Pediatric Annals

Neonatal Sepsis

Sophie Pierog, MD; Sarvesh Nigam, MD

Abstract

The presence of similar organisms in the nose and throat, umbilical cord, and external auditory canal cultures in the first few hours of life may indicate considerable contamination, but not necessarily sepsis.7,30 However, careful consideration should be extended to the infant whose superficial cultures show the presence of beta-hemolytic streptococci, group B. These infants, contaminated by the organisms from the maternal vaginal tract, may become critically ill with sepsis and meningitis.10-12

There is no rapid laboratory test that is reliable in diagnosing sepsis of the newborn.1,5,7,10,31 Table 3 lists three groups of laboratory procedures that should be done to complete a "sepsis work-up." Group 1 tests, such as the white blood count and differential, may yield immediate confirmatory results. Group 2 tests may yield confirmatory results, but some time must elapse before the results are known. Group 3 tests must be included, since the clinical signs and symptoms of sepsis are nonspecific; these tests are necessary in making a differential diagnosis.

The newer experimental laboratory tests that may be of some value in diagnosing bacterial sepsis of the newborn are the modified nitroblue tetrazolium (NBT) test32 and the leukocyte lactate dehydrogenase (LDH) test. With the modified NBT test, investigators found that the absolute number of NBT-positive cells was greater in infected low-birth-weight infants than in noninfected ones. The leukocyte LDH values reported by Powers and Ayoub33 in a preliminary study seemed to be significantly elevated in newborn infants with neonatal meningitis and sepsis.

TREATMENT

Since there is frequently no rapid or reliable laboratory test to confirm the diagnosis of infection in the newborn, vigorous treatment with antimicrobials should be initiated at once, pending laboratory results. The infant with sepsis is sick and in need of supportive treatment. He is usually hypothermic and is not taking adequate amounts of fluids; caloric intake is poor; and such gastrointestinal symptoms as diarrhea, vomiting, and abdominal distention may be present.

General Supportive Treatment

1. The infant should be placed in an isolette in a special-care nursery, since careful observation and monitoring of vital signs are necessary.

2. The infant's temperature should be maintained at homoiothermic levels (97-98° F. or 36-37° C).

3. Oxygen should be provided for an infant who is in respiratory distress, and there should be frequent monitoring of blood gases.

4. Usually, in the first few hours to days, the infant's oral feedings may be withheld; parenteral fluids should be given, and there should be frequent monitoring of serum electrolytes.

5. A nasogastric tube may need to be passed to aspirate fluid or air if there is vomiting or abdominal distention.

Antimicrobial Treatment

1. Bactericidal rather than bacteriostatic drugs should be used.

2. Antimicrobials should be given parenterally (intramuscularly or intravenously); the use of oral antibiotics in the neonate should be discouraged.

Special Management1,7,10

1. The infant with neonatal sepsis may develop hypoglycemia. Be prepared to monitor blood sugar levels and to give glucose intravenously.

2. The infant may develop hyperbilirubinemia. Be prepared to treat by an exchange transfusion or by phototherapy.

3. The infant may develop disseminated intravascular coagulopathy.34 After the appropriate tests are taken, recommended (but still somewhat controversial) treatment is:

a) Fresh whole blood may be given (10-20 ml./kg. of body weight) or

b) Fresh heparinized blood may be given as an exchange transfusion (double volume is 160 ml./kg. of body weight) or

c) Heparin (1.0 mg./kg. of body weight) should be given intravenously every four to six hours.

4. The infant may be in shock (endotoxic). Be prepared to give fresh blood at 10-20 ml./kg. of body weight.

CONCLUSIONS

Because of the seriousness of the consequences, the…

Neonatal sepsis, a generalized bacterial disease of infants documented by a positive blood culture, is a common problem in most nursery services. The incidence of neonatal sepsis is difficult to determine, since proof of infection is not always possible and many factors may influence the occurrence of sepsis or predispose the infant to infection. The frequency of occurrence of sepsis varies in different areas of the country and in different hospitals. The type of hospital (urban, rural, medical center, community, etc.), the obstetric and nursery routines in each hospital, the socioeconomic status of the population, and the prematurity rate are all important factors in determining the incidence and mortality for neonatal sepsis.1-4

The overall incidence is estimated to be 1 to 1.8 per 1,000 live births.3 Buetow et al.5 report the rate of sepsis in infants of 1,001-1,500 gm. as 164 per 1,000 live births; others6 report the risk of sepsis in preterm infants as 1 per 250 live births. Male newborns are more often afflicted than females.

