Objective. Streptococcus pneumoniae infections in the neonate (SPIN) are relatively unusual events (1%–11% of neonatal sepsis) but are associated with substantial morbidity and mortality. Previous reports suggest that invasive SPIN is associated with prolonged rupture of membranes, maternal colonization/illness, prematurity, early-onset pneumonia presentation (<72 hours), and high mortality (50%). The aim of this study was to review the current epidemiology and clinical course of SPIN.
Methods. The US Pediatric Multicenter Pneumococcal Surveillance Group has been prospectively monitoring S pneumoniae infections since 1993 in 8 children’s hospitals. For this report, data were gathered retrospectively from the charts of neonates who were 30 days of age and younger and had SPIN from September 1993 to February 2001. All pneumococcal isolates were sent to a central laboratory for serogrouping/typing and susceptibility testing.
Results. Twenty-nine cases of SPIN were identified from a total of 4428 episodes of S pneumoniae infection in children. Sixty-six percent were male, and 55% were white; the mean age was 18.1 day (±8.2). Ninety percent of infants were ≥38 weeks’ gestation. Two mothers had bacterial infections at delivery; 1 had S pneumoniae isolated from both blood and cervix, and 1 had clinical amnionitis. The primary diagnoses in the neonates were bacteremia (8), meningitis (8), bacteremic pneumonia (4), septic arthritis/osteomyelitis (1), and otitis media (8). Thirty percent of infants with invasive SPIN presented with leukopenia/neutropenia, but this did not predict poor outcome. The infecting pneumococcal serogroups were 19 (32%); 9 (18%); 3 and 18 (11% each); 1, 6, and 14 (7% each); and 5 and 12 (3.5% each). Twenty-six percent of invasive neonatal infections were caused by serogroups 1, 3, 5, and 12, which are not contained in the heptavalent pneumococcal vaccine. In contrast, 6% of invasive nonneonatal disease was caused by these same nonvaccine serogroups. Susceptibility testing demonstrated that 21.4% of isolates were penicillin nonsusceptible and 3.6% were ceftriaxone nonsusceptible. Three (14.3%) neonates with invasive SPIN died; all deaths occurred within 36 hours of presentation. Deaths did not appear to be related to pneumococcal serogroup or susceptibilities.
Conclusions. Compared with previous studies of neonates with pneumococcal infection, this series showed that infants with SPIN were usually 2 to 3 weeks of age at presentation; likely to be full term; and ill with pneumonia, meningitis, and otitis media. This late-onset presentation was associated with an overall mortality rate of 10.3% (14.3% for invasive disease).
The recent development, licensure, and routine use of the heptavalent pneumococcal conjugate vaccine (PCV7) in infants is having a substantial impact on the epidemiology of pneumococcal disease in children.1–4 In infants younger than 1 year, rates of invasive pneumococcal disease have decreased from 50 to 100/100 000 to 9/100 000 person-years, and in children younger than 2 years from 82 to 114/100 000 to 38/100 000 person-years.4 A challenge that remains is to develop a preventive approach for infants who are younger than 2 months and are below the age of administration of PCV7. In addition, the increasing resistance of pneumococci to antimicrobials, including penicillin and ceftriaxone reported by our group5 and others,6 may complicate decisions regarding therapy for suspected pneumococcal disease.7
Previous reports suggest that Streptococcus pneumoniae infections in the neonate (SPIN), including sepsis, pneumonia, and meningitis, are relatively unusual events (1%–11% of neonatal sepsis cases)8–12 but are associated with considerable morbidity and mortality. Published reviews of SPIN consist of single cases and small series,13–26 most of which describe early-onset disease akin to neonatal group B Streptococcus (GBS) disease. These infections were associated with respiratory distress within the first day of life, prolonged rupture of membranes (PROM), prematurity, leukopenia, and high mortality.23–25 In addition, several series described positive maternal vaginal cultures for S pneumoniae23,25 and even concomitant maternal pneumococcal infections, including pneumonia, meningitis, and amnionitis.25 In contrast, Gomez et al26 in 1999 reported that gestational age (GA), low birth weight (LBW), and PROM were not risk factors in 5 neonates with invasive SPIN from their institutions and their analysis of 119 cases from the literature.
The purpose of this study was to characterize further the current epidemiology, clinical presentations, and outcome of SPIN. This report details the experience of 29 neonates, ≤30 days of age, with S pneumoniae infection identified between 1993 and 2001 by the US Pediatric Multicenter Pneumococcal Surveillance Group.
The US Pediatric Multicenter Pneumococcal Surveillance Group is composed of investigators at 8 children’s hospitals in the United States who have been prospectively collecting data and bacterial isolates from pediatric patients with pneumococcal infections at their institution.5,27 Isolates of S pneumoniae are from normally sterile body sites (blood, pleural fluid, cerebrospinal fluid, synovial fluid/bone, etc) collected since September 1993 and from middle ear fluid since September 1994 (spontaneous drainage, tympanocentesis, myringotomy or at time of placement of pressure equalization tympanostomy tubes). This study was approved by the Institutional Review Board for each institution.
The current study used original case report form data designed for the identification of all patients with pneumococcal disease. Additional information was abstracted retrospectively from the medical record to obtain additional demographic, obstetric, epidemiologic, and outcome data for SPIN patients identified from September 1993 through February 2001. “Invasive SPIN” is defined as bacteremia, pneumonia, meningitis, and bone and joint infections. “All infants with SPIN” also includes neonates with middle ear fluid from which S pneumoniae was isolated.
All pneumococcal isolates were sent to the Infectious Diseases Research Laboratory, Texas Childrens Hospital (Houston, TX), where serogrouping/serotyping and susceptibility testing for penicillin and ceftriaxone were performed. Strains were serogrouped by commercially available antisera (Statens Seruminstitut, Copenhagen, Denmark; Dako Inc, Carpenteria, CA).28 Serotyping was performed on vaccine serogroups only. Susceptibility testing was performed by standard microbroth dilution with Mueller-Hinton media supplemented with 3% lysed horse blood. Susceptibility breakpoints for all isolates described in this report, from 1993 to 2001, were defined according to the 2002 National Committee for Clinical Laboratory Standards Guidelines29,30 (breakpoints: penicillin ≤0.06 μg/mL = susceptible [S], 0.1–1.0 μg/mL = intermediate [I], ≥2.0 μg/mL = resistant [R]; ceftriaxone (meningeal isolates) ≤0.5 μg/mL = S, 1.0 μg/mL = I, ≥2 μg/mL = R; ceftriaxone (nonmeningeal isolates) ≤1 μg/mL = S, 2 μg/mL = I, ≥4 μg/mL = R). Isolates that were I and R to penicillin or ceftriaxone were considered nonsusceptible to that drug.
Statistical analyses were performed using NCSS 6.2 (Number Crunching Statistical Systems). Descriptive statistics were calculated on all outcome variables in addition to the relevant classification variables and covariate age. Normality and equal variance were tested on data when t test, analysis of variance, or regression analysis was applied. In cases in which data were nonnormal or variance was unequal, transformations were attempted and nonparametric analyses were conducted. Mann-Whitney U test was used for 2-group comparisons of medians along with the Kolmogorov-Smirnov test for difference in distributions. Fisher exact test was applied to test equality of 2 proportions. χ2 tests were performed on 2-way tables. Pearson correlation coefficients and Spearman rank correlation coefficients were applied to normal and nonnormal data, respectively. All hypotheses were tested at α = 0.05 level of significance.
