Patricia R Chess MD

1
Improved Perinatal Outcomes through Reduction of
Elective Deliveries Prior to 39 Weeks
Patricia R Chess MD Associate Professor of
Pediatrics and Biomedical Engineering University
of Rochester Golisano Childrens Hospital at
Strong
2
Humankind has been delivering babies for millions
of years

• Homo Habilis- Handy man walked the earth 14
  million- 1,750,000 years ago
• Homo Sapiens- Wise Man 130,000 years ago
• Biblical Adam- 6000 years ago
• The wise man has managed to shorten the length
  of human gestation an average of 1 full week over
  just 10 years!

3
Changing Distribution of US Live Births
Percent Singleton Live Births
Davidoff et al Sem Perinatology 2006
4
Gestational age-specific total cesarean section
and labor induction rates among all singleton
live births, United States, 1992 and 2002
M.J. Davidoff et al. Semin Perinatol 2006
5
Medically Indicated Factors to Deliver

• Placental abruption/ previa with bleeding
• Infection
• Maternal medical conditions hypertension,
  cancer, transplant, SLE etc
• Preeclampsia
• Idiopathic preterm labor
• Premature rupture of membranes
• Intrauterine growth restriction
• Multiple gestations
• Fetal congenital anomalies-
• eg Gastroschisis with edema of intestines
• Poor placental function/ severe oligo
• Isoimmunization with anemia

6
When to electively schedule delivery

• Preterm
• ie before the 259th day counting from the first
  day of the last menstrual period
• Late preterm34 0/7-36 6/7 weeks
• Term 37 0/7- 41 6/7
• Low risk of morbidity after 37 weeks
• Ergo 37 weeks is close enough when scheduling an
  elective delivery, or is it

7
Elective Delivery

• Scheduled, nonurgent
• C/S
• Induction
• Indications
• Elective repeat C/S
• Relative medical reasons-LGA, twins etc
• Doctor/ patient schedule
• work schedule/ OR or LD availability/ vacations
• Patient anxiety/ discomfort

8
Elective Delivery

• 30.2 of births C/S (1.2 million in 2005)
• Trial of labor after C/S not without risk
• Up to 50 of C/S are repeat C/S
• 10 of infants in US delivered by elective repeat
  C/S
• Inductions for non-medical reasons also on the
  rise
• Timing of delivery is critical and can lead to
  iatrogenic potentially preventable morbidity and
  mortality

ACOG Comm Op 2007
9
The good survival 90 after 27 weeks
10
The badNeonates born at 36-39 weeks at
increased risk of

• Transient Tachypnia of the Newborn
• Respiratory Distress Syndrome
• Temperature instability
• Hypoglycemia
• Hyperbilirubinemia/ Kernicterus
• Higher rates of rehospitalization
• Feeding problems
• Apnea/ SIDS
• Seizures

Riskin et al, Am J Perinat 2005
11
Hypothermia increases with decreasing gestational
age

• Increased need for intervention in DR
• Decreased subcutaneous tissue
• Increased proportion of surface area to total
  body size
• Increased risk of infection

Raju et al, Pediatrics 2006
12
Hypoglycemia

• 18 at 35-36 weeks gestation
• 4 at term
• Decreased subcutaneous tissue
• Cold stress
• Poor po feeding
• Infection

Raju, Pediatrics 2006
13
Hyperbilirubinemia

• Decreased hepatic uptake of bilirubin from
  plasma, delayed bilirubin conjugation, increased
  enterohepatic circulation of bilirubin,
  dehydration
• 54 receive Rx _at_ 35, 36 wks, 38 _at_ FT
• Narrow margin of safety, especially with LGA late
  preterm infants

Maisels, Pediatrics 1998
14
Brain

• At 35 weeks of gestation significantly fewer
  sulci and brain weight is only 60 of term
  infants
• Over last 4 weeks of gestation there is a
  dramatic increase in gyri, sulci, synapses,
  dendrites, axons, oligodendrocytes, astrocytes,
  microglia
• Higher rate of seizures at earlier gestation due
  to immaturity of neurons
• May be at increased risk of bilirubin-induced
  brain injury

Kinney et al, Sem Perinat 2006
15
Recurrent Apnea

• 4-5 of late-preterm infants, close to 0 at term
• Neurodevelopmental immaturity- increased REM
  sleep
• GER
• Infection