Frequently, sepsis is suspected clinically but not confirmed by a positive blood culture. If only the clinical criteria are used for the diagnosis of sepsis, the overall incidence is then reported to be as high as 20 per cent, with a death rate ranging from 13 to 45 per cent.7

PATHOGENESIS

The infant may acquire infection in utero or during or after delivery.8-11 Though it is an uncommon occurrence, bacteria may cross the placenta through the bloodstream as a result of maternal bacteremia (as with Listeria monocytogenes or syphilis). More commonly, the fetus in utero is jeopardized by an ascending (vaginal) infection if the amniotic membranes of the mother have been ruptured for more than 24 hours. Gotoff and Behrman7 report evidence of chorioamnionitis in 7 to 15 per cent of all placentas, and Blanc9 reports the ratio of infected amniotic fluid to systemic infection in the neonate as between 30 and 100 to 1. The longer the duration of membrane rupture, the more likely is fetal infection; at times, however, such infection may occur with intact membranes.

In the passage through the vaginal canal and over the perineal area of the mother, the gastrointestinal or respiratory tract of the infant may be contaminated by maternal secretions. Following birth, the infant becomes infected with bacterial organisms from fomites or by personnel in the delivery room and nursery areas. Bacteria enter the infant mainly through the gastrointestinal and respiratory tracts and the umbilical cord area. Current intensive-care nursery practices - such as resuscitation, umbilical vessel cannulation, capillary or venous blood sampling, injections, and fluid administration - increase the risk of infection in the sick newborn.

ETIOLOGY AND CHANGING ECOLOGY

The most common organisms in the etiology of newborn infections are the gram-negative organisms.5-7 Current intensive newborn care practices and widespread use of antibiotics have encouraged the emergence of resistant strains of gramnegative bacteria (Klebsiella, Pseudomonas, etc.). The earlier practice of bathing the infants with hexachlorophene may have caused the disappearance of nursery epidemics caused by staphylococci and the emergence of gram-negative organisms as the causative agents of newborn infections.

Recently, however, more evidence has been accumulating of newborn infections caused by gram-positive organisms, especially group B betahemolytic Streptococcus.12-14

THE INFANT'S DEFENSE MECHANISMS AGAINST INFECTIONS

The reason for the newborn's vulnerability to infection is not completely understood; there are a number of problems in the inflammatory and immunologic response to infections by the infant.15 Though the infant's leukocytic response may be variable, there seems to be evidence that the bone marrow does respond with an increase in production and release of the immature (band) neutrophil.16-18 There is some conflicting evidence about the efficiency of phagocytosis of these leukocytes; most investigators agree, however, that there is a somewhat decreased phagocytic activity of the leukocyte - especially in the premature infant - which may be due to the lack of a serum factor.19-22

In general, however, while the leukocytes of the infant may ingest bacteria, the bactericidal ability is definitely decreased.19 Of special importance is the finding by Forman and Stiehm20 that in sick infants particularly, the phagocytic and bactericidal activity of the polymorphonuclear cell is definitely impaired.

The infant is not immunologically incompetent; immunoglobulin G (IgG) and immunoglobulin M (IgM) are synthesized in small amounts in the fetus as early as 10 to 12 weeks of gestation.23,24 The bulk of the infant's IgG, however, is derived from maternal sources in the last trimester of pregnancy; when challenged with an antigen (even in utero), the infant responds promptly with an elevation of the serum IgM.25 Synthesis of IgA is delayed until the 30th week of gestation and only slowly rises to detectable levels several days after birth.22 Complement production by the fetus also begins in utero, even before the production of immunoglobulins; however, complement levels are only 50 per cent of normal adult values.26,27 Poor suck, gag, and cough reflexes and increased skin permeability also contribute to the newborn's predilection to infection.