During the study period September 1993 through February 2001, 29 infants ≤30 days were identified with SPIN (21 invasive infections and 8 otitis media [OM]), from a total of 4428 episodes of S pneumoniae infections (0.65%; 3200 invasive infections, 1228 OM). Cases per year ranged from 1 to 6, with a mean (±standard deviation [SD]) of 3.43 (2.1). Sixty-six percent of the infants were male; 55% were white, 24% were black, 17% were Latino, and 3% were other. The mean age (±SD) of all infected infants was 18.1 days (8.2) and 17.8 days (9.2) for infants with invasive disease (not significant). Ninety percent of the infants were ≥38 weeks’ GA. The mean (±SD) birth weight was 3318 g (402) for all infants (N = 26) with SPIN, 3206 g (411) for infants with invasive disease (N = 18), and 3571 g (251) for infants with OM (N = 8). Birth weights were not available on 3 infants with invasive disease. The lower birth weight of infants with invasive SPIN was statistically significant when compared with those with isolated OM (P = .030).
Sites of Infection and Clinical Presentations
The primary diagnoses and age of presentation are listed in Tables 1 and 2; meningitis, bacteremia without localization, and OM were the most common diagnoses. One infant had pyoarthrosis of the elbow with contiguous osteomyelitis. The mean age (±SD) at presentation for each primary diagnosis is shown in Table 1. One-way analysis of variance of age at presentation by primary diagnosis did not reveal any significant differences.
Table 2 lists the associated diagnoses (see Other in Table 2). Of 8 infants with meningitis, 3 were bacteremic (patients 6, 10, and 11) and 6 had another focus of infection (2 with OM [patients 11 and 12] and 4 with pneumonia [patients 5, 6, 8, and 9]). Three infants with primary bacteremia had coexistent respiratory syncytial virus infections present on admission (patients 13, 14, and 15).
Three infants presented with early-onset systemic disease (within the first 72 hours of life); only 1 presented in the first 24 hours of life (Table 2). Two of these infants were full term and were diagnosed with bacteremic pneumonia on day of life (DOL) 1 and 2, respectively (patients 1 and 2). The third (patient 5) presented on DOL 3 with meningitis; this infant was born at 37 weeks’ GA. The mothers of infants 2 and 5 had indications of intrauterine and/or systemic infections (see below).
Three infants had an underlying disease: a 16-day-old infant with a cleft lip/palate presented with bilateral OM (patient 24), a 26-day-old who had human immunodeficiency virus infection presented with bacteremia (patient 18), and a 28-day-old who was status post inguinal hernia-related small bowel obstruction presented with bacteremia (patient 19). Three of 29 infants were premature; 1 was born at 34 weeks’ GA (patient 9), and 2 were born at 37 weeks’ GA (patients 5 and 16; Table 2).
Fever was observed in 55.2% (16 of 29) of all infants; 71.4% of infants with invasive disease presented with documented rectal temperatures >38.0°C/100.4°F31 (Table 1). Fevers ranged from 38.2°C to 40.3°C (100.6°C–104.6°F). One of 8 patients with isolated middle ear infection was febrile, whereas fever was present in 7 of 8 patients with meningitis. All 8 patients with primary bacteremia had fever at presentation, but, notably, none of 4 patients with bacteremia and pneumonia presented with fever.
The majority of infants were noted to be irritable, often with signs of respiratory distress (grunting, flaring, retractions) and ill in appearance, but did not have any distinguishing signs or symptoms that might differentiate the illness from serious infections as a result of other bacteria in this age group. Most infants (70%) with OM had bilateral infections (Table 2).
White blood cell counts among infected infants varied from profound leukocytosis to leukopenia. Seven (33%) of 21 infants with invasive disease had WBC counts <5.0 × 103/μL and/or absolute neutrophil count <1.0 × 103/μL (Table 1). Two of these infants had bacteremic pneumonia, and 5 had meningitis and sepsis. All 4 infants ≤7 days of age had low WBC counts and/or absolute neutrophil count (patients 1, 2, 5, and 6). There was a significant difference in the mean ages (±SD) of infants with invasive disease who presented with WBC counts <5.0 × 103/μL (mean age: 11.0 days [12.2; N = 6]) when compared with those with invasive disease with WBC counts >5.0 × 103 /μL (mean age: 20.5 days [6.41; N = 15]; P = .030).