Hunt et al, Sem Perinatology 2006
16
Apparent Life-Threatening Events

• 8-10 incidence in preterm infants
• 1 incidence in full-term infants
• 30 of ALTE infants preterm and of these 87 were
  late preterm

Hunt et al, Sem Perinatology 2006
17
Sudden Infant Death Syndrome

• Late-preterm infants at 2 fold higher risk
• 1.4 cases/1000 at 33-36 weeks gestation
• 0.7/1000 at term
• Monitors not found to decrease morbidity or
  mortality from SIDS
• Place infant, including late-preterm infants, on
  back to sleep

Hunt et al, Sem Perinatology 2006
18
Respiratory morbidity4 fold increase at 38
weeks, 5 fold at 37 weeks
Hansen et al BMJ 2008
19
(No Transcript)
20
Risk of PneumothoraxDecreases as Gestational Age
Increases
Zanardo et al J Pediatrics 2007
21
Need for Respiratory Resuscitation in DR
1284 ECS, 1284 matched vaginal deliveries
Zanardo et al, Ped Crit Care Med, 2004
22
Long- term Sequelae
Lindstrom et al, Pediatrics 2007
23
Total Number of Neonatal Respiratory ECMO Runs
Decreasing
data from ELSO registry
24
Delivery Trends in ECMO Patients
data from ELSO registry
25
ECMO trends looked at another way
pts
Jain et al, Sem Perinat, 2006
26
And the ugly
Mortality on ECMO Neonates with respiratory
conditions
C cesarean EC elective cesarean V vaginal
p 27
ECMO after elective C/SNeonates with respiratory
conditions

p 28
Mortality Related to Gestational Age
Young et al, Pediatrics 2007
29
Were casualties of our own success

• As medical care improves resulting in improved
  outcomes at lower gestational ages, and lives get
  more hectic, people trade a perceived negligible
  risk of elective delivery prior to 39 weeks for
• Convenience
• Decreased discomfort
• Relief of anxiety
• .

30
Choosing When to Deliver

• ACOG recommends elective deliveries be avoided
  prior to 39 weeks, and if they are scheduled
  prior to 39 weeks an amniocentesis be done to
  assess lung maturity
• Many elective deliveries occur prior to 39 weeks,
  most without amniocentesis for lung maturity- how
  does one change practice?

ACOG Practice Bulletin 10 11/99
31
Scheduled Cesarean Gestational AgesLow risk,
not in laborFinger Laker Region
N3661 (40 below 39 weeks)
32
NICU Admission after Scheduled Low-risk Cesarean
SectionFinger Lakes Region
33
How to address the issue

• Education of risks directed to
• Medical care providers
• Families

34

• Aim to decrease scheduled deliveries prior to
  39 weeks, and increase amniocenteses to assess
  for lung maturity if scheduled before 39 weeks-
  as easy as teaching an old dog new tricks

35
Approach

• Identify factors contributing to the choice to
  delivery electively prior to 39 weeks
• Develop educational materials and outreach
  efforts related to risks of such deliveries
  directed to care providers and families aimed at
  decreasing this choice
• Assess the effectiveness of the efforts using the
  PDS

36
Objective 1. Determine the incidence of elective
delivery prior to 39 weeks

• Identify elective deliveries regional hospital
• Validate PDS data on elective delivery by
  conducting a PDS data audit in which indications
  for elective delivery identified in the medical
  record are compared with PDS data.

37
Objective 2. Identify key factors influencing
decision for elective delivery

• Using the RPC medical record and health care
  provider review, identify factors related to
  choice to deliver electively prior to 39 weeks
  (both physician and patient)

38
Objective 3. Statewide definition of elective
delivery using PDS data

• Using PDS data develop a standard definition of
  elective delivery
• To serve as baseline to track incidence

39
Objective 4. QI Bundle

• Develop educational brochures for mothers related
  to benefit of full 9 month pregnancy and risk of
  early inductions for nonmedical reasons and
  distribute throughout 9 county region
• Develop educational program for providers
  consisting of grand rounds, outreach, pamphlets
  after obtaining provider feedback of optimal
  format/ mode of dissemination of material

40
Objective 5. Implement QI bundle

• Distribute educational material to 9 county area
  covered by RPC
• to OB offices
• to offices where prenatal labs are drawn
• include with OB admission packets
• Meet with OBs to discuss ACOG guidelines, risks
  of delivering electively before 39 weeks
• Track elective deliveries
• Provide feedback to caregivers who continue to
  deliver electively prior to 39 weeks