Table

TABLE 1CLASSIFICATION OF CLINICAL SIGNS AND SYMPTOMS OF INFECTION ACCORDING TO SYSTEMS

TABLE 1

CLASSIFICATION OF CLINICAL SIGNS AND SYMPTOMS OF INFECTION ACCORDING TO SYSTEMS

Table

TABLE 2PERINATAL EVENTS THAT PREDISPOSE THE INFANT TO INFECTION

TABLE 2

PERINATAL EVENTS THAT PREDISPOSE THE INFANT TO INFECTION

CLINICAL DIAGNOSIS OF INFECTION

Since the morbidity and mortality of neonatal sepsis are high and the clinical course of the disease may be rapidly fulmonating, emphasis must be placed on early detection. Clinical signs and symptoms of sepsis are often minimal, vague, and nonspecific and may mimic other pathologic conditions. Typically, the infant is noted by the nursery staff (or by the mother) to be "doing poorly," and there may be subtle changes in behavior of the infant. Manifestations of sepsis may involve several organ systems as summarized in Table 1. Maternal history and perinatal events may suggest a predisposition of the infant to sepsis (Table 2) and should be considered when a preliminary diagnosis of newborn infection is being made.

LABORATORY EVIDENCE TO SUPPORT THE DIAGNOSIS

After the clinical evidence for sepsis is weighed, a laboratory evaluation should be initiated to confirm the diagnosis (Table 3). Most of the laboratory tests are nonspecific; bacteriologic confirmation is necessary to establish the diagnosis of newborn sepsis. Care must be taken in obtaining specimens for bacteriology. Blood cultures should be drawn from a peripheral vein following careful preparation of the skin with an iodinated solution. It would be desirable to obtain at least two blood cultures from the infant before treatment is started.28,29 From a small infant, no more than 1 ml. of blood can usually be withdrawn and the ratio of blood to culture medium should be about 1 to 10. The value of a single positive blood culture is sometimes questioned (though not completely disregarded), since it may represent bacterial presence secondary to skin contamination or a bacteremia without actual sepsis. The presence of the same organism in two or more cultures almost always confirms a diagnosis of sepsis.

Table

TABLE 3WORK-UP FOR SEPSIS IN THE NEWBORN

TABLE 3

WORK-UP FOR SEPSIS IN THE NEWBORN

Spinal fluid from a nontraumatic tap can reveal valuable information. Xanthochromia may normally be present in 50 per cent of spinal taps at this age and may be due to hemolyzed red blood cells; turbid or cloudy fluid may represent the pleocytosis of meningitis. The fluid should be cultured immediately, and sediment should be examined for the presence of leukocytes and gramstained for the presence of bacteria. A leukocyte count of 30 or more white blood cells per high-power field may be indicative of meningitis. The presence of bacteria would serve as a prompt indicator of appropriate antibiotic therapy; bacteria may be present in early stages of sepsis with little or no cellular response. The protein concentration of newborn cerebrospinal fluid (CSF) may not be helpful, since the content of protein in newborn CSF may normally be as high as 200 mg./100 ml.

Table

TABLE 3WORK-UP FOR SEPSIS IN THE NEWBORN

TABLE 3

WORK-UP FOR SEPSIS IN THE NEWBORN

The presence of similar organisms in the nose and throat, umbilical cord, and external auditory canal cultures in the first few hours of life may indicate considerable contamination, but not necessarily sepsis.7,30 However, careful consideration should be extended to the infant whose superficial cultures show the presence of beta-hemolytic streptococci, group B. These infants, contaminated by the organisms from the maternal vaginal tract, may become critically ill with sepsis and meningitis.10-12

There is no rapid laboratory test that is reliable in diagnosing sepsis of the newborn.1,5,7,10,31 Table 3 lists three groups of laboratory procedures that should be done to complete a "sepsis work-up." Group 1 tests, such as the white blood count and differential, may yield immediate confirmatory results. Group 2 tests may yield confirmatory results, but some time must elapse before the results are known. Group 3 tests must be included, since the clinical signs and symptoms of sepsis are nonspecific; these tests are necessary in making a differential diagnosis.

The newer experimental laboratory tests that may be of some value in diagnosing bacterial sepsis of the newborn are the modified nitroblue tetrazolium (NBT) test32 and the leukocyte lactate dehydrogenase (LDH) test. With the modified NBT test, investigators found that the absolute number of NBT-positive cells was greater in infected low-birth-weight infants than in noninfected ones. The leukocyte LDH values reported by Powers and Ayoub33 in a preliminary study seemed to be significantly elevated in newborn infants with neonatal meningitis and sepsis.

