Introduction

Journal of Neonatology

0973-21870973-2179

National Neonatology Forum

jn-20-1-003

REVIEW ARTICLE

Laboratory diagnosis of Neonatal Sepsis

Datta

Sanghamitra

Oberoi

Jaswinder K.

Chugh

T.D.

Department of Microbiology, Sir Ganga Ram Hospital, New Delhi - 110 060.

*E-mail: chughtd@hotmail.com

1–32006

2011623

Introduction

Neonatal septicaemia with its high incidence and its grave prognosis, in spite of adequate treatment with modern antibiotics, has been a challenge for all times. Optimal diagnosis and treatment strategies are difficult to define. The signs and symptoms are protean with a high mortality and thus there is an urgent need to know whether the baby has sepsis, to institute treatment as quickly as possible. Confirmation of the diagnosis may take time, and diagnostic tests are usedto obtain a rapid indication of the infection. Definitive culture tests are not rapid or particularly sensitive. Of all newborns, only 3 to 5 per 1000 admitted to a newborn intensive care unit have culture-proven septicaemia (1,2). The screening tests such as WBC count and acute phase reactants have at best a positive predictive value of 40% (3). Thus, the result of a diagnostic test must be evaluated in the light of the clinical condition of the baby.

Classification

Neonatal sepsis may be categorized as early or late-onset sepsis (EOS or LOS). Early-onset neonatal sepsis has a frequency of 0.5 to 5 cases per thousand live birth. The disease becomes more severe as the gestational age decreases. 85% of newborns with early-onset infection present within 24 hours, 5% present at 24–48 hours, and a smaller percentage of patients present beyond 48 hours of life. Early-onset sepsis syndrome is associated with acquisition of microorganisms from the mother. Transplacental infection or an ascending infection from the cervix may be caused by organisms that colonize in the mother's genitourinary tract. The infant may acquire the microbe by passage through a colonized birth canal during delivery. The microorganisms most commonly reported to be associated with early-onset infection include group B Streptococcus (GBS), Escherichia coli, Haemophilus influenzae, and Listeria monocytogenes (4). However, gram negative bacilli are the predominant causative organisms for early onset sepsis in the developing world.

Late-onset sepsis syndrome is acquired from the environment. Organisms that have been implicated in causing late-onset sepsis syndrome include coagulase-negative Staphylococci, Staphylococcus aureus, E coli, Klebsiella, Pseudomonas, Enterobacter, Candida, GBS, Serratia, Acinetobacter, and anaerobes. The infant's skin, respiratory tract, conjunctiva, gastrointestinal tract, and umbilicus may become colonized from the environment, leading to the possibility of late-onset sepsis from invasive microorganisms. Vectors for such colonization may include vascular or urinary catheters, other indwelling lines, or contact from caregivers with bacterial colonization (5,6). In India 60% of the neonatal sepsis is caused by Gram-negative organisms (1). The rogues gallery is enumerated in Table 1. Table 1

National Neonatal-Perinatal Database 2002–2003 (1)

Organisms	Incidence

K.pneumoniae	27.52%
S. aureus	14.89%
E.coli	11.6%
Pseudomonas sp.	12.98%
Coagulase negative Staphylococcus	3.76%
Enterobacter sp.	5.96%
Acinetobacter sp.	3.40%
S.viridans	0.78%
Others	17.16%
Citrobacter sp.	1.99%
GroupB Streptococcus	nil

Pneumonia is more common in early-onset sepsis, whereas meningitis and/or bacteremia are more common in late-onset sepsis. Premature and ill-infants have an increased susceptibility to sepsis and subtle nonspecific initial presentations; therefore, they require a vigilant watch so that sepsis can be identified and treated effectively (7,8). Sepsis in very low birth weight babies (VLBW) is classified as early-onset neonatal sepsis, late-onset (>3 days) and late-late-onset sepsis (>3 months) which frequently occur in babies who require prolonged intensive care and have a central venous catheter, gastrointestinal disease, or chronic lung disease has a better outcome.

Neonatal sepsis syndrome can also be caused by infection with adenovirus, enterovirus, or coxsackievirus. Additionally, sexually transmitted diseases and viral diseases, such as gonorrhoea, syphilis, herpes simplex virus (HSV), cytomegalovirus (CMV), hepatitis, HIV, rubella, toxoplasmosis and Trichomonas vaginalis, have all been implicated in neonatal infection (3).

