01 Ehrenkranz Impact of Growth on Neonatal … of Growth on Neonatal Outcomes ... Yale Estimated...
Transcript of 01 Ehrenkranz Impact of Growth on Neonatal … of Growth on Neonatal Outcomes ... Yale Estimated...
1
Impact of Growth on Neonatal Outcomes
Richard A. Ehrenkranz, MDProfessor of Pediatrics
Yale University School of Medicine
Disclosure StatementsRichard A. Ehrenkranz, MD
I have no relevant financial relationships to disclose or conflicts of interest to resolve.
I will not discuss any unapproved or off-label, experimental or investigational use of a product, drug, or device.
Objectives● To discuss curves used to monitor postnatal
growth of preterm infants, especially extremely preterm (EPT) infants
● To review the data that: postnatal growth reflects the adequacy of nutrient postnatal growth reflects the adequacy of nutrient
intake growth correlates with neurodevelopmental
outcomes
● To recommend the implementation of feeding guidelines
Plotting Serial Anthropometric Measurements on Growth Curves
• Provide an assessment of nutritional status
• Permit comparisons with established norms
• Identify alterations in growth patterns Symmetric vs asymmetric Growth along a percentile vs
faltering Growth below the 10th percentile
• Assess response to interventions
Nutritional Management of Preterm Infants: Current Recommendations
To provide nutrients to approximate the rate of growth and composition of weight gain f l f t f thfor a normal fetus of the same postmenstrual age and to maintain normal concentrations of nutrients in blood and tissue.
AAP CON. Pediatric Nutrition Handbook, 6th ed. 2009; p.79.
2
Growth Curves and Postnatal Growth of Preterm Infants
Controversy exists over feasibility and appropriateness of achieving AAPCON goalgoal.
Descriptive curves based upon:Intrauterine growth dataPostnatal growth data
3000
3500
4000
4500
s)
Yale Estimated Fetal Weight Growth Curve
0
500
1000
1500
2000
2500
22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40
Fet
al W
eig
ht
(gra
ms
Gestational Age
3rd %
50th %
97th %
IU Growth from Live Births at 26-42 Weeks’ Gestation
Lubchenco. Pediatr 1966; 37: 403-408
IU Growth from Live Births at 25-44 Weeks’ Gestation
Mean + 2 sd
Usher, McLean. J Pediatr. 1969; 74:901-910
A New Growth Chart for Preterm Infants
Fenton. BMC Pediatrics, 2003.
New IU Growth Curves Based on US Data
Olsen. Pediatrics, 2010.
3
Limitations of Intrauterine Growth Curves for Monitoring Postnatal Growth
Curves represent cross-sectional data derived from newborns, not serial measures over time
Preterm infants may not be “normal”
Gestational age dating is imprecise
Curves maybe population specific, and not generalizable
Postnatal Weight Changes in Low Birth Weight Infants
Shaffer. Pediatrics 1987; 702-705
Longitudinal Growth of VLBW Infants
1500
1750
2000
ram
s)
501
601
701
801
901
100
1 1
101
120
1
130
1
140
1
Postnatal Age (Days)
1 14 28 42 56 70 84 98
500
750
1000
1250
Wei
ght
(gr
NICHD Growth Observational Study, 1999
Ehrenkranz. Pediatr 1999
Longitudinal Growth of VLBW InfantsLongitudinal Growth of VLBW Infants
40
45
50
cm)
501 601 701 801 901 1001 1101
1201 1301
1401
Postnatal Age (Weeks)
1 2 3 4 5 6 7 8 9 11 13 15 17
25
30
35
Leng
th (
c
NICHD Growth Observational Study, 1999NICHD Growth Observational Study, 1999
Ehrenkranz. Pediatr 1999
Longitudinal Growth of VLBW InfantsLongitudinal Growth of VLBW Infants
30
35
renc
e (c
m)
501
601
701
801
901
1001
1101
1201
1301
1401
Postnatal Age (Weeks)
1 2 3 4 5 6 7 8 9 11 13 15 17
20
25
Hea
d C
ircum
fer
NICHD Growth Observational Study, 1999NICHD Growth Observational Study, 1999
Ehrenkranz. Pediatr 1999
4
Postnatal Growth of VLBW Infants vs Expected Intrauterine Growth
1500
2000
ram
s)
50th 10th
Postmenstrual Age (Weeks)
Intrauterine growth24-25 weeks26-27 weeks28-29 weeks
24 28 32 36
500
1000
Wei
ght
(gr
•
•
•
NICHD Growth Observational Study, 1999NICHD Growth Observational Study, 1999
Ehrenkranz. Pediatr 1999
Postnatal Weight Gain After Very Preterm Birth: A UK Population Study
Postnatal growth of 2577 infants < 32 wks GA plotted on UK-WHO growth charts. (Cole. ADC/RCPCH Abst. 2011.)
