Andreas Ohlin

Evaluation of a commercial multiplex PCR test (SeptiFast) in the etiological diagnosis of community-onset bloodstream infections

The commercial polymerase chain reaction (PCR) test, SeptiFast, is designed to identify the DNA of individual bacterial and fungal pathogens in whole blood. We aimed to evaluate the usefulness of the test for the detection of community-onset bloodstream infections. We prospectively included adult patients who were subjected to blood culture (BC) at an infectious diseases department. For the evaluation, one BC/PCR set (two BC bottles and one PCR tube) per patient was used. When several sets were obtained and analyzed, the first set with any positive result was evaluated. Among 1,093 consecutively included patients, BC was positive in 138 and PCR was positive in 107. Fifty positive PCR results were supported by BC in the same BC/PCR set, ten were supported by other cultures, and, additionally, ten were supported by the clinical presentation. Compared with BC, PCR showed specificities and negative predictive values of >97% for all detectable pathogens. The following sensitivities and positive predictive values (PPVs) were noted: Staphylococcus aureus, 67% and 43%; Streptococcus pneumoniae, 12% and 67%; other Streptococcus species, 43% and 77%; Escherichia coli, 53% and 56%; and Klebsiella species, 43% and 23%. If support from other cultures and the clinical presentation were included in the reference standard, the PPVs for the detection of these bacteria were 57%, 100%, 92%, 75%, and 69%, respectively. Although the specificities were high, the low sensitivities and suboptimal PPVs noted in the present study discourage routine use of the test in its present form for the detection of community-onset bloodstream infections.

Real‐time PCR of the 16S‐rRNA gene in the diagnosis of neonatal bacteraemia

Objective: To evaluate a real‐time PCR assay for the diagnosis of neonatal bacteraemia.

Multiple group-specific sequencing primers for reliable and rapid DNA sequencing

Pyrosequencing technology is a bioluminometric DNA sequencing method that employs a cascade of four enzymes to deliver sequence signals. To date this technology has been limited to the sequencing of short stretches of DNA. As an improvement to this technique, we have introduced a bacterial group-specific, multiple sequencing primer approach that circumvents sequencing of less informative semi-conservative regions of the 16S rRNA gene. This new approach is suitable for challenging templates, improving sequence data quality, avoiding sequencing of non-specific amplification products, lessening sequencing time, and moreover, this strategy should open the way for many new applications in the future. The group-specific, multiple sequencing primers can be applied in the Sanger dideoxy sequencing method as well. In addition, we have improved the chemistry of the Pyrosequencing system enabling sequencing of longer stretches of DNA, which allows numerous new applications.

Diagnosis of neonatal sepsis by broad-range 16S real-time polymerase chain reaction.

Background: The standard diagnostic test (blood culture) for suspected neonatal sepsis has limitations in sensitivity and specificity, and 16S polymerase chain reaction (PCR) has been suggested as a new diagnostic tool for neonatal sepsis. Objectives: To develop and evaluate a new real-time PCR method for detection of bacterial DNA in blood samples collected from infants with suspected neonatal sepsis. Methods: Immediately after blood culture, a study sample of 0.5–1.0 ml whole blood was collected and used for a novel 16S real-time PCR assay. All positive samples were sequenced. Detailed case studies were performed in all cases with conflicting results, to verify if PCR could detect pathogens in culture negative sepsis. Results: 368 samples from 317 infants were included. When compared with blood culture, the assay yielded a sensitivity of 79%, a specificity of 90%, a positive predictive value of 59%, and a negative predictive value of 96%. Seven of the 31 samples with a positive PCR result and a negative blood culture had definite or suspected bacterial sepsis. In five samples, PCR (but not blood culture) could detect a pathogen that was present in a blood culture collected more than 24 h prior to the PCR sample. Conclusions: This study presents an evaluation of a new real-time PCR technique that can detect culture-positive sepsis, and suggests that PCR has the potential to detect bacteria in culture-negative samples even after the initiation of intravenous antibiotics.

Clinical signs and CRP values associated with blood culture results in neonates evaluated for suspected sepsis.

Aim:  To identify which clinical signs at presentation are most predictive of sepsis subsequently confirmed by blood culture and to investigate whether the predictive power of the clinical signs varies by gestational age.

Suboptimal care and metabolic acidemia is associated with neonatal encephalopathy but not with neonatal seizures alone: a population‐based clinical audit

To determine the incidence of moderate to severe neonatal encephalopathy (NE) and neonatal seizures without encephalopathy, and the association with metabolic acidemia. Secondly, to investigate the occurrence of suboptimal intrapartum care and its impact on neonatal outcome.

Sepsis as a risk factor for neonatal morbidity in extremely preterm infants.

This study evaluated sepsis as a risk factor for neonatal morbidities and investigated the association between specific pathogens and neonatal morbidities.