Risk factors for neonatal sepsis

The major potential risk factors of neonatal sepsis along with the incidence of proven sepsis given the presence of risk factors are listed in Table 2. Premature rupture of membrane (PROM) plus two other risk factors raises the risk of sepsis to 25-fold (9). Table 2

Risk factors of Neonatal Sepsis (9,10).

Conditions	Incidence of proven sepsis

PROM > 18 hours	1%
Maternal + GBS	0.5–1%
Maternal + GBS prophylaxis	0.2–0.4%
Maternal + GBS + PROM	4–7%
Chorioamnitis	3–8%
PROM + Preterm	4–6%
PROM + Low APGAR	3–4%

The rates of infection in neonates and their mortality are listed below (10).

Proven sepsis: 2.5 per 1000 live births, mortality 8.7%. Clinical sepsis: 3.6 per 1000 live births, mortality 4.3% VLBW (very low birth weight) proven sepsis: 26.5 per 1000 live births. VLBW clinical sepsis: 32.4 per 1000 live births. Early-onset GBS: 0.5 per 1000 live births, mortality 10%. Other early-onset infection: 0.5 per 1000 live births.

Laboratory diagnosis

The clinical signs of neonatal sepsis are subtle and nonspecific and often the diagnosis is difficult to establish on the basis of clinical findings alone (7). Hence, the laboratory has a very significant role to play in the diagnosis of neonatal septicemia. A two-pronged approach is used for the evaluation of neonates with possible sepsis. Non-specific screen tests are used to evaluate the likelihood of infection, and specific diagnostic tests are performed to confirm the presence of a specific pathogen in body fluids.

Definitive, specific diagnostic tests are isolation of bacteria from a central body fluidBlood culture

The “gold standard” for diagnosing sepsis remains the blood culture though the sensitivity is 50–80%. The increased use of maternal antibiotics has reduced the rate of positive blood cultures in early-onset sepsis (11). In addition bacteraemia may be transient in early stages of the disease, and the small volume of blood may be insufficient to detect low bacterial density sepsis in infants. The successful recovery of microorganisms dependent on many factors like specimen collection, blood volumes and the number and timing of blood cultures (12).

Table 3 gives the profile of organisms isolated at SGRH during the year 2004. Klebsiella species are the most common causative organisms for bacterial sepsis both amongst infants with early onset and late onset sepsis. Table 3

Profile of Organisms in Babies with Sepsis in NICU SGRH - 2004 n(%)

ORGANISM	EOS n(%)	LOS n(%)

Candida	_____	4 (14.8)
E.coli	4 (36.4)	3 (11.1)
Enterobacter		4 (14.8)
Klebsiella	5 (45.5)	14 (51.9)
Strept.	1 (9.1)
Citrobacter		1 (3.7)
Acinetobacter	1 (9.1)	1 (3.7)

Specimen collection

Skin preparation before the collection of blood is of paramount importance to reduce the risk of introducing contaminants into the blood culture medium. In spite of aseptic collection of blood samples 1–3% of blood cultures become contaminated with organisms such as diphtheroids, Propionobacterium sp. and coagulase negative staphylococci which are skin colonizers (13). The vein from which the blood is to be drawn must be chosen before the skin is disinfected. It is not desirable to draw blood through a vascular shunt or a catheter, because these prosthetic devices are difficult to decontaminate completely.

Universal precautions are taken while drawing the blood. Using 70% isopropyl or ethyl alcohol the venipuncture site is cleansed in a circle approximately 5 cms in diameter and then allowed to air dry. Starting at the center of the site, the swab is moved concentrically with 1–2% tincture of iodine or chlorhexidine and allowed to dry on the skin for at least one minute. Chlorhexidine is preferred in neonates because the tender skin may be sensitive to iodine. The timing is critical and a watch or a timer should be used. The phlebotomist must also disinfect the gloved fingers in an identical manner before withdrawing the blood. After drawing the blood the site should be cleaned with 70% alcohol again as many patients are sensitive to iodine. The needle should not be changed before injecting the blood in the culture bottles. Iodine should not be used to cleanse the septa of the blood culture bottles (12).

Specimen volume

In infants the concentration of microorganisms per ml of blood is higher than adults i.e. the yield is 10 CFU of bacteria per ml of blood. 1–5 ml is usually required for bacterial culture but quantities less than 1 ml may not be adequate to detect pathogens. In BacT/ALERT Paediatric bottle an inoculum volume of 1–4 ml is recommended.

Culture techniques

Conventional blood culture techniques are labour-intensive and time consuming. During these times of cost constraints of health care, and a corresponding requirement of clinically relevant care, the development of instrumentation for blood cultures has been accomplished. Laboratories that process many blood cultures can provide results cost effectively.

BacT/ALERT automated blood culture system measures the CO ₂ derived pH changes by a colometric sensor in the bottom of each bottle. The sensor is separated from the broth medium by a membrane that is only permeable to CO₂. As organisms grow they release CO₂ which diffuses across the membrane and is dissolved in water present in the matrix of the sensor. As CO₂ is dissolved, free hydrogen ions are generated. These free hydrogen ions cause a color change in the sensor which is read by the instrument. Within 618 hours of incubation most bacteria responsible for a clinically significant disease are present in numbers large enough to give a positive signal (12). The advantage of continuous monitoring blood-culture system is that the subcultures are done as soon as the bottle flags positive along with microscopy and reporting of the slide enables a rapid institution of empiric therapy prior to giving a sensitivity report as compared to the blind subcultures done from conventional bottles at definite interval. The supplements present in the culture broths enable the growth of fastidious organisms. The mechanical shaking of these bottles at regular intervals also hasten the growth of the bacteria. The presence of resin, charcoal and polyanathol sulfonate reduces the deleterious effects of antibiotics and other inhibitory substances in blood which inhibit the growth of bacteria.

Lumbar puncture

Early-onset meningitis is a rare event and occurs in 0.25/1000 live-births in the era before maternal GBS prophylaxis. But upto 37% of cases can be missed if lumbar puncture (LP) is not routinely performed as a part of sepsis work up (14). A reasonable recommendation is to perform LP in neonates with symptoms of meningitis, in symptomatic babies in whom sepsis is the leading diagnosis, and in babies with positive blood cultures. Once CSF is obtained bacterial cultures are very reliable in the absence of pretreatment. CSF may contain very few organisms per ml of fluid so concentration of the specimen is recommended. CSF collected with universal precautions under aseptic technique following LP is taken in three bottles. The first bottle for biochemical parameters, second for microbiological examination and the third for haematological and cytological parameters. CSF specimens should be transported to the laboratory without delay and should not be refrigerated. Normal CSF is clear and colourless with a viscosity similar to water. Turbidity or cloudiness begins to appear with leukocyte counts over 200 cells/ml or red blood cell counts over 400 cells/ml. Cell count are performed on undiluted CSF in a manual counting chamber. The normal leukocyte cell counts in neonates is 0–30 cells/ml. All positive CSF gram stain are reported immediately. A positive gram stain is often confirmed by a CSF antigen test which enables the early institution of the right antibiotic therapy prior to the sensitivity report which can be done only if the culture is positive.

Urine culture

Urine cultures are most appropriate while investigating late-onset sepsis. About 1% neonates have bacteria in bladder urine. Supra-pubic aspiration is preferred. More than 10 leucocytes per cmm or >10 leucocyte/ml. in uncentrifuged urine is pyuria and a colony count of more than or equal to 10⁵ is significant bacteriuria and correlates with presence of 1 bacterium per oil immersion field (examination of 20 fields is required.). In suprapubic aspirates any number is significant. A straight non-dwelling catheter is also used for collection of urine sample under aseptic measures though risk exists of urethral organisms being introduced in the bladder, lower urethral colonizers might also give a false positive report (12).

Tracheal aspirate culture

Although differentiation of colonization from infection can be problematic in chronically ventilated patients, it has proven useful when obtained within first 12 hours of life of the baby. A positive culture may be found in 44% of infants with clinical pneumonia. It increases diagnostic accuracy when obtained from infants with suspected sepsis who require intubation and ventilation for pneumonia or respiratory failure (3). An acceptable specimen yields less than 10 squamous epithelial cells per low power field or the presence of 25 or more polymorphs with few squamous cells denotes an excellent sample quality. Growth with a quantitation of 10⁴–10⁵ CFU/ml is significant in a good quality specimen.

