Necrotizing

Enterocolitis: An

Update

Josef Neu, M.D.

A

C

B

“Classic” NEC

Neu, J. and Walker , W. A. New England Journal of Medicine, Jan. 2011

Agenda

• Background

• Different forms of “NEC”

• Pathogenesis: “Classic NEC”

• Can we develop predictive and/or

diagnostic biomarkers?

• Preventative Strategies

Necrotizing Enterocolitis: Scope

of the Problem

• In US and Canada, it affects approximately 7% of babies weighing between 500-1500 grams (approximately 20-30% die).

• Approximates the number of children fewer than 15 years of age who die of leukemia or meningitis.

• Survivors may be left with significant sequelae.

• Very costly.

• Scares neonatologists!

Being led astray: 50 years---

not much progress • Lumping of several

diseases called

“NEC” into the same

data set.

• Animal models that

do not represent the

disease.

• Human surgical

samples that

probably represent

degrading tissue.

More than one disease or one

disease with many origins?

• Ischemic intestine—cardiogenic,

polycythemia, severe IUGR with

abnormal Doppler flow.

• Intestinal anomalies---

Hirschprung’s, malrotation,

intussusception, etc.

• Spontaneous Intestinal

Perforations (SIPs)

More than one disease or one

disease with many origins?

• Variant of food protein induced

enterocolitis syndrome (FPIES)

• pRBC infusions

• Inflammatory, Dysbiosis---

”Classic” NEC in preterms.

Right after birth 12 hours later

Isolated Non-NEC Perforations

• Occurs early.

• Not related to feedings (except for

possibly the lack of feedings).

• Related to combined early

dexamethasone (or

hydrocortisone)and indomethacin (Stark, et al. NEJM,2001;Gordon, et al. J.

Perinatol., 1999: Watterbert, 2004).

Is there a Clear Definition of

NEC? Bells is Broken • Stage 1-Too non-

specific and the term

should not be used.

• Stage 3- Could

signify intestinal

necrosis or

Spontaneous

Intestinal Perforation

What Causes NEC?

Postmenstrual Age Difference

Neu, J. Acta Paediatrica, 2005

94 (Supple 449): 100-105 A.R. Llanos, et al., Paediatr Perinat

Epidemiol ,2002 16 (4): 342–349.

NEC: Relation to Inflammatory

Mediators

• Increased plasma

pro-inflammatory

cytokine (IL-8. Il1,

IL-6, TNF-α )

concentrations.

• Increased platelet

activating factor.

A

C

B

“Classic”NEC: Do the

Microbiota Differ?

Neu, J. and Walker , W. A. New England Journal of Medicine, Jan. 2011

FECAL MICROBIOTA: NEC Mai V, Young C. PLOS One, May 2011

• Proportions of the four major phyla two

weeks before and the week of diagnosis

Microbial Shift Prior to NEC

From Claud, E. et al . Microbiome, 2013

Proteobacteria

• Major Phylum of bacteria

• Includes a wide variety of pathogens (E.coli,

Salmonella, Vibrio, Helicobacter, Klebsiella,

Pseudomonas,etc.).

• Gram negative with outer membrane composed

of lipopolyscaccharide.

• Have been associated with exacerbations of

intestinal inflammation and IBD.

Inappropriate Colonization “DYSBIOSIS”

Type of Diet: Human Milk

versus Formula

Lack of Enteral

Feeding; TPN,

Intestinal pH

Antibiotics and

Microbial Environment

Does Feeding or Lack of Enteral Feeding Affect the Microbiome?

• Excuses To Withhold ENTERAL

“Feedings

Low APGAR scores.

Umbilical catheters.

Apnea and Bradycardia.

Mechanical ventilation.

CPAP.

Vasoactive drugs.

TPN is available.

Demehri, FR., et al. Cellular and Infection Microbiology, Dec. 2013

Effect of Total Parenteral

Nutrition (TPN) in Mice

Demehri, FR., et al. Cellular and Infection Microbiology, Dec. 2013

Clark RH, et al. Pediatrics 2006;117:1979-87

TOP TEN LIST OF NICU

MEDICATIONS

Odds ratio of NEC with

increased days on antibiotics

Alexander, V.N. J. Pediatrics, Sept. 2011

Effect of Antibiotics on Colonization Resistance

Ng, KM., et al. Nature. Vol 502, 3 Oct 2013, and

Stecher, B. Cell Host & Microbe 14, September 11, 2013

Does Intestinal pH Matter for the Microbiome? Gastric Acid Secretion

Hyman, PE, et al J Pediatr. 1985 Mar;106(3):467-71.