Embleton, et al. Pediatrics 2001; 107:270-3.
RCT: Aggressive Nutritional Support in VLBW Infants
Intervention Control(n=64) (n=61)
Protein 0.5 g/kg/d (d1) 0.5 g/kg/d (d3)to 3.5 g/kg/d to 2.5 g/kg/d
Lipid 0.5 g/kg/d (d2) 0.5 g/kg/d (d5)to 3.5 g/kg/d to 2.0 g/kg/d
Enteral 0.5 mL/hr (d1) started when daily, rarely d/c stable
Wilson, et al. Arch Dis Child. 1997; 77;F4-F11.
RCT: Aggressive Nutritional Support in VLBW Infants
Intervention Control(n=64) (n=61)
Age regain BW 9 d 12 d* NEC 8 % 7 %Bacteremia 50 % 66 %CoNS bacteremia 44 % 56 %At d/c or death:
Body wt < 10th% 59 % 82 %*Length < 10th% 56 % 74 %HC < 10th% 14 % 30 %
[* p<0.05]
Wilson, et al. Arch Dis Child. 1997; 77:F4-F11.
RCT of Aggressive Nutritional Support in VLBW Infants
Wilson, Arch Dis Child 1997
5
Early & Aggressive Nutrition Strategy in VLBW Infants: Historical Control Study
Aggressive Conventional(n=117) (n=65)
Protein 1.5 g/kg/d (hrs) 0.5 g/kg/d (d3)to 4.0 g/kg/d to 3.0 g/kg/d
Lipid 0 5 g/kg/d (24 hrs) 0 5 g/kg/d (d3-4)Lipid 0.5 g/kg/d (24 hrs) 0.5 g/kg/d (d3-4)to 3.5 g/kg/d to 3.0 g/kg/d
Enteral 10 mL/kg/d HM or started when PF (d1), by 10 stable; mL/kg/d x 7 d, then advanced per15-20 mL/kg/d to team.180 mL/kg/d.
Dinerstein, et al. J Perinatolog 2006; 26:436-442.
Early and Aggressive Nutritional Strategy (Parenteral & Enteral) Decreases Growth
Failure in VBLW Infants
Dinerstein, et al. J Perinatology 2006; 26: 436-442.
Early & Aggressive Nutrition Strategy in VLBW Infants: Historical Control Study
Aggressive Conventional(n=117) (n=65)
Age regain BW 10 d 12 d* Age to full EN 15 d 20 d*NEC 2 % 3 %L t t i 26 % 23 %Late-onset sepsis 26 % 23 %At 40 wks PMA:Body wt < 10th% 53 % 77 %*Body wt (median) 2.95 kg 2.70 kg*Length (mean) 46.6 cm 45.6 cm*HC (mean) 35.2 cm 34.3 cm*[* p<0.01]
Dinerstein, et al. J Perinatology. 2006;26:436-442.