41
Objective 6. Measure effectiveness of educational
program by using perinatal database to measure
rate of elective deliveries prior to 39 weeks
after educational initiative

• Analyze Perinatal Database for number of elective
  deliveries prior to 39 weeks beginning 12 months
  after completion of project

42
Project Design

• Retrospective review
• Population
• Mothers presenting for delivery to Strong
  Memorial Hospital
• Singleton deliveries occurring from January 2006
  to December of 2007 (one year to develop
  definition of elective delivery from PDS, second
  year to test definition)
• Gestational age 36 0/7 - 38 6/7 weeks gestation
  from QS system
• Design
• Review of maternal charts
• Data collection of specific information
• Reason for admission
• Indication of Delivery
• Outcome

43
Delivery Classified in QS as Medicalany of
following listed as reason for maternal admission
in electronic medical record

• Labor
• Bleeding
• Decreased fetal movement
• NST performed
• PIH
• ROM
• Version

44
Delivery Classified in QS as Possible Elective
any of following listed as reason for maternal
admission in electronic medical record- all paper
charts in these categories reviewed

• No reason listed
• Induction
• Observation
• Repeat cesarean section
• Primary cesarean section

45
Medical record data collection form
46
Results

• 1707 patients were screened using QS
• 130 omitted duplicates, multiple gestations in
  the QS
• 725 identified as possible elective deliveries
  and charts reviewed
• 459 determined to be elective
• 266 determined to be medically indicated
• 1118 patients were deemed nonelective based on
  predetermined criteria listed
• 852- evidence for medical indication from QS
  system
• 266- based on chart review after initial
  identification as possible elective

47
Reason for Maternal Admission at Gestations 36
0/7-38 6/7 weeks
Probable Nonmedical
Probable Medical
48

• 
  
• 
  n1577 P

49

• Comparing variables
• of/to elective deliveries

50
(No Transcript)
51
Revised Project Designregarding maternal/ OB
reason for delivery

• Initial question to Mother/ OB resident/ OB
  Nurse, if admitted from 36 0/7- 38 6/7 Why did
  you come in to the hospital to deliver at this
  time? met with resistance from OB attendings,
  so
• A survey questionnaire was developed and
  distributed to the obstetricians in the region to
  clarify the rationale and reasons for scheduling
  deliveries prior to 39 weeks

52
OB Questionnaire
53
OB questionnaire

• Distributed to OBs at regional hospitals during
  OB Grand Rounds
• ACOG guidelines reviewed regarding timing of
  elective deliveries
• Questionnaires collected at end of meeting
• Total of 74 responses (out of 90 regional OBs)
  were obtained

54
OB Response Reported Frequency Scheduling
Elective Deliveries
55
(No Transcript)
56
OB reported experience with adverse outcomes with
elective delivery prior to 39 weeks
n74
57
OB report of practice
58
Community Educational, QA/QI Venues

• Formal presentations at Perinatal Outreach visits
  in the Finger Lakes region on subject presented.
• University of Rochester CME conference, "The
  Risks of Late Preterm Delivery highlighting the
  medical risks of delivery prior to 39 weeks.
• Discussions with Chairs of OB Departments in the
  community- to date 2 of 4 larger delivery service
  hospitals have adopted a protocol related to
  timing of elective deliveries (one listing 39
  weeks, one 38 weeks)

59
Community Educational, QA/QI Venues cont.

• Discussions one- on- one with private practice
  OBs- questionnaires to ancillary medical
  providers was met with extreme resistance and
  needed to be prematurely aborted so general
  questionnaires to OBs being used.
• Routine feedback loop established. If a woman is
  delivered electively prior to 39 weeks and the
  infant requires additional support, a letter is
  sent directly to the OB provider describing the
  nature of the infant's needs and reinforcing the
  ACOG guidelines to avoid elective delivery prior
  to 39 weeks without evidence of lung maturity-
  already seeing a decrease in need for follow-up
  letters!

60
Feedback letter template
Dear Dr. Recently a patient, _____ , was
admitted to the NICU at xxx/ SCN at xxx due to
___________________. The infant required
______________. Review of the records indicate
that the baby was delivered on _________ by
scheduled elective induction/ cesarean section at
_____ weeks. In our QA/QI role as the Regional
Perinatal Center we are reminding providers of
The American College of Obstetricians and
Gynecologists Bulletin on elective delivery
which outlines that non-emergent deliveries
should not be scheduled before 39 weeks without
documentation of lung maturity by amniocentesis.
This is especially true for planned cesarean
sections which are associated with higher
respiratory morbidity than in infants born after
labor. If there is a medical/obstetrical
indication to deliver earlier than 39 weeks, it
should be documented in the chart that the risks
of early delivery are outweighed by the
anticipated benefits. Although obstetricians
often are lulled into a false sense of security
by the fact that most infants born at 37-38 weeks
do well, population studies demonstrate higher
risks than at 39 weeks, including a higher rate
of need for ECMO (extracorporeal life support or
heart-lung bypass) our NICU receives babies
every year with significant respiratory morbidity
associated with elective delivery at 37-38 weeks.
Because such gestational age dependent morbidity
can be eliminated by following ACOGs guidelines,
purely elective deliveries before 39 weeks
without documenting lung maturity in advance
should be avoided. Granted, some morbidity
occurs even at 39 weeks or with mature lungs,
but avoidance of scheduled elective deliveries
before 39 weeks, although not a guarantee of good
outcome, is an ACOG standard of care. It is
possible that the NICU records do not include
some portion of the patients medical history
that justified delivery before 39 weeks without
documentation of lung maturity. If so, please
bring these details to our attention. Respectfully
, XXX
61
Informational trifold pamphlet for families- page
1
62
Informational trifold pamphlet for families- page
2
63
Using SPDS to track delivery practices

• 1. Main Method of Delivery, route_main
• Vaginal (includes forceps_low/outlet,
  forceps_mid, spontaneous, vacuum)
• Cesarean Section
• 2. Trial of Labor for C-Section, trial_lab
• Yes
• No
• 3. Induction combination of two variables
• ind_arom Induction of Labor AROM (Yes/No)
• ind_med Induction of Labor Medical
  (Yes/No)
• 4. Premature Rupture of Membrane (1 hr), prom
• Yes
• No

64
Statistical analysis of sensitivity, Specificity,
Positive Predictive Value and Negative Predictive
Value

• Using 2007 chart review data, develop modeling
  equation to take into account low incidence
  diagnoses

65
Step 1 Determine if a woman presented in labor
(Yes/No) by combining variables 1-3.

• Yes Vaginal Delivery No Induction (AROM or
  Medical)
• Yes CS w/ Labor No Induction
• No Vaginal Delivery Induction
• No CS w/ Labor Induction
• No CS w/out Labor

66
Step 2 Determine if woman had PROM by using
variable 4 directly.
67
Step 3 Determine crude elective delivery
prevalence

• ED Elective Delivery did not present in
  labor no PROM
• MI Medically Indicated presented in labor,
  or did present in labor did have PROM
• Crude ED Prevalence ( ED)/(total births in
  sample)

68
Step 4 Adjust elective delivery prevalence for
the sensitivity specificity of the definition.

• Pcorrected Pobserved Sp 1/Se Sp 1
• Pcorrected Pobserved 0.22/0.712
• Example Suppose, in 2009, we estimate the
  prevalence of elective deliveries to be 44 using
  the simple definition above.
• ? Pobserved 0.44
• Pcorrected 0.44-0.22/0.712 0.309 30.9
• Always use the known sensitivity and specificity
  rates for 2007 because this is the sample the
  definition was developed from.

69
Summary

• Elective deliveries before 39 weeks are common
  (currently 40 of deliveries 36 0/7-38 6/7 at
  FLRPC), with concurrent increased morbidity and
  potentially mortality
• What is reported as being done and what is done
  are not always the same (eg performing amnio for
  fetal lung maturity)
• Increased education and white ethnicity appear to
  be related to increased elective delivery prior
  to 39 weeks, suggesting a potential target
  population for educational focus
• A multispecialty, multifaceted approach founded
  in data may be effective in affecting change in
  decreasing this practice (and if not, having it
  as a quality indicator may)
• Remember the 3 Ps- You can change practice, but
  it takes patience, persistence, and a pleasant
  approach
• Accuracy of SPDS coding is key to determining
  efficacy of educational programs in decreasing
  elective deliveries before 39 weeks

70
Babies do not chose when to be born- we need to
be their advocates
71
Acknowledgements
Deb Pittinaro Taha BenSaad Claire
Hoffmire Tim Stevens Kathryn Clark
Chris Glantz Keri Cockman