Pediatrics, 2012, 129. e-40-45

Gastric Acid Inhibition

Duncan, S.H., et al. Environmental Microbiology, 2009, 11(8), 2112-2122

Dysbiosis and NEC:

Conclusions • May start before birth.

• We have suggested associations between

certain taxa of bacteria and NEC which

make sense.

• Causation (fulfillment of Koch’s postulates)

not complete.

• Many factors still unkown: transcriptomics,

metabolomics, microbial host interactions.

NEC: A Diagnostic Dilemma

Why are new biomarkers needed

for NEC?

• Current diagnosis based on history,

physical exam, radiographic criteria

detects the disease after it’s occurrence

when it may already be too late.

• Diagnostic biomarkers would enhance our

diagnostic capabilities.

• Predictive biomarkers would tell us which

infants are at highest risk so that we could

provide targeted prevention.

Diagnostic Biomarker Types • Non Specific Inflammatory Biomarkers

– Platelet count

– White Blood count and Immature: Total ratios

– C-reactive protein

– Procalcitonin

– Serum amyloid A

– Cytokines

– Neutrophil CD64

Not helpful in differentiating NEC from

sepsis!

Diagnostic Biomarker Types

• Enhanced non-specific Biomarkers

–Fecal Calprotectin (S100A9)

–Fecal Calgranulin C(S100A12)

• Gut Associated Biomarkers

–Intestinal Fatty Acid Binding Protein

(I-FABP)

–Claudin 3

CRP, WBC and PLATELETS Thuijls, et al. Annals of Surgery, 251 (6), June 2010

PLATETLETS

C-REACTIVE PROTEIN

WHITE BLOOD CELL

Claudin

Thuijls, et al. Annals of Surgery, 251 (6), June 2010

Zo-1

Zo-2

Zo-3

Actin filaments

myosin

occludin

Paracellular space

claudin

JAM

Calprotectin Thuijls, et al. Annals of Surgery, 251 (6), June 2010

•Released by intestinal

neutrophils with intestinal

inflammatory damage and can be

used as a marker for gut wall

inflammation

•It is resistant to enzymatic

degradation, and can be easily

measured in feces.

NEC No NEC

INTESTINAL FATTY ACID BINDING PROTEIN (I-

FABP) Thuijls, et al. Annals of Surgery, 251 (6), June 2010

• Small , water –soluble

protein limited to

mature enterocytes of

small and large

intestine.

• Released into

circulation when cell

membrane integrity is

compromised.

• Can be measured in

urine

Marker Cutoff Point Sensitivity Specificity LR+ LR- AUC

(95%CI)

P

I-FABP 2.25 pg/mmole

creatinine

0.93 0.90 9.3 0.08 0.98 (0.94-

1.0)

<0.001

Claudin-3 8OO.8 INT 0.71 0.81 3.74 0.36 0.76 (0.59-

0.94)

0.016

Calprotectin 286.2

microgram/gram

feces

0.86 0.93 12.29 0.15 0.94 (0.85-

1.0)

0.001

NEC versus Non NEC Differentiation Thuijls, et al. Annals of Surgery, 251 (6), June 2010

PREDICTIVE BIOMARKERS:

Who is at greatest risk?

Buccal swab or Salivary analyses:

Predictive Biomarkers

Biomarker Discovery and

Validation (Murgas-Tarrazzo, Neonatology, 2013)