Neurodevelopmental Outcomes at 18-22 mos by Weight Gain Quartile
40%
50%
60%
ents
wit
h
me
MDI <70 PDI <70 Impairment
0%
10%
20%
30%
12.0 (2.1) 15.6 (0.8) 17.8 (0.8) 21.2 (2)
Weight Gain Groups (grams/kg/day)
Per
cen
t o
f P
ati
Ou
tco
m
Ehrenkranz. Pediatrics 2006:117:1253-61
Cerebral Palsy at 18-22 mos by Weight Gain Quartile
20%
25%
wit
h O
utc
om
e
0%
5%
10%
15%
12.0 (2.1) 15.6 (0.8) 17.8 (0.8) 21.2 (2)
Weight Gain Groups (grams/kg/day)
Per
cen
t o
f P
atie
nts
w
Ehrenkranz. Pediatrics 2006:117:1253-61
Growth Outcomes at 18-22 mos by Weight Gain Quartile
Ehrenkranz. Pediatrics 2006:117:1253-61
6
Early Provision of Parenteral AA in ELBW Infants: Growth Outcomes at 36 wks PMA
Early
n=182
Late
n=836
OR (95% CI)
p-value
Weight (gms) 1958 ± 383 1819 ± 320 <0.0001
Wt <10th %tile 127 (82%) 681 (95%) 4.2 (2.4-7.5)
Wt <5th %tile 108 (70%) 605 (85%) 2.1 (1.4-3.2)
Length 41.7 ± 2.7 41.0 ± 2.3 0.0108
HC 30.9 ± 1.8 30.3 ± 1.6 <0.0001
Poindexter et al. J Pediatr 2006; 148:300-305
Head Circumference at 18-22 mos CA
* P < 0.05 Poindexter et al. J Pediatr 2006; 148:300-305
Neurodevelopment at 18-22 mos CA
Early
(n=182)
Late
(n=836)
Adjusted OR
(95% CI)
# Evaluated 154 (85%) 712 (85%)
Bayley MDI<70 33% 32% 0.9 (0.6-1.4)
B l PDI<70 17% 19% 1 1 (0 6 1 9)Bayley PDI<70 17% 19% 1.1 (0.6-1.9)
Cerebral Palsy (any)
17% 14% 0.6 (0.4-1.0)
Hearing Disability 3% 3% 1.1 (0.4-3.3)*
Blindness (L or R) 0% 1% (p=0.224)*
NDI 32% 35% 1.0 (0.6-1.5)
*Unadjusted Poindexter et al. J Pediatr 2006; 148:300-305
First-Wk Protein Intake Are Associated with 18-mos CA Developmental Outcomes in ELBW Infants
[Stephens. Pediatrics. 2009]
2.42.6
2.9
2
2.5
3
3.5
4
g/da
y
0.4
1
1.5
2
0
0.5
1
1.5
2
1 2 3 4 5 6 7Day of Life
mea
n g/
kg
124 infants born 2000-2001. Wk #1 mean protein intake = 1.8 g/kg/day
First-Wk Protein Intakes Are Associated with 18-mos CA Developmental Outcomes in ELBW Infants
[Stephens. Pediatrics. 2009]
Regression analyses controlling for BW, gender, IVH, BPD, NEC and maternal education demonstrated:
Each 1 gm/kg/d increase in protein intake: ▪ 8.2-point increase in MDI
A t d f 17% f th i i MDI▪ Accounted for 17% of the variance in MDI
Each 10 kcal/kg/d increase in energy intake:▪ 4.6-point increase in MDI ▪ Accounted for 18% of the variance in MDI
Intrauterine, Early Neonatal, and Postdischarge Growth & Neurodevelopmental Outcome at 5.4 Yrs in Extremely Preterm Infants After Intensive
Nutritional Support►Franz, et al. Pediatrics 2009; 123:e101-e109● Study Design: 219 (83%) of 263 surviving VLBW
infants born between July 1996-June 1999 were evaluated at about 5 yrs of age with:
1) standardized neuroexam2) Kaufmann Assessment Battery for Children3) Gross Motor Function Classification Scale
● All infants received “intensive early nutritional support” that included initiation of parenteral protein (2 g/kg/d) on day 1 and enteral feeds (~16 mL/kg/d) on day 1. Parenteral protein was increased to 3 g/kg/d; enteral feeds by ~ 16 mL/kg /d.