Figure 1 compares primary diagnoses in neonates, with infection in nonneonates during the study period. Meningitis alone was significantly more common in the neonates than in the nonneonates (27.6% vs 10.0%; P = .006).
Maternal History and Other Risk Factors
Obstetric history was available from 17 (58.6%) of 29 of all cases and in 13 (61.9%) of 21 invasive cases. Two mothers had presumed bacterial infections at the time of delivery. One mother had clinical amnionitis, with PROM >18 hours. Neither she nor her infant received peripartum antibiotics. The infant presented on DOL 2 with respiratory distress and was diagnosed with pneumonia and bacteremia (patient 2). The other mother had S pneumoniae isolated from blood and cervix (serotype unknown) and rupture of membranes for 14 hours. Although the mother received intrapartum antibiotics, this infant presented with meningitis on DOL 3 (patient 5). Two additional women were reported to have received intrapartum chemoprophylaxis for GBS colonization. These infants presented on DOL 14 with penicillin-susceptible bacteremia (patient 14) and on DOL 18 with penicillin-resistant S pneumoniae OM and purulent otorrhea (patient 26), respectively. Neither ill contacts in the family (upper respiratory infection) nor having multiple siblings or siblings in child care was a risk factor for SPIN (data not shown), although the latter may have influenced the S pneumoniae serotype and susceptibilities causing the infection (see below).
Serotypes/Serogroups and Antibiotic Susceptibilities
The results of serogroups/types are listed in Fig 2. Serogroup 19 was the most common cause of infections in neonates for both invasive (30%) and total infections, including OM (32%). This was the third most common serogroup causing disease in the nonneonatal children’s group from 1993 to 2001.
Twenty-six percent of invasive neonatal isolates were caused by serogroups 1, 3, 5, and 12, serogroups not contained in the PCV7 pneumococcal vaccine. In contrast, only 6.0% of invasive nonneonatal disease was caused by these same nonvaccine serogroups (179 of 2996; P = .005). Neonates who had at least 2 siblings and/or siblings in child care were more likely to have infections caused by vaccine serogroups than neonates without siblings or siblings in child care (82% [9 of 11] vs 50% [3 of 6]).
Figure 3 depicts the antimicrobial susceptibilities of S pneumoniae isolates recovered from neonates compared with nonneonatal isolates from infected children during this study period. Overall, the neonatal isolates were more susceptible to penicillin, for all infections, OM and invasive infections. These differences in susceptibility did not reach statistical significance, likely because of small neonatal sample size. Only 1 neonatal isolate was ceftriaxone nonsusceptible. All neonatal nonsusceptible isolates occurred in vaccine strain serogroups 6, 9, 14, and 19. Five of 6 penicillin-nonsusceptible neonatal isolates occurred since 1998; the single ceftriaxone-nonsusceptible isolate occurred in 1998. Of note, 3 of 11 neonates with multiple siblings and/or siblings in child care were infected with penicillin-nonsusceptible organisms, whereas none of the 6 infants without siblings or without siblings in child care had nonsusceptible isolates.
Clinical Course and Outcome
Seven patients required intubation. These included all 4 patients with bacteremic pneumonia and 3 patients with meningitis, 2 of whom had concomitant pneumonia. Several patients required hemodynamic support, including intravenous bolus fluids, albumin infusions, and inotrope support (dopamine, dobutamine, epinephrine).