Detection of bacterial antigens

Culture studies though specific are time-consuming. Bacterial antigen detection in CSF even if the organisms are killed after institution of therapy is a rapid method of diagnosis. The sensitivity is 72–80% for CSF. It is slightly lower for urine. Urinary antigen tests for E.coli, S. pneumoniae and GBS can be done. The test has a sensitivity of 67% and a positive predictive value of 56% thus falling short of criteria expected for definitive diagnostic test (15). CSF antigen by latex agglutination test is a rapid card test and can confirm a gram stain as cultures though the gold standard may be negative if samples are collected after institution of antibiotics.

Adjunctive, nonspecific diagnostic and screening tests

The difficulties in identifying the septic neonate has prompted evaluation of the adjunctive tests that may indicate possibility of infection, but do not identify the inciting organism. It is important for these tests to miss no cases (high sensitivity) and rule out cases when disease is not present (high negative predictive accuracy). Many acute phase reactants and cytokines have shown to correlate with the diagnosis of neonatal sepsis but very few have a high specificity. Some tests have performed well in research setting but require special handling or are not readily available in hospital laboratories. Discussion will focus on readily available tests.

Total leucocyte count

Total leucocyte count has a low predictive value because of a wide range of normal counts from 8000–20,000/cmm. Leukopaenia less than 5000/cmm or absolute neutropenia less than 1,500/cmm is usually associated with neonatal sepsis although many septic infants have higher counts giving this test a low sensitivity of 29% but reasonable specificity of 91% (3). Neutrophils may show abnormal morphology with appearance of Dohle bodies (appearance of rough endoplasmic reticulum which stain light blue with Giemsa stain), toxic granulations (eosinophilic granules in cytoplasm of neutrophils and vacuolization.

ESR

Micro ESR is a simple, inexpensive though not very reliable marker of neonatal sepsis. During neonatal period a value of more than 15mm is considered suggestive of infection. This is done by collecting capillary blood in a standard preheparinised microhaematocrit tube and reading the fall of erythrocyte column after one hour.

C-Reactive protein (CRP)

CRP has been shown to have a higher sensitivity and negative predictive value than leukocyte indices. It is synthesized in the liver in response to and as a part of inflammatory response. It is released 4–6 hours after onset of the stimulus, peaks at 24–48 hours and then diminishes over time as the inflammation resolves. Ability to produce CRP does not appear to be affected by gestational age (11). It is an inexpensive readily available automated test (16). Elevated CRP is most useful for indicating neonatal infection when determined serially with at least two measurements 12–24 hours apart. The sensitivity increases from 48–63% when a single test is performed to 84–90% when multiple CRP screens are performed over 24–48 hours following onset of symptoms.

If the CRP was > 6.0 mg/l, incidence of proven sepsis is 10% making it useful in predicting disease. Normalisation of CRP is a good indicator of resolution of infection with treatment. Sensitivity for predicting bacteraemia is organism dependant. Sensitivity is lower for gram positive than gram negative bacteria (16). Foetal distress, hyperbilirubinaemia, intraventricular haemorrhage can result in elevation of CRP levels. So it should not be used as a sole indicator for starting or stopping of therapy but viewed in context of blood culture report, clinical findings and other laboratory studies.

When this reagent is mixed with serum containing CRP at a level greater than 6 mg/l the particles agglutinate which is interpreted as being a positive sample. The reagent is also used for semiquantitation of CRP. For this purpose the sample is diluted over a range of dilutions and each is tested qualitatively. The CRP level is estimated from the last dilution with visible agglutination. The CRP is calculated in mg/l by the product of reagent sensitivity i.e. 6 mg/l and the highest dilution with a positive reaction.

Interleukins

Other acute phase reactants like cytokines namely interleukins have been studied as markers for early diagnosis of neonatal sepsis. Current evidence indicates that mot of the physiologic effects generated by bacterial infections are mediated by complex interaction of cytokines activated in response to the presence of microbial components with in the vascular compartment. Cytokines are intracellular signaling polypeptides produced by activated cells during an episode of sepsis (17). Interleukins are produced by various cells types including activated monocytes, B & T lymphocytes, fibroblasts and endothelial cells (18). IL-6 is a prototype of a family of cytokines that are chemotactic for neutrophils and monocytes. It has a very short half life and the concentrations fall precipitously with treatment and become undetectable in most infected patients within 24 hours. For early onset sepsis and determination of IL-6 within 48 hours o life, sensitivities have been reported ranging from 69% – 100% and specificities ranging from 39%–92% at different cut off values of 10, 70 and 100 pg/ml (19, 20).