What is neonatal sepsis

In this issue of Acta Paediatrica, Patel et al. (1) present interesting results from the ELGAN study (2). This material consists of 1106 newborns delivered before 28 weeks of gestation, recruited from 14 neonatal units. The study is interesting as it focuses not only on the infants with sepsis but also on the very large group of prematures that do not have proven bacteraemia but still receive sepsis treatment. In this study, these newborns are referred to as ‘presumed bacteraemia’ and that group is of course larger than the definite bacteraemia group (555 and 358 prematures, respectively). These numbers reflect the problematic diagnostic situation for neonatal sepsis; there simply is no available diagnostic test that can detect or exclude neonatal sepsis with an acceptable sensitivity and specificity, meaning that neonatologists must initiate antibiotic treatment on the slightest suspicion. A weakness in the study is that the 206 infants (14%) that did not survive until postnatal day 28 were excluded from the original cohort. It is reasonable to believe that a majority of the newborns that had fatal neonatal sepsis died during this period, meaning that the sickest infants were excluded and the material is hence lacking patients with severe sepsis, septic chock and fatal sepsis. This might also mean that aggressive bacteria such as group B Streptococci, gram negatives and Listeria monocytogenes might be underrepresented in this study and contaminants and less aggressive bacteria such as coagulase-negative staphylococci might be over-represented. Sepsis is normally defined as a positive blood culture in combination with systemic inflammatory response syndrome. In 2005, Haque et al. (3) published suggestions for a neonatal sepsis definition in Pediatric Critical Care Medicine; in this paper, Haque concludes that there is a lack of consensus in this area, but the importance of studying clinical signs and the foetal inflammatory response is also stressed. This lack of consensus means that so far no accepted definition of neonatal sepsis is available, neither for clinical use nor for scientific studies, and in earlier studies, a wide variety of definitions has been used. As the ELGAN study did not collect detailed data on symptoms, bacterial species or serum measurements of inflammatory response, signs of foetal inflammatory response could not be used in this study. Hence, bacteraemia was studied instead of sepsis. In the definite bacteraemia group, all positive blood cultures were included irrespectively of what bacterial species the culture yielded. This means that a number of patients in this group did not have sepsis, they only had a contaminated blood culture. Furthermore, all patients with a false-negative blood culture who actually had sepsis were included in the group presumed bacteraemia. Unfortunately these systematic misclassifications are shared with several other large studies. The definition of the group ‘presumed bacteraemia’ is even more troublesome, since all infants with antibiotic treatment for more than 3 days were included in this group. To define a diagnosis merely on the treatment that the patient receives is indeed doubtful, as treatment strategies vary between centres and individual doctors. This weakness becomes especially evident when statistical comparisons with factors preceding the diagnosis are made. How can we, for example, know if the increased risk for early presumed bacteraemia associated with vaginitis is because of illness in the infants and not because of a liberal use of antibiotics when the physician knows that the mother had vaginitis? In neonatal intensive care, antibiotics are normally used for several other indications than bacteraemia, but in this study all infants on antibiotics are classified as presumed Invited Commentary for Patel et al. Presumed and definite bacteremia in extremely low gestational age newborns, pages 36–41. Acta Pædiatrica ISSN 0803–5253

Scrubbing the hub of intravenous catheters with an alcohol wipe for 15 sec reduced neonatal sepsis

The aim of this study was to investigate whether scrubbing the hub of intravenous catheters with an alcohol wipe for 15 sec could reduce the incidence of neonatal sepsis in a level‐three neonatal intensive care unit.

Rapid typing of neonatal Staphylococcus epidermidis isolates using polymerase chain reaction for repeat regions in surface protein genes.

Staphylococcus epidermidis is a significant pathogen in neonatal sepsis and other nosocomial infections. For further investigations of the colonisation patterns and invasive pathways, typing methods that are applicable on large populations of bacterial isolates are warranted. In the present study, a genotyping method based on polymerase chain reaction (PCR) for the repeat regions of four genes (sdrG, sdrF, aap and sesE) that encode for bacterial surface proteins was developed and applied to a sample of well-characterised neonatal blood isolates of S. epidermidis (n = 49). The PCR products were visualised on agarose gel (sdrG, sdrF and sesE) or by fragment analysis (aap). The discriminatory index (D-index) for genotyping of the different genes was compared to genotyping by pulsed-field gel electrophoresis (PFGE). The highest D-index for the PCR-based typing methods was found for the combination of sdrF, sdrG and aap (D-index 0.94), whereas the optimal two-gene combination (sdrF and aap) resulted in a D-index of 0.92. We conclude that the described method can be used for the genotyping of large populations of S. epidermidis isolates with a sufficient discriminatory capacity, and we suggest that the combination of sdrF and aap is the most suitable to use.