The overall mortality in the neonates with invasive disease was 14.3% (3 of 21) compared with 1.3% in the nonneonatal group (P = .003), the majority of which occurred in children with underlying conditions.5 Fatal neonatal infections occurred in 2 patients with septic shock. The third fatal infection was caused by septic shock/meningitis in a premature infant who was of 34 weeks’ GA and presented at DOL 26 (patient 9). All deaths occurred within 36 hours of presentation. Mortality did not appear to be related to isolate serogroup or susceptibility. Fatal infections occurred with serogroups 3, 6, and 9; 2 organisms were fully susceptible, and 1 was penicillin nonsusceptible (MIC 1) but ceftriaxone susceptible. The WBC count at presentation did not predict outcome. Six of the 7 infants with leukopenia and/or neutropenia survived. The other 2 of 3 infants who died had the highest peripheral WBC counts (43.0 × 103/μL and 27.0 × 103/μL) at presentation (patients 3 and 4). At the time of discharge, 2 of 7 surviving infants with meningitis had cerebral infarcts (patients 6 and 7), 1 of whom also had an abnormal brainstem auditory response at hospital discharge (patient 7). In addition, the infant with pyoarthrosis/osteomyelitis had mild decreased range of motion in the affected arm at discharge (patient 21).
Although not the subject of this report, an additional 44 infants in the database were between 31 and 60 days of age. Therefore, 73 infants identified during the study period were below the age of receipt of PCV7. These infants presented with bacteremia (41%), meningitis (25%), pneumonia (6.8%), mastoiditis (6.8%), bone and joint infection (2.3%), cellulitis (2.3%), and OM (15.9%). Twenty-two percent of isolates were nonsusceptible to penicillin, and all but 1 isolate was ceftriaxone susceptible. Serogroups 19 (24.4%), 14 (13.0%), 3 (11.1%), and 6 (8.9%) were the 4 most common serogroups in this age (N = 40). Eighteen percent (6 of 33) of invasive infections in this age group were caused by nonvaccine serogroups 1, 3, 5, and 12, compared with 26% of neonatal infections and 6% of the nonneonatal infections.
SPIN has previously been reported as single case reports and small series. With the exception of 1 report,26 most of these cases describe neonates presenting in the first several days of life with invasive disease, especially pneumonia and sepsis, and often accompanied by leukopenia. These cases were associated with LBW, premature delivery, and obstetric complications. Mortality rates varied from 20% to 60%.23–26
In contrast to most previously published reports, only 3 neonates in this series presented in the first 3 days of life, with risk factors for early-onset neonatal sepsis present in 2 of them. One infant had no known risk factors, and none of these 3 neonates with early-onset presentation died. The mean age (±SD) of presentation for all cases of SPIN and for invasive disease was in the third week of life, 18.1 days (8.2) and 17.8 days (9.2), respectively, and only 1 of the infants in this series was of LBW. In this study population, 8 (38%) of 21 neonates who presented with invasive disease had meningitis and 38% had primary bacteremia. Pneumonia remained a common presentation; 4 infants presented with bacteremic pneumonia, and 4 neonates with meningitis had concomitant pneumonia. Leukopenia and/or neutropenia was present in 30% of neonates with invasive SPIN; low WBC count did not predict poor outcome.
The presentation of isolated OM occurred in 27.6% of infants in this series. Previous data from the 1970s support the finding of S pneumoniae as a common causative agent for OM in neonates.32,33 As in the present report, other studies observed that young infants with isolated pneumococcal OM were full term and often had bilateral disease. In the study of Turner et al,34 70% of the infants were febrile, and none of them had concomitant serious bacterial infection caused by S pneumoniae. In this present report, only 1 of 8 neonates with isolated OM had fever. However, 2 additional cases of bilateral OM (not culture-proven S pneumoniae) were associated with concomitant S pneumoniae meningitis in febrile neonates.