TREATMENT

Since there is frequently no rapid or reliable laboratory test to confirm the diagnosis of infection in the newborn, vigorous treatment with antimicrobials should be initiated at once, pending laboratory results. The infant with sepsis is sick and in need of supportive treatment. He is usually hypothermic and is not taking adequate amounts of fluids; caloric intake is poor; and such gastrointestinal symptoms as diarrhea, vomiting, and abdominal distention may be present.

General Supportive Treatment

1. The infant should be placed in an isolette in a special-care nursery, since careful observation and monitoring of vital signs are necessary.

2. The infant's temperature should be maintained at homoiothermic levels (97-98° F. or 36-37° C).

3. Oxygen should be provided for an infant who is in respiratory distress, and there should be frequent monitoring of blood gases.

4. Usually, in the first few hours to days, the infant's oral feedings may be withheld; parenteral fluids should be given, and there should be frequent monitoring of serum electrolytes.

5. A nasogastric tube may need to be passed to aspirate fluid or air if there is vomiting or abdominal distention.

Antimicrobial Treatment

1. Bactericidal rather than bacteriostatic drugs should be used.

2. Antimicrobials should be given parenterally (intramuscularly or intravenously); the use of oral antibiotics in the neonate should be discouraged.

Special Management1,7,10

1. The infant with neonatal sepsis may develop hypoglycemia. Be prepared to monitor blood sugar levels and to give glucose intravenously.

2. The infant may develop hyperbilirubinemia. Be prepared to treat by an exchange transfusion or by phototherapy.

3. The infant may develop disseminated intravascular coagulopathy.34 After the appropriate tests are taken, recommended (but still somewhat controversial) treatment is:

a) Fresh whole blood may be given (10-20 ml./kg. of body weight) or

b) Fresh heparinized blood may be given as an exchange transfusion (double volume is 160 ml./kg. of body weight) or

c) Heparin (1.0 mg./kg. of body weight) should be given intravenously every four to six hours.

4. The infant may be in shock (endotoxic). Be prepared to give fresh blood at 10-20 ml./kg. of body weight.

CONCLUSIONS

Because of the seriousness of the consequences, the physician probably tends to overdiagnose neonatal sepsis and many infants may needlessly be subjected to antimicrobial therapy.

It should be emphasized that the aim is always toward prevention; good obstetric care will reduce the incidence of delivery of compromised infants who are susceptible to infection. Strict attention should be paid to hand washing and vigilance maintained over the cleanliness of the nursery environment and equipment used in the nursery. Periodic bactériologie surveillance should be established, and the sensitivity pattern of the bacterial flora of the nursery and delivery room should be determined. Nursery practices should be evaluated and updated periodically. In this way, the quantity of contaminants that are potential pathogens to the sick infant will be kept at a minimum.

BIBLIOGRAPHY

1. Wilson, H. D.. and Eichenwald, H. F. Sepsis neonatorum. Pediatr. Clin. North Am. 21 (1974). 571-582.

2. Gluck, L. Wood, H. R. and Fousek. M. D. Septicemia of the newborn. Pediatr. Clin. North Am. 13 (1966), 1131-1148.

3. Overall, J. C, Jr. Neonatal bacterial meningitis: Analysis of predisposing factors and outcome compared with matched control subjects. J. Pediatr. 76 (1970), 499-511.

4. Naeye, R , and Blanc, W. A. Relation of poverty and race to antenatal infection N. Engl. J. Med. 283 (1970), 555-560.

5 Buetow. K. C. Klein, S. W , and Lane, R B. Septicemia in premature infants Am. J. Dis. Child. 110 (1965), 29-41.

6 McCracken. G. H . and Shinefield. H. R. Changes in the pattern of neonatal septicemia and meningitis. Am. J. Dis. Child. 112 (1966), 33-39.

7. Gotoff. S. P.. and Behrman. R. E. Neonatal septicemia. J. Pediatr. 76 (1970). 142-153.

8. Plotkin. S. A. Routes of fetal infection and mechanisms of fetal damage. Am. J. Dis. Child. 129 (1975). 444-449.

9. Blanc, W. A. Pathways of fetal and early neonatal infection. J. Pediatr. 59 (1961). 473-496.

10. Oavies, P. A. Bacterial infection in the fetus and newborn. Arch. Dis. Child. 46 (1971), 1-27.

11. Ray, C. G., and Wedgwood. R. J. Neonatal listeriosis. Six case reports and a review of the literature. Pediatrics 34 (1964). 378-392.