37 identified and 20 mapped

3 chosen for validation

spot # Protein Name

Fold

case/norma

l p-value

mass

kDa/pI

mass

kDa/pI score coverage

unique

peptides

3013

Interleukin-1 receptor antagonist (IL-

1RA) -3.15 0.029 16.23/5.01 20.04/5.82 256 28 5

2971 -2.19 0.01 18.09/5.05 525 24 4

2892 Peroxiredoxin-1 4.18 0.004 23.53/8.59 22.09/8.27 274 39 8

998 Isoform 1 of Alpha-1-antitrypsin 3.18 0.045 72.77/5.53 46.72/5.37 117 6 3

1005 3.48 0.021 72.49/5.59 127 6 4

1203 Clusterin isoform (apolipoprotein J) -2.16 0.0087 65.50/7.62 52.48/5.88 140 8.91 3

1146 -1.95 0.019 67.23/7.24 159 8.9 3

2777 Proteosome subunit alpha type 2 1.18 0.047 26.04/5.05 25.90/6.91 85 13.2 2

854 Gelsolin (isoform 2-cytoplasmic) 1.78 0.019 81.26/5.61 80.62/5.58 698 22 11

855 1.76 0.037 80.66/5.68 810 23 16

1897

Cleaved Peroxisomal multifunctional

enzyme type 2 -3.61 0.0016 38.60/8.78 33.59/8.6 298 9.92 5

(3R)-hydroxyacyl-Co A dehydrogenase)

2892

Phosphatidylethanolamine-binding protein

1 4.18 0.004 23.53/8.59 21.05/7.01 243 26 4

1975 Alpha-2-glycoprotein 1, zinc precursor -2.24 0.0062 38.70/5.41 34.26/5.71 112 26 9

805 Polymeric immunoglobulin receptor 2.08 0.0026 86.63/5.24 83.260/5.59 238 7.3 6

2442 cDNA FLJ75207 -2.22 0.00047 31.15/7.35 29.12/8.27 297 12 3

2436 -2.45 0.0025 31.12/7.0 326 12 4

3013 Prolactin-inducible protein -3.15 0.029 16.23/5.01 16.56/8.26 278 26 3

3027 -2.76 0.0053 8.85/4.63 172 4 2

1146

Protein-glutamine gamma-

glutamyltransferase E -1.95 0.019 67.23/7.24 76.62/5.61 727 23 10

2892

Neutrophil gelatinase-associated

lipocalin (lipocalin-2) 4.18 0.004 23.53/8.59 22.59/9.02 503 26 8

2181 N-acetylglucosamine kinase -1.47 0.026 34.70/5.70 37.36/5.82 1014 30 14

1975

cDNA FLJ53019, highly similar to Serpin

B13 -2.24 0.0062 38.70/5.41 45.28/5.56 830 26 9

1146

cDNA FLJ54957, highly similar to

transketolase -1.95 0.019 67.23/7.24 68.74/7.58 258 12 5

1203 -2.16 0.0087 65.50/7.62 222 8.7 3

1694 Placental protein 11 -1.49 0.017 49.35/5.23 42.10/5.26 368 14 4

1173

Isoform 1 of heat shock cognate 71 kDa

protein -2.37 0.018 65.92/6.3 70.88/5.37 202 8.2 4

2425 s-formylglutathione hydrolase (esterase) -2.49 0.00027 31.88/6.6 31.45/6.54 86 7.45 2

2932 Small proline-rich protein 3 2.37 0.014 20.47/8.34 18.15/8.86 89 28 4

Discovery Phase

Validation Phase

• Western Blots showed IL-1 RA lower

in babies who subsequently

developed NEC.

• ELISA-no difference.

• Still searching: Other promising

biomarkers being evaluated.

• Lipocalin-2

Prevention with Mother’s Milk: NEC in Premature Infants (UK)

Lucas & Cole, Lancet 1990;336:1519-23

Formulas Only (n=236) 24 (10.2%) 17 (7.2%)

Formulas plus

Mother’s Milk (n=437) 16 (3.7%) 11 (2.5%)

Human Milk (n= 253) 11 (4.3%) 3 (1.2%)

In-Hospital Diet All Cases Confirmed Cases

Preventative Measures

• Questionable measures

– Formula acidification.

– IgG-IgA.

– Oral antibiotics.

– Glucocorticoids.

– Anticytokine therapy

– Growth Factors (EGF)

– Arginine/glutamine

– Artificial “amniotic fluid”

– Pre, pro and postbiotics

Routine Use of Probiotics

Probiotic-Definition

• The definition of probiotics as

defined by the World Health

Organization is "live

microorganisms which when

administered in adequate

amounts confer a health

benefit on the host.“

• There are well over 100 types

of probiotics, with over 50

Lactobacillus species alone.

Food Supplement or Drug?

• It depends!

• Health claim usually should be considered

a drug.

• Drugs that are sold by prescription are

subjected to rigorous testing.

• Foods can be sold by anyone and not

subjected to rigorous standards.