The serogroup distribution of isolates of S pneumoniae recovered from neonates and nonneonates was compared (Fig 2). In our study, 75% of SPIN and 86% of nonneonatal infections were caused by serogroups in the currently licensed PCV-7 (data not shown). When only invasive infections were examined, in neonates, 26.3% (5 of 19) were caused by the nonvaccine serogroups 1, 3, 5, and 12, whereas these same serogroups represented only 6% of invasive disease in all children (P = .005). Several previous studies from the United States and Europe have also demonstrated that a substantial proportion of neonatal disease may be caused by nonvaccine serogroups.24–26 These serogroups are rarely found to cause invasive disease in young children in the United States and Europe, although they represent more common serogroups causing disease in older children and adults.5,35,36 It is interesting that they are important causes of disease in young children in Asia, Africa, and Latin America.36 These serogroups may therefore be acquired in the United States from parents and adult caregivers of these infants. No neonate was infected with serogroup 23, the fourth most common serogroup (11%) for infected nonneonates during this study period but uncommon in children ≤2 years of age5 (data not shown).
Neonatal isolates were more likely to be penicillin susceptible than isolates recovered from nonneonates (21.4% vs 33.8% nonsusceptible). This difference was most pronounced when comparing middle ear isolates (25% vs 51% nonsusceptible; Fig 3), although none of these differences reached statistical significance, likely because of the small sample size of the neonatal group.
Pneumococcus is not considered to be normal vaginal flora. However, pneumococcal pelvic infections are known to occur in women associated with pneumonia, surgery, foreign bodies, and parturition.10,37 The organism has been a rare cause of puerperal infections for >100 years38 and can cause morbidity and mortality in mother and newborn infant.10,37,39–42 In a review by Westh et al,37 7 (30.4%) of 23 mothers of infants with early-onset (<5 days) invasive pneumococcal disease had clinical signs of infection, 5 with endometritis and 2 with meningitis.
Several studies have demonstrated genital colonization with S pneumoniae to be exceptionally rare (≤0.03%).43,44 As several authors have previously noted, the overall rarity of genital tract colonization with S pneumoniae suggests a high invasion to colonization rate in neonates for this organism. Therefore, these authors suggest that serious consideration be given to intrapartum prophylaxis or strategies similar to those used with GBS disease for newborns born to women with positive S pneumoniae cultures.43,44 Of note, some investigators have questioned whether early-onset SPIN may be an increasing problem in infants, most notably in developing countries.25,45,46
Neonates in this series demonstrated manifestations of either early- or late-onset sepsis as has been described for other neonatal pathogens. The majority of neonates reported here presented in the second or third week of life with invasive disease, such as meningitis, bacteremia, and pneumonia, and did not have obvious perinatal risk factors. In addition, bilateral OM with perforation was common. The source of acquisition of pneumococci in these infants is not clear. Our data suggest a role for maternal vertical transmission presenting as early invasive disease and horizontal transmission from siblings or perhaps adult close contacts, leading to local disease (OM) or invasive disease with nonvaccine serogroups.
CHILDREN`S NURSES AND DOCTORS AGAINST THE EXPLOITATION OF POOR COUNTRIES THROUGH ARMS TRADING
“The inhuman trade in weapons has devastating effects on the health and well being of children. Governments in rich ‘developed’ countries are the main source of arms. These are initially traded legally. Later they are acquired by criminals who sell them illegally to some of the most barbarous and dangerous factions usually in poor countries.
“This campaign, AK47MC, aims to reduce arms trading, both legal and illegal, and thereby save lives and reduce suffering in mothers and children. The effects of the arms trade on mothers and children 1,2 have been recently addressed in an article in the British Medical Journal. Our campaign, utilizing the support of as many national and international organizations of children’s nurses and pediatricians as possible, seeks to modify this trade. …
“All we are asking at this stage is a written expression of support for the aims listed on our website: http://www.ak47mc.org.”
Noted by JFL, MD
This study was supported, in part, by a grant from Roche Laboratories.
We thank Tricia Morphew, MS, Christine Johnson DeRosa, PhD, and Earl Leonard, MS, for statistical support and Samantha Stone for data entry. We also acknowledge the following individuals for invaluable help with data collection: Susana Aragon, RN, Nancy Tucker, RN, Andrea Forbes, RN, and Bev Petrites, RN, CCRC.