12. Franciosi. R. ?.. Knostman, J. D.. and Zimmerman. R. A. Group B streptococcal neonatal and infant infections. J. Pediatr. 82 (1973). 707-718.

13. McCracken. G. H. Group B streptococci: The new challenge in neonatal infections. J. Pediatr. 82 (1973). 703-706.

14. Eickhoff, T. C. et al. Neonatal sepsis and other infections due to group B beta-hemolytic streptococci. N. Engl. J. Med. 271 (1964), 1221-1228.

15. Stiehm, E. R. Fetal defense mechanisms. Am. J. Dis. Child. 129 (1975). 438-443.

16. Xanthou, M. Leukocyte blood picture in ill newbom babies. Arch. Dis. Child. 45 (1970), 242-249.

17. Gregory. J., and Hey, E. Blood neutrophil response to bacterial infection in the first month of life. Arch. Dis. Child. 47 (1972). 747-753.

18. Akenzua, G. I., et al. Neutrophil and band counts in the diagnosis of neonatal infections. Pediatrics 54 (1974). 38-42.

19. Coen, R., Grush, A., and Kauder, E. Studies of bactericidal activity and metabolism of the leukocyte in full-term neonates. J. Pediatr. 75 (1969), 400-406.

20. Forman, M. L, and Stiehm, E. R. Impaired opsonic activity but normal phagocytosis in low-birthweight infants. W. Engl. J. Med. 281 (1969), 926-931.

21 . Gluck, L., and Silverman, W. A. Phagocytosis in premature infants. Pediatrics 20 (1957). 951-953.

22. Dossett. J. K1 Williams. R. C1 Jr.. and Quie, P. G. Studies on interaction of bacteria, serum factors and polymorphonuclear leukocytes in mothers and newborn. Pediatrics 44 (1969), 49-61.

23. Lawton. A. R.. and Cooper. M. D. Development of immunity: Phytogeny and ontogeny. In Stiehm, E. R., and Fulginiti, V. A. (eds.). Immunologic Disorders in Infants and Children. Philadelphia: W. B. Saunders Company, 1973, pp. 28-41.

24. Van Fruth, R.. Schmit, H. R. E., and Hijmans. W. The immunologic development of the human fetus. J. Exp. Med. 122 (1965). 1173-1187.

25. Alford, C A., Stagno, S.. and Reynolds. D. W. Diagnosis of chronic perinatal infections. Am. J. Dis. Child. 129 (1975). 455-463.

26. Köhler. P. F. Maturation of the human complement system. J. Clin. Invest. 52 (1973), 671-677.

27. Fireman, P.. Zuchowski, D. A., and Taylor, P. M. Development of human complement system. J. Immunol. 103 (1969). 25-31.

28. Eitzman, D. V., and Smith, R. T. The significance of blood cultures in the newborn period. Am. J. Dis. Child. 94 (1957). 601-609.

29. Franciosi. R. A., and Favara, B. E. A single blood culture for confirmation of the diagnosis of neonatal septicemia. Am. J. Clin. Pathol. 57 (1972), 215-219.

30. Evans. H. E., et al. Bacterial flora of newborn infants in the external auditory canal and other sites. N.Y. State J. Med. 73 (1973), 1071-1072.

31. Evans, H. E., Glass, L., and Mercado, C The micro-erythrocyte sedimentation rate in newborn infants. J. Pediatr. 76 (1970), 448-451.

32. Kalpaktsoglou, P. K., et al. Evaluation of nitroblue tetrazolium test in low-birth-weight infants. J. Pediatr. 84 (1974), 441-443.

33. Powers. D. W.. and Ayoub. E. M. Leukocyte lactate dehydrogenase in bacterial meningitis. Pediatrics 54 (1974), 27-33.

34. Abildgaard, C F. Recognition and treatment of intravascular coagulation. J. Pediatr. 74 (1969). 163-176.

TABLE 1

CLASSIFICATION OF CLINICAL SIGNS AND SYMPTOMS OF INFECTION ACCORDING TO SYSTEMS

TABLE 2

PERINATAL EVENTS THAT PREDISPOSE THE INFANT TO INFECTION

TABLE 3

WORK-UP FOR SEPSIS IN THE NEWBORN

TABLE 3

WORK-UP FOR SEPSIS IN THE NEWBORN

10.3928/0090-4481-19760201-08

Sign up to receive

Journal E-contents