2/15/2018

1

Neonatal Early Onset SepsisRisk based approach to management

Anette Kunig, MD

February 6, 2018

Neonatal Early Onset Sepsis (EOS)

• Definition

• Invasive bacterial infection of the blood and/or CSF within the first week after birth

• Pathogenesis

• Normal GI and GU flora

• Ascending colonization maternal genital tract and uterus

• Colonization and infection of the fetus/newborn

• Presentation

• Sick infant requiring respiratory and/or hemodynamic support

• Nonspecific signs also associated with normal transition to extrauterine life

2/15/2018

2

Neonatal SepsisMajor Advances Over Last 30 Years

Primarily Driven OB/GYN Innovations

• Advances

• Prenatal

• CDC: Universal GBS carrier ID

• CDC: Aggressive intrapartum antibiotic prophylaxis and chorioamnionitistreatment

• Postnatal

• Improved management of sick neonates

• Effects

• 10 to 100-fold drop in early onset sepsis rate: currently 3/10,000

• 2 to 20-fold drop in sepsis mortality: currently as low as 2% in term babies

February 15, 2018 3

Benitz 2015, Schrag 2000, Bromberger 2000, Escobar 2014, Braun 2015

Signs of neonatal sepsis?

Blood Cx , CBC, and LPAbx Treatment

Maternal chorioamnioniti

s?

Blood Cx and CBCAbx Treatment

Observation for ≥ 48 hrs

Observation for ≥ 48 hrs

+/- CBC @ 6-12 hours

Blood Cx and CBCObservation for ≥ 48

hrs

< 37 weeks orROM ≥18 hrs?

Routine Clinical Care

≥ 37 weeks andROM < 18 hrs?

GBS prophylaxis indicated?

GBS Tx ≥ 4 hrsbefore delivery?

YES

YES

YES

NO

YES

YES

NO

NO

NO

YES

NO

Who to evaluate and who to treat:

2010 CDC Prevention of GBS

Algorithm- choriomnionitis /

maternal fever

2/15/2018

3

Critique of CDC recommendations

• Good:

• Prenatal management

• Large reduction in incidence of EOS

• Not so good:

• Expert opinion, not efficacy studies

• Based on prevention of neonatal group B strep sepsis only

• Identification of carriers

• Intrapartum antibiotic prophylaxis

5

Critique of CDC recommendations

• Not so good

• Based on data prior to 2002

• No change between 2002 and 2010 in CDC recommendation

• Before widespread use of intrapartum antibiotic prophylaxis

• EOS risk 5-10 fold higher

• Decreased incidence of sepsis

• More data on poor positive predictive value of CBC

• Large number infants being evaluated for sepsis, 15-20%

• Large number infants receiving empiric antibiotics, 5-8%

2/15/2018

4

Critique of CDC recommendations

• Don’t account for newborns clinical presentation

• Newer algorithms developed for evaluation and treatment of at-risk infants

• Well, equivocal, and clinically ill

• Reduction in evaluation and treatment

• Persistent high rates of treatment in a setting where the incidence of EOS is decreased

Imbalance between Infection and Sepsis Evaluation / Antibiotic Treatment

EOS incidence

0.3/1000 Live Births

Blood Cultures 14.4%

Antibiotics 5%

Data from KPNC

2010-2012

Disturbed maternal bonding

Disturbed microbiome

2/15/2018

5

EOS CalculatorInitial probability

based on maternal risk factors

EOS CalculatorBaseline risk modified by clinical presentation

2/15/2018

6

Clinical Presentation Risk Groups Studied for Likelihood Ratios

Clinical

Presentation

Description

Clinical illness 1. Persistent need for NCPAP / HFNC / mechanical ventilation (outside of the delivery room)

2. Hemodynamic instability requiring vasoactive drugs

3. Neonatal encephalopathy /Perinatal depression• Seizure

• Apgar Score @ 5 minutes < 5

4. Need for supplemental O2 > 2 hours to maintain oxygen saturations > 90% (outside of the delivery room)

Equivocalpresentation

1. Persistent physiologic abnormality > 4 hrs

• Tachycardia (HR > 160)