- 3.Black S, Shinefield H, Ling S, et al. Effectiveness of heptavalent pneumococcal conjugate vaccine in children younger than five years of age for prevention of pneumonia. Northern California Kaiser Permanente Vaccine Study Center Group. Pediatr Infect Dis J.2002;21 :810– 815
- 6.↵Applebaum PC. Resistance among Streptococcus pneumoniae: implications for drug selection. Clin Infect Dis.2002;34 :1613– 1620
- 7.↵Kaplan SL, Mason EO. Management of infections due to antibiotic-resistant Streptococcus pneumoniae.Clin Microbiol Rev.1998;11 :628– 644
- 8.↵Klein JO. Bacterial sepsis and meningitis. In: Remington JS, Klein JO, eds. Infectious Diseases of the Fetus and Newborn Infant. Philadelphia, PA: WB Saunders; 2001:943–998
- 23.↵Bortolussi R, Thompson TR, Ferrieri P. Early-onset Pneumococcal sepsis in newborn infants. Pediatrics.1977;60 :352– 355
- 27.↵Kaplan SL, Mason EO, Barson WJ, et al. Three-year multicenter surveillance of systemic Pneumococcal infections in children. Pediatrics.1998;102 :538– 545
- 28.↵Mason EO Jr, Lamberth L, Lichenstein R, Kaplan SL. Distribution of Streptococcus pneumoniae resistant to penicillin the USA and in vitro susceptibility to selected oral antibiotics. J Antimicrob Chemother.1995;36 :1043– 1048
- 29.↵National Committee for Clinical Laboratory Standards. Performance Standards for Antimicrobial Susceptibility Testing; 12th Informational Supplement. M100-312. Wayne, PA: National Committee for Clinical Laboratory Standards; 2002
- 30.↵Ginocchio CC. Role of NCCLS in antimicrobial susceptibility testing and monitoring. Am J Health Syst Pharm.2002;59(8 suppl 3) :S7– S11
- 31.↵Herzog LW, Coyne LJ. Normal temperature in infants less than 3 months of age. Clinical Pediatr.1993;32 :142– 146
- 32.↵Tetzlaff TR, Ashworth C, Nelson JD. Otitis media in children less than 12 weeks of age. Pediatrics.1977;59 :827– 832
- 35.↵Hausdorff WP, Bryant J, Kloek C, Paradiso PR, Siber GR. The contributions of specific pneumococcal serogroups to different disease manifestations: implications for conjugate vaccine formulation and use, part II. Clin Infect Dis.2000;30 :122– 140
- 36.↵Hausdorff WP, Bryant J, Paradiso PR, Siber GR. Which pneumococcal serogroups cause the most invasive disease: implications for conjugate vaccine formulation and use, part II. Clin Infect Dis.2000;30 :100– 121
- 39.↵Tempest B. Pneumococcal meningitis in mother and neonate. Pediatrics.1974;53 :759– 760
- 40.Endometritis and neonatal sepsis due to Streptococcus pneumoniae.Obstet Gynecol1979;53(3 suppl) :47S– 49S
- 41.Tarpay MM, Turbeville DF, Krous HF. Fatal Streptococcus pneumoniae type III sepsis in mother and infant. Am J Obstet Gynecol.1980;136 :257
- 44.↵Primhak RA, Tanner MS, Spencer RC. Pneumococcal infection in the newborn. Arch Dis Child.1993;69 :317– 318
- 46.↵Molyneux E, Walsh A, Phiri A, Molyneux M. Acute bacterial meningitis in children admitted to the Queen Elizabeth Central Hospital, Blantyre, Malawi in 1996–97. Trop Med Int Health.1998;8 :610– 618
- Copyright © 2003 by the American Academy of Pediatrics