Time to Positivity of Coagulase Negative Staphylococcus in Neonatal Blood Cultures as a Useful Tool to Discriminate Between Sepsis and Contamination

doi:10.21203/rs.3.rs-4701584/v1

Download PDF

Article

Time to Positivity of Coagulase Negative Staphylococcus in Neonatal Blood Cultures as a Useful Tool to Discriminate Between Sepsis and Contamination

https://doi.org/10.21203/rs.3.rs-4701584/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Objective: to assess the usefulness of time to positivity (TTP) to distinguish between sepsis and contamination in coagulase-negative staphylococci (CoNS) isolates.

Study design: unicentric retrospective observational. