3939s1 Klontz

Enterobacter sakazakii Case Reports and Outbreaks Involving Infants as Reported in the Peer-Reviewed English Medical LiteratureKarl C. KlontzEpidemiology TeamCenter for Food Safety and Applied NutritionFood and Drug Administration: 

Enterobacter sakazakii Case Reports and Outbreaks Involving Infants as Reported in the Peer-Reviewed English Medical Literature Karl C. Klontz Epidemiology Team Center for Food Safety and Applied Nutrition Food and Drug Administration

Case Definition- E. sakazakii infection: 

Case Definition- E. sakazakii infection An infant meeting at least one of the following criteria: • E. sakazakii recovered from > 1 of the following normally sterile sites: ▫ Blood ▫ Cerebrospinal fluid ▫ Brain tissue ▫ Urine • Outbreak of necrotizing enterocolitis and E. sakazakii recovered from blood, stool, or stomach aspirate of > 1 infant • Bloody diarrhea, E. sakazakii recovered from stool in pure culture

Number Reports E. sakazakii Infection – English-Language Peer-Reviewed Literature: 

Number Reports E. sakazakii Infection – English-Language Peer-Reviewed Literature Age group No. references No. cases (%) Infants 20 48 (83) Children (>1 yr), adults 7 10 (17) Age group No. cases Percent 0 – 1 month 37 64 >1 – 12 months 10 17 >1 – 4 years 4 7 > 4 years 6 10 (median age = 74 yr) > Not stated 1 2 (infant, but ? age) > Total 58 100

Initial Reports of E. sakazakii infection: Chronology: 

Initial Reports of E. sakazakii infection: Chronology Urmenyi (1961) Jöker (1965) Monroe (1979) Kleiman (1981) Adamson (1981) Muytjens (1982) Jiménez (1982)* Muytjens (1983) Case series (1st study to propose possible link to powdered formula) Case reports

E. sakazakii Infections in Infants: Selected Clinical Features – Peer-Reviewed Literature (N=48): 

E. sakazakii Infections in Infants: Selected Clinical Features – Peer-Reviewed Literature (N=48) Gender: 52% female Syndromes: Meningitis (58%), Sepsis/bacteremia (17%); Necrotizing enterocolitis (29%) Overall case-fatality: 33% (15/46 for whom info available) Birth weight (g) Number No. Died (%) < 1000 10 2 (20) 1000-1499 9 2 (22) 1500-2499 11 7 (64) > 2500 10 3 (30)

Muytjens et al. (1983): Analysis of 8 cases of neonatal meningitis and sepsis due to E. sakazakii: 

Muytjens et al. (1983): Analysis of 8 cases of neonatal meningitis and sepsis due to E. sakazakii Reanalyzed 20 Enterobacter strains isolated from CSF (going back 6 yrs) and 25 strains from blood (2 yrs) in the Netherlands Hypothesis: E. sakazakii may not have been identified originally because organism produces minimal pigment at 36° C Using various biochemical and growth parameter systems (25 ° C), 8 strains identified as E. sakazakii All 8 strains confirmed by CDC as E. sakazakii Clinical data for E. sakazakii cases obtained retrospectively

Muytjens et al. (1983): Clinical features of 8 cases of neonatal meningitis and sepsis due to E. sakazakii: 

Muytjens et al. (1983): Clinical features of 8 cases of neonatal meningitis and sepsis due to E. sakazakii 3/8 cases delivered by C-section Progress in first few days of life was normal First signs of illness occurred between days 4-8 of life 2/8 cases had necrotizing enterocolitis + meningitis 75% case-fatality rate; 2 survivors “retarded” Death occurred within hours – several days “Hemorrhagic encephalitis”

Muytjens et al. (1983): Epidemiologic features of 8 cases of neonatal meningitis and sepsis due to E. sakazakii: 

Muytjens et al. (1983): Epidemiologic features of 8 cases of neonatal meningitis and sepsis due to E. sakazakii 6/8 cases birth weights < 2,500 grams (lowest: 850 g) 5/8 cases premature (gestation < 36 wk) 5/8 cases occurred at same general hospital Other 3 cases born in 3 other hospitals Temporal clustering: 3/5 cases in general hospital became ill within 3 months; 2/3 cases in other hospitals within 2 months

Muytjens et al. (1983): Environmental Sampling Results: 

Muytjens et al. (1983): Environmental Sampling Results Environmental sampling at pediatric ward of general hospital E. sakazakii recovered from prepared formula, dish brush, and stirring spoon, but not from powdered formula and water Plasmid profiles: “3 or 4 of 5 isolates from patients at the general hospital were probably the same strain” Plasmid profiles differed between cases vs. prepared formula (1990 note: “No cases observed at general hospital since powder formula replaced by liquid formula 8 years ago.”)

E. sakazakii and Powdered Infant Formula: Developments From 1961-1988: 

E. sakazakii and Powdered Infant Formula: Developments From 1961-1988 1980 (Farmer): 1/57 isolates came from “can of dried milk” 1983 (Muytjens) ▫ E. sakazakii recovered from prepared formula ▫ Recommended sampling of “milk” as part of investigations 1988 (Muytjens) ▫ 141 powdered infant formula products obtained from 35 countries tested for presence of Enterobacteriaceae ▫ 52% samples + ▫ 20 powders (14%) from 13 countries yielded E. sakazakii ▫ Concentration of Enterobacteriaceae uniformly < 1 CFU/g

Three Cases of Neonatal Meningitis Caused by E. sakazakii in Powdered Formula (Biering [1989-Iceland]): 

Three Cases of Neonatal Meningitis Caused by E. sakazakii in Powdered Formula (Biering [1989-Iceland]) Illnesses in 1986, 1987 (2 within one month) 2 infants normal at birth (36- and 38-wk gestation); 1 Down’s All 3 did well until day 5 when they became ill 1 died, 2 recovered with severe neurologic sequelae All 3 fed powdered infant formula 1st study to show, in illness setting, E. sakazakii recovery (low numbers) from freshly prepared formula from previously unopened cans No recovery from environment (milk kitchen, utensils, ward)

Three Cases of Neonatal Meningitis Caused by E. sakazakii in Powdered Formula (Biering [1989-Iceland]): 

Three Cases of Neonatal Meningitis Caused by E. sakazakii in Powdered Formula (Biering [1989-Iceland]) In addition to 3 ill cases, 1 colonization 4 E. sakazakii strains isolated from neonates had same plasmid profile as 22 formula strains Anecdotal evidence that formula bottles occasionally kept at 35-37 C for extended periods in bottle heaters But in 1 instance, recovery (direct plating) from bottle that had been refrigerated “…milk powder can be the mode of transmission for E. sakazakii meningitis or sepsis in neonates…”

Outbreaks of E. sakazakii infections linked to powdered infant formula: Historical cohort studies: 

Outbreaks of E. sakazakii infections linked to powdered infant formula: Historical cohort studies Simmons et al. (1989): Enterobacter sakazakii Infections in Neonates Associated with Intrinsic Contamination of a Powdered Infant Formula (Infect Control Hosp Epidemiol) Van Acker et al. (2001): Outbreak of Necrotizing Enterocolitis Associated with Enterobacter sakazakii in Powdered Milk Formula (J Clin Microbiol) CDC. Enterobacter sakazakii Infections Associated with the Use of Powdered Infant Formula-Tennessee, 2001. MMWR 2002;51:297-300

