence, non-fatal MI or CHD death
* Invasive breast ca incidence
Secondary:
* Stroke
* Pulmonary embolism
* Colorectal ca
* Endometrial ca
* Hip fracture
* Death from other causes
WHI Risks + Benefits of HRT - Results?
Absolute risk / 10,000 person years:
* CHD: 1.29 (1.02 – 1.63)
* Breast ca: 1.26(1.00 – 1.59)
* Stroke: 1.41 (1.07 – 1.85)
* PE: 2.13 (1.39 – 3.25)
* CRC: 0.63 (0.43 – 0.92)
* Endometrial ca: 0.83 (0.47 – 1.47), not signif
* Hip #: 0.66 (0.45 – 0.98)
* Death other causes: 0.92 (0.74 – 1.14)
100 more per 10 000 experienced a global index event
* ↑ 30% CHD, 25% breast Ca, 40% stroke, 2x risk DVT/PE
* ↓ 35% CRC
* No impact endometrial Ca
WHI Risks + Benefits of HRT - Conclusions
Overall health risks exceeded benefits from use of combined E + P for an average 5.2yr follow up among healthy postmenopausal US women.
CEE + MPA should not be initiated for primary prevention of CVD
WHI Risks + Benefits of HRT - Limitations?
- 1 drug regimen of CEE + MPA tested, results do not apply to lower dosages, other formulations
- Does not distinguish effects of oestrogen from progestin
- High rates of discontinuation of study medications, 42% E + P, 38% placebo
- Dropout rates exceeded design projections