7
Contribution of Perinatal Risk Factors on Mental Processing Composite from
Multiple Linear Regression Models
Parameter (in weight model) Partial R2 P-value
IVH > 3 vs < 3 0.213 < 0.001
Duration of MV, > 7 vs < 7 d 0.113 < 0.001Measure of weight gain
SDS t bi th0.035
0 005SDS at birth∆SDS (D/C – birth)∆SDS (F/U – D/C)
0.0050.0470.884
PVL, yes vs no 0.020 0.029Highest maternal education,
lowest vs higher sch deg 0.016 0.009Total R2 of model 0.396
Franz. Pediatrics 2009; 123:e101-e109
Infant Growth Before & After Term:Effects on Neurodevelopment in Preterm Infants
► Belfort, et al. Pediatrics 2011; 128:e899-e906● Design: Observational study of 613 infants < 33 wks GA
born between 4/01-10/05; participants of DINO (DHA for Improvement of Neurodevelopmental Outcome) Trial
1) Anthropometrics at term, 4, 12, & 18 mos CA2) Bayley Scales of Infant Development-II at 18 mos CA2) Bayley Scales of Infant Development-II at 18 mos CA
● Findings:1) Between 1 wk-term, greater wt gain → higher BSID scores 2) Between 1 wk-term w/ BW < 1250 g, stronger associations w/MDI & PDI
3) Between term - 4 mos, greater wt gain & linear growth→ higher PDI scores
4) Between 4 – 12 mos, no growth associations with BSID
Infant Growth Before & After Term: Effects on Neurodevelopment in Preterm Infants
Belfort. Pediatr 2011; 128:e899-e906.
Neonatal Caloric Intake Influences Preterm Infant Head Growth
● AGA infants calorically deprived (< 85 kcal/kg/d) for > 4 wks (SGA infants for > 2 wks) had HC growth -1 to -2 SD below standard HC growth curves at 12 mos CA; AGA infants also had motor developmental delay at 12 mos CAdelay at 12 mos CA.
Georgieff, J Pediatr 1985; 107:581-587
● Early high-energy nutrient intake (> 95 kcal/kg/d) between postnatal days 2-10 by VLBW, SGA infants promoted catch-up HC growth & significantly higher DQ/IQ scores from 6 mos to adulthood, when compared to SGA infants w/o catch-up HC growth.
Brandt, J Pediatr 2003; 142:463-468
HC (m+sd) of AGA preterm infants compared with SGA HC catch-up group (CU) & group w/o HC catch-up (no-CU) from 35 wks PMA to adulthood. (Brandt, J Pediatr 2003; 142:463-468)
DQ/IQ (m+sd) of AGA preterm infants compared with SGA HC catch-up group (CU) & group w/o HC catcup (no-CU) from 6 mos to adulthood. (Brandt, J Pediatr 2003; 142:463-468)
8
Improving Head Growth in Preterm Infants: A RCT
►Objectives: To explore the relationships between early nutrition, postnatal head growth, quantitative brain MRI, and developmental outcome during the 1st yr of life among infants born < 29 wks gestation.
►Methods: Infants were randomized to hyperalimented or standard parenteral and enteral nutrition. Growth was monitored from birth to 36 wks PMA.