• Tachypnea (RR > 60)

• Temperature instability (> 100.4˚F or < 97.5˚F)

• Respiratory distress (grunting, flaring, or retracting) not requiring supplemental O2

2. Two or more physiologic abnormalities lasting for > 2 hrs

Well appearing • No persistent physiologic abnormalities

EOS CalculatorBaseline risk modified by clinical presentation

2/15/2018

7

Implementing the EOS Calculator and applying recommendations based on risk

EOS risk

(per 1000)

Intervention Observation Location

<0.4 none routine FCC

0.4-0.9 none Q4h x 24h FCC

1-3 Blood culture Q4h x 24h FCC

>3 Antibiotics and Blood culture

Per NICU NICU

SET TREATMENT THRESHOLDS

Clinical recommendations for evaluation and treatment

2/15/2018

8

Study Characteristics

• 204,485 infants January 1, 2010 to December 31, 2015

• Northern California Kaiser patients

• Gestational age 35 weeks or later

• 3 study periods

• Baseline period:

• clinical care based on CDC guidelines

• Learning period

• EOS calculator incorporated using only maternal data

• clinical presentation of infant not incorporated

Study Characteristics

• 3 study periods

• EOS Calculator period

• Newborn’s clinical presentation was incorporated

• Well, equivocal, clinically ill

• Recommendations made based on probability of EOS calculated from maternal data and clinical presentation

• Observation: routine versus close vital signs

• Evaluation with blood culture

• Empiric antibiotics

2/15/2018

9

Study ResultsSepsis evaluation

Study ResultsAntibiotic Use

2/15/2018

10

Study Results

• No increase in EOS between the baseline period and the EOS calculator period

• Infants with sepsis are not being missed by decreasing rates evaluation and treatment

• No increase in antibiotic usage over 24-72 hour across all study periods

• Antibiotics weren’t being delayed until infants showed signs of illness

Conclusions

• CDC guidelines have greatly reduced the incidence of EOS

• Developed prior to widespread use of intrapartum antibiotics

• Empiric antibiotics for all newborns with maternal chorio ( temperature >38C) regardless of clinical condition

• imbalance between the incidence of EOS and percentage of infants having sepsis evaluations or receiving antibiotics

• EOS calculator uses clinical appearance of infant to stratify maternal risk factors

• Decreases sepsis evaluations

2/15/2018

11

Conclusions

• Decreases empiric antibiotic use

• Asthma, autoimmune disorders, obesity

• No difference in antibiotics use at 24-72 hours

• No increase in early onset sepsis (positive cultures within first 7 days)

Limitations

• Larger studies needed to evaluate safety of delaying antibiotic treatment until infant becomes symptomatic

• no adverse events seen in study

• Scenario is uncommon

• Risk of delaying treatment weighed against treating large number infants unnecessarily

• Prior probability EOS in various insitutions may differ from what was used in this study

• Risk threshold for clinical interventions can be adjusted based on center characteristics

• Lower risk threshold in centers with poor prenatal care, low rates of GBS screening, lack of peds follow up, etc

2/15/2018

12

Link to calculator

• https://neonatalsepsiscalculator.kaiserpermanente.org/

• App store: EOS Calculator

Examples

• 37 weeks

• Maternal temp 102.8 with pushing / not diagnosed with chorio

• ROM 34 hours

• GBS negative

• Intrapartum antibiotics just prior to delivery

• Equivocal: tachypneic

2/15/2018

13

Examples

• 40.2 weeks

• Maternal temp 100.8

• ROM 5 hours

• GBS negative

• Amp and gent 3 hours PTD

• (broad spectrum)

• Maternal chorio diagnosed

• Maternal fever, fetal tachycardia

• Well infant

Thank you

2/15/2018

14

Elements of Newborn

EOS Calculator