Probiotics significantly reduce mortality in VLBW infants

by 41% (from 8.3% down to 3.4%) " NNT 21

Is the result valid and reliable? Can it be generalized?

Meta-Analysis-Death

Deshpande, G. Pediatrics, 2010

Summary of 2010 Meta Analysis

• 11 studies evaluated.

• 10 different probiotic preparations.

• Risk for NEC and death significantly lower in probiotic group.

• Sepsis did not differ.

• “Overall evidence indicate that additional placebo controlled trials are unnecessary if a suitable probiotic product is available”.

Deshpande,G. Pediatrics, May 2010

Examples of Misadventures and

Over-Exuberance • High Oxygen/minimal oxygen=retinopathy and

increased mortality.

• 100% oxygen for resuscitation.

• High dose vitamin K and kernicterus

• Sulfizoxazole prophylaxis and kernicturus

• Chloramphenicol and cardiovascular collapse

• Steroids and cerebral palsy

• Furosemide and pathologic fractures.

Some regulatory concerns

• Is there a ‘‘suitable probiotic product’’

with adequate quality control that is

approved by the Food and Drug

Administration (FDA) (as safe and

effective for NEC) currently on the

market?

• To this author’s knowledge there is not

such a product.

Neu, J. Journal of Pediatrics, April 2011

Questions Which probiotics actually decrease NEC?

Is there really a decrease in mortality and is this

causally related to probiotics?

Are there populations that might benefit the

most?

Are there risks (e.g. increased sepsis, long term

risks, etc.)?

Multicenter Trial of Probiotics in

Taiwan

Lin, et al PEDIATRICS Vol. 122 No. 4 October 2008, pp. 693-700

Multicenter Trial of Probiotics in

Taiwan (750 g)

Lin, et al PEDIATRICS Vol. 122 No. 4 October 2008, pp. 693-700

Fatal gastrointestinal mucormycosis in an

infant following use of contaminated ABC

Dophilus from Solgar Company

CDC, FDA, and the Connecticut Departments

of Public Health and Consumer Protection, are

investigating a fatal case of GI mucormycosis

in a premature infant of 29 weeks gestation

following the use of a probiotic supplement

called ABC Dophilus distributed by Solgar,

Inc., Leonia, NJ.

FDA ALERT!!

Probiotic NEC study - sample size

calculation

NEC incidence Infants per group

One-sided Two -sided

5% baseline 792 984

7% baseline 557 693

Parallel group design

One-sided (two-sided) testing

α=5%, ß=80%

50% reduction in NEC incidence

Largest Studies so far

• ProPrems---Pediatrics, 2013-powered for late onset

sepsis.

• 1099 very preterm infants from Australia and New

Zealand were randomized.

• No significant difference in definite late-onset sepsis or

all-cause mortality was found.

• probiotic combination reduced NEC of Bell stage 2 or

more (2.0% versus 4.4%; relative risk 0.46, 95%

confidence interval 0.23 to 0.93, P = .03; number needed

to treat 43, 95% confidence interval 23 to 333).

• COSTELLO STUDY IN GREAT BRITAIN:NO

DIFFERENCES

• Although the prevalence is not known, many

neonatologists have started prescribing probiotics

in NICUs for the prevention of NEC.

• Anecdotally, some of the agents being used are

not even probiotics previously studied in preterm

neonates.

• In the United States, there are no probiotics that

are “licensed” or for which there are well

developed access schemes for routine use in the

prevention of NEC as suggested by one author for

their use.

Current Status in U.S.

• There are also no current standards for

“quality control of a reconstituted

product”.

• Good manufacturing practices (GMPs)

specifically for use of probiotics as drugs

to prevent a specific disease such as

NEC are not available.

Current Status in U.S.

Probiotics: Conclusions and

Recommendations

• So far no product has been licensed.

• Confirmation of the available results in large,

adequately powered, possibly multi-center, well

designed RCTs with proven NEC or mortality as

primary outcome is required. • Abrahamsson TR, Rautava S, Moore AM, Neu J, Sherman PM.J

Pediatr. 2014 Aug;165(2):389-94

Take Home Messages

• “NEC” is more than one disease.

• Pathogenesis of “classic NEC” is multifactorial, but some factors (e.g., GI microbial ecology) may be more important than others.

• Predictive and diagnostic biomarkers are needed for targeted prevention and therapy.

• Promising preventative measures may be on the horizon but we need to proceed carefully.