Procalcitonin assay

Another acute phase marker that has attracted much attention recently is procalcitonin (PCT). The increase in circulating PCT concentration is independent of calcitonin, and PCT has been shown to be associated with neurotransmission, immunomodulation,and vascular control during infection and in the systemic inflammatory response syndrome (SIRS). Although the exact sites of production of PCT in sepsis have not been identified, monocytes and hepatic cells are believed to be potential sources. Serum concentrations of PCT begin to rise four hours after exposure to bacterialendotoxin, peak at six to eight hours, and remain raised for at least 24 hours. The half-life is estimated to be about 25–30 hours, and the serum concentrations do not appear to be affected by the gestational age. Several studies have shown that serum PCT concentrations increase appreciably in systemic bacterial infection and necrotising enterocolitis during early and late onset neonatal infection. Its diagnostic profile has been claimed to be superior to other acute phase proteins,including CRP, with sensitivity and specificity ranging from 87% to 100%. In contrast, infants with viral infection, bacterial colonisation, and noninfected inflammatory stress, such as birth trauma, aspiration syndrome, and hypoxaemia, have normal or only slightly raised concentrations. PCT has been reported to be useful in indicating the severity of infection, following progress of treatment, and predicting outcomes. Optimum cut-off values of PCT is 2.5 μg/L for the diagnosis of sepsis at birth. Determination of PCT is of value in excluding bacterial infection in neonates since it has a negative predictive value of 93%. (21)

Sepsis screens

To improve the diagnostic capability of adjunctive diagnostic tests there are several sepsis screen strategies (22). Table 4 lists the combinaton of CRP with WBC indices a sepsis screen used in Pennsylvania Health System Hospitals. 100% ability to rule out sepsis can be achieved by these screens in infants with risk of infection obtained after 12–24 hours after birth. A commonly observed pattern in a noninfected infant born from a stressful labour shows an elevation of WBC or I/T (Immature/Total neutrophil ratio) at 12 hours of age with normal CRP, and an increase in CRP at 24–48 hours of age but with normalising WBC indices. This scenario is most often benign. Most infants with sepsis have progressively abnormal WBC and CRP indices. Table 4

Sepsis Screen (screen positive if > = 2 points) (17)

Test	Point value

Absolute neutrophil <1750/mm³	1
Total WBC count <7500 or >40,000/mm³	1
Immature/Total neutrophil ratio >0.20	1
Immature/Total neutrophil ratio >0.40	2
CRP+ ve > 1mg/dl	1
CRP+ ve ≥ 5mg/dl	2

Molecular methods

Polymerase Chain Reaction (PCR) has proved to be a valuable adjunct for detection of neonatal viral infections such as human immunodeficiency virus, herpes simplex virus etc. when used in conjunction with other diagnostic tests like serology. However its use to detect bacteraemia and fungaemia is challenging. Detection of bacterial DNA in blood has been accomplished by PCR amplification of the gene for 16SrRNA which is universally present in bacteria and absent in humans (23). Those sequences are amplified with PCR, using an automated method allowing a rapid diagnosis. Studies have been conducted by using Multiplex-PCR with simultaneous amplification of more than one genetic locus to differentiate various etiological agents responsible for sepsis based on the molecular weight of the fragments amplified by PCR. Sensitivity of 96–100%, and specificity of 95.6%–99.4% has been reported (24).

PCR detection of fungaemia is another important area of investigation as preterm infants are at high risk of mortality and morbidity from invasive candida disease and earlier treatment of candidal sepsis,including central venous catheter removal is associated with improved outcome. The target of amplification is 18SrRNA gene present in all clinically relevant fungal spp. Studies have shown compared to blood cultures, PCR was 100% sensitive and 98% specific (25).

The Future

A promising approach is DNA microarray technology. Devices are developed which incorporate automated DNA analysers with integrated sample preparation and biosensing elements which could allow early, rapid and sensitive detection of small number of organisms in minute volumes of blood (26). In the near future microchip detection of blood stream infection may allow practitioners to tailor antibiotic therapy towards specific pathogens and avoid unnecessary antibiotic therapy. Given the poor outcome of neonatal sepsis despite current optimal antimicrobials and intensive care, concerted efforts and focus should be on prevention and reliable detection methods for early detection of this critical syndrome.