E. sakazakii Infections in Neonates Associated with Intrinsic Contamination of Powdered Infant Formula (Simmons-1989): 

E. sakazakii Infections in Neonates Associated with Intrinsic Contamination of Powdered Infant Formula (Simmons-1989) Feb-Mar 1988: bacteremia (2 infants), UTI (1), bloody diarrhea (1) E. sakazakii recovered from stool of all 4 infants Investigation revealed all 4 had been fed same powdered formula Formula prepared in blender; rinsed with tap water between uses Blender yielded heavy growth of E. sakazakii No further clinical isolates once blender sterilized Historical cohort study in NICU assessed risk factors for infection and colonization (2/7-3/13)

E. sakazakii Infections in Neonates Associated with Intrinsic Contamination of Powdered Infant Formula (Simmons-1989): 

E. sakazakii Infections in Neonates Associated with Intrinsic Contamination of Powdered Infant Formula (Simmons-1989) Infected/Colonized Yes No Yes 4 1 5 Formula No 0 40 40 Relative risk = undefined; p = 0.0006

E. sakazakii Infections in Neonates Associated with Intrinsic Contamination of Powdered Infant Formula (Simmons-1989): 

E. sakazakii Infections in Neonates Associated with Intrinsic Contamination of Powdered Infant Formula (Simmons-1989) Implicated elemental formula given 1° to most premature infants Samples obtained from opened, yet unmixed, can powdered formula Powder cultured using method of Muytjens et al. E. sakazakii (~ 8 CFU/100 g), E. cloacae (48 CFU/100 g) recovered Same plasmid, multilocus enzyme profile (formula and patients)

van Acker et al. (2001): Necrotizing enterocolitis associated with E. sakazakii in powdered formula: 

van Acker et al. (2001): Necrotizing enterocolitis associated with E. sakazakii in powdered formula 12 cases NEC diagnosed June-July, 1998 in Belgium NICU 50 neonates admitted to NICU during this period All 12 had birth weights <2,000 g, fed powdered infant formula 6/12 neonates had + cultures [blood, anal swab, or stomach aspirate] for E. sakazakii vs. 0/38 without NEC (p<0.001) 10/12 neonates with NEC had been fed the same powdered formula: Alfaré (semi-elemental formula with low osmolarity)

van Acker et al. (2001): Necrotizing enterocolitis associated with E. sakazakii in powdered formula: 

van Acker et al. (2001): Necrotizing enterocolitis associated with E. sakazakii in powdered formula E. sakazakii + (NEC) Yes No Yes 6 8 14 Formula No 0 36 36 Relative risk = undefined; p = 0.0002

van Acker et al. (2001): Necrotizing enterocolitis associated with E. sakazakii in powdered formula: 

van Acker et al. (2001): Necrotizing enterocolitis associated with E. sakazakii in powdered formula E. sakazakii recovered from prepared formula, unopened cans Molecular typing (AP-PCR) confirmed partial strain similarity between powdered formula, patient isolates No further cases of NEC observed after use of implicated powdered formula stopped Inadvertent challenge test: Implicated formula stopped on 7/10/98 Released again on 7/20 7/23: new case NEC linked to formula (intrinsic formula contamination shown at same time)

van Acker (2001): NEC-associated with E. sakazakii in powdered formula: 

van Acker (2001): NEC-associated with E. sakazakii in powdered formula Manufacturer’s quality control data for formula: Of 5 samples analyzed: 1 yielded 20 coliforms/g; 4 < 1 coliform/g Results fulfilled requirements of Codex Alimentarius: minimum of 4/5 control samples with < 3 coliforms g; maximum of 1/5 with >3 but < 20 coliforms/g Did not meet Belgian law: < 1 coliform/g in all samples Product recalled

Manufacturer Actions: van Acker (2001): NEC Outbreak: 

Manufacturer Actions: van Acker (2001): NEC Outbreak Production facility in The Netherlands upgraded More stringent standards adopted for products (< 0.3 coliform/g, 0 E. sakazakii isolates/10 g) 1999: implicated powdered formula reintroduced As of publication date, no further cases of NEC associated with E. sakazakii