Primary outcome: OFC at 36 wks PMA.Secondary outcomes: Quantitative MRI at 40 wks
PMA; Bayley Scales at 3 & 9 mos post-term
Tan. Arch Dis Child Fetal Neonatal Ed, 2008
RCT: Improving Head Growth in Preterm Infants < 29 wks’ GA
Hyperalimented(n=68)
Control(n=74)
PN compositionEnergy (kcal/kg/d)
Glucose (g/kg/d)
Protein (g/kg/d)
11716.3
4
9313.5
3Protein (g/kg/d)
Fat (g/kg/d)
44
33
Age PN initiated < 24 hrs < 24 hrs
PN advancement Stepwise from 1 g/kg/d P & F to 4 gm/kg/d by d7
Stepwise from 1 g/kg/d P & F to 3 gm/kg/d by d5
Enteral Nutrition EBM, preferably w/in 48 hrs;
↑ by 6-12 mL/d
EBM, preferably w/in 48 hrs;
↑ by 6-12 mL/dTan. Arch Dis Child Fetal Neonatal Ed, 2008
Clinical, Nutritional & Growth Outcomes of All Survivors
Intervention (n=55)
Control(n=59)
Days to regain BW 10.3 (6.1)* 13.9 (6.3)***
Median days on PN 17 (12) 12 (12)
Median days to FEN 19 (11) 15 (11)
Median # sepsis evts/pt 2 (2) 1 (2)
@Oxygen @ 36 wks PMA 40 36
NEC 6 6Energy intake @ 4 wks 2766 (233) 2621 (191)**Protein intake @ 4 wks 73 (7) 64 (5)***
OFC @ 36 wks (cm) 31.1 (1.5) 31.4 (1.3)
Length @ 36 wks (cm) 42.9 (2.3) 42.4 (2.1)
Weight @ 36 wks (g) 2136 (345) 2090 (293)
*Mean (sd); **p<0.01; ***p<0.001
Tan. Arch Dis Child Fetal Neonatal Ed, 2008
Results: 80% of the infants in the Intervention group vs 97.8% of infants in the Control group were still in overall energy and protein deficit after 4 wks. Therefore, pooled data were used in these scatter plots.
Tan. Arch Dis Child Fetal Neonatal Ed, 2008
Improving Head Growth in Preterm Infants: Results
● After pooling the study groups, significant correlations existed between:▪ Energy intake & energy deficit during the first 28 d
and:▫ TBV @ 40 wks PMA@▫ MDI & PDI @ 3 mos post-term,
but not 9 mos post-term▪ Anthropometrics @ 36 wks PMA and TBV & CBV
@ 40 wks PMA
▪ Weight @ 36 wks PMA and both mental and motor outcomes during year 1
Tan. Arch Dis Child Fetal Neonatal Ed, 2008
Poor Linear Growth Velocity & 2 Yr Neurodevelopmental Outcomes in VLBW Infants
Cognitive scores at 24 mos CA: Not associated with WZ at any time points Associated with HCZ at birth and at 4 and 24 mos CA Associated with LZ at 4 and 12 mos CA
Ramel. Neonatology 2012; 1o2:19-24.
9
Objectives● To discuss curves used to monitor postnatal
growth of preterm infants, especially extremely preterm (EPT) infants
● To review the data that: postnatal growth reflects the adequacy of nutrient
intake growth correlates with neurodevelopmental
outcomes
● To recommend the implementation of feeding guidelines
Implement Feeding Guidelines to Minimize Interruption of Antenatal Nutritional Support
● Feeding guidelines should include:
▪ “Consensus” early PN and EN evidence-based strategies (ie, potentially better practices)p )
▪ Standardized plan to manage “feeding intolerance”
▪ Strategy to maintain steady rate of postnatal growth, adjusting nutritional support if growth parameters not met
Stages of Nutritional Support After Birth of Extremely Preterm Infants
● Initiation
● Transition to full enteral nutrition (FEN)
● Maintenance of steady growth rate on FEN in the NICU
● Post-Discharge nutrition
Stages of Nutritional Support: Initiation
● Target GIR ~ 6 mg glucose/kg/min● Early initiation of parenteral AA
▪ Start within hrs of birth▪ Provide as a “To Deliver” amount
1 / / ( )▪ 1.5 g/kg/d (minimum)
● Total Fluid for first 24 hrs to provide 80-100 mL/kg/d
▪ 50 mL/kg/d of a “starter” (“vanilla”, off-hrs”) PN
▪ 30-50 mL/kg/d of glucose, minimal electrolyte soln
Off-Hours TPN
● Glucose 10 mg/dL (D10W)
● Protein (amino acids) 6.0 g/dL
● Minimal electrolytes
● Ca++ 100 mg/dL
● No vitamins, minerals, lipids
● @ 50mL/kg/d: GIR is 3.5 mg/kg/min, protein delivered is 3.0 g/kg/d, & energy delivered is 23 kcal/kg/d
Stages of Nutritional Support: Transition to Full Enteral Nutrition
● Optimize parenteral nutrition (PN)▪ Increase protein intake to ~ 4 g/kg/d
◦ Daily increases of 0.5-1.0 g/kg/d▪ Increase GIR to ~ 10 mg/kg/min during first week of lifeduring first week of life
▪ Initiate IFE within 24 hrs of birth◦ Start with at least 0.5 g/kg/d ◦ ↑by 0.5-1.0 g/kg/d to ~ 3.0 g/kg/d
▪ Increase % Total Daily Fluids (TDF) as PN over 1st 3-4 days of life
10
Stages of Nutritional Support: Transition to Full Enteral Nutrition
● Initiate Minimal Enteral Nutrition [(MEN); trophic feeds, gi stim feeds]
▪ Start within first 24 - 96 hrs of age▫ Colostrum, if possible
▪ Select duration of MEN▪ Select duration of MEN▫ 24 hrs vs several days
▪ Select volume of MEN▫ ~ 10% total daily fld intake (~ 12 mL/kg/d)
▪ Select rate of advancement to FEN▫ ~12 mL/kg/d▫ Fortify HM @ ~ 100 mL/kg/d
Stages of Nutritional Support: Maintenance of Growth on FEN
● Monitor growth▪ Weight gain ~20 gm/kg/d (over 5-7 days)
▪ Length ~ 1 cm/wk
▪ HC ~ 1 cm/wk
● Identify causes of decreased rates of growth▪ Energy needs > energy intake
◦ Fluid restriction (eg, BPD)
◦ Malabsorption◦ ↓ intake during transition to nipple feedings
▪ ↓ protein content of fortified MOM (or DM)
● Maintain adequate energy & protein intakes▪ Energy intake ~ 120 kcal/kg/d (minimum)
▪ Protein intake : ~ 4.0 gm/kg/d < 30 wks PMA [P/E=3.3]~ 3.5 gm/kg/d > 30- 36 wks PMA [P/E=2.9]
Stages of Nutritional Support:Post-Discharge Nutrition
● Special nutritional requirements persist after D/C
▪ Fortified HM
▪ Nutrient-enriched post-D/C formula
● Monitor growth with WHO growth curves from term CA to 24 mos CA
Implementation of Feeding Guidelines Leads to Improved Outcomes, including:
▪ BW regained sooner
▪ FEN achieved sooner (reduced need for PN)
▪ Cumulative energy & protein deficits reduced
Decreased rates of late onset infection & NEC▪ Decreased rates of late-onset infection & NEC
▪ Improved anthropometrics at 36 wks PMA and at discharge (eg, fewer infants < 10th %tile
▪ Mediation of severity of illness
▪ Reduction of hospital stay
Summary and Conclusions● Extrauterine growth restriction (EUGR) may be
unavoidable for many EPT infants.
● Since postnatal growth reflects nutrient adequacy, growth must be monitored.
● Implementation of evidence-based feeding guidelines leads to improved outcomes.
● Postnatal growth is correlated with growth and neurodevelopmental outcomes in early childhood.