References1National Neonatology forum of India.National Neonatal Perinatal Database-Report for year 2002, New Delhi: NNF India; 2004.2

Gerdes

Polin

Early diagnosis and treatment of neonatal sepsis. Indian J Pediatr1998; 65: 63–78.3

Gerdes

Diagnosis and management of bacterial infections in neonate. Paediatric Clin N Am2004; 51: 939–959.4

Eichenwald

Perinatally transmitted neonatal bacterial infections. Inf Dis Clin N Am1997; 11: 223–239.5

Chugh

Ghafoor

Kuruvilla

Bishibishi

Colonisation and infections of neonates by K. pneumoniae in an intensive care unit. J Trop Paediatr1985; 31: 200–203.6

Chugh

Ghafoor

Kuruvilla

Bishibishi

Neonatal colonisation and infections with streptococci in an intensive care unit. Annals of Trop Paediatr1986; 15: 14–18.7

Gerdes

Clinicopathological approach to the diagnosis of neonatal sepsis. Clin Perinatol1991; 18 (2): 361–381.8

Chugh

Ghafoor

Kuruvilla

Bishibishi

Neonatal colonisation and infection with streptococci in an intensive care unit in Kuwait. Paediatric Infectious Diseases in Arab countries. Ed. Elzouki

A.Y.

Publisher Jon Wiley and Sons Ltd.

1987; 13: 87–93.9

Lopez Sastre

Coto Cotallo

Neonatal sepsis of vertical transmission: an epidemiological study from the “Grupo de Hospitales Castrillo.”

J Perinat Med2000; 28: 309–315.10

Isaacs

Royle

Intrapartum antibiotics and early onset neonatal sepsis caused by group B streptococcus and other organisms in Australia. Australasian Study Group for Neonatal Infections. Pediatr Infect Dis J1999; 18: 524–528.11

Kaufman

Fairchild

Clinical Microbiology of Bacterial and Fungal sepsis in very low Birth Weight Infants. Clin Microbiol Rev2004; 17 (2): 638–680.12

Betty

Daniel

Alice

Bailey & Scott's Diagnostic Microbiology

11^th ed.13American Academy of Pediatrics: Red Book 2000. 25^th ed. 2000; 20–26, 241–2, 537–543.14

Wiswell

Ganon

No lumbar puncture in the evaluation for early neonatal sepsis: will meningitis be missed?

Paediatrics1995; 95: 803–806.15

Rabalais

Bronfin

Daum

Evaluation of commercially available latex agglutination test for rapid diagnosis of Group B streptococcal infection. Paediatr Infect Dis J1987; 6: 177–181.16

Gabay

Kushnor

Mechanism of disease:acute phase proteins and other systemic responses to inflammation. N Engl J med1999; 340: 448–454.17

Gabay

Kushner

Acute phase proteins and other systemic responses to inflammation. NEJM1999; 340: 448–454.18

Santana Reyes

Garcie

Munoz

Role of cytokines (IL-1b, 6, 8 TNF –a & IL – 2) and CRP in the diagnosis of neonatal sepsis. Acta Pediatr2003; 92: 221–227.19

Messer

Eyer

Donato

Gallati

Evaluation of IL-6 and soluble receptors of tumor necrosis factor for early diagnosis of neonatal infection. J Pediatr1996; 129: 574–580.20

Panero

Pacifico

Rossi

IL-6 in neonates with early and late onset infection. Pediatr Infect Dis J1997; 16: 370–375.21

Claudio

Osborn

Diagnosis of neonatal sepsis : A clinical and laboratory challenge. Clinical chemistry2004; 50: 279–287.22

Madan

Philip

Frequency and timing of symptoms in infants screened for sepsis: effectiveness of a sepsis-screening pathway. Clin Paediatr2003; 42 (1): 11–18.23

McCabe

Khan

Zhang

Mason

McCabe

ERB

Amplification of bacterial DNA using highly conserved sequences, automated analysis and potential for molecular triage of sepsis. Pediatrics1995; 95: 165–169.)24

Yadav

Wilson

Menon

Polymerase chain reaction in rapid diagnosis of neonatal sepsis. Indian Paediatrics2005; 42: 681–685.25

Jordan

PCR identification of four medicallyimportant Candida spp. by using a single primer pair. J Clin Microbiol1994; 32: 2962–2967.26

Nissen

Sloots

Rapid diagnosis inpaediatric infectious diseases, the past, present and future. Paediatr Infect Dis J2002; 21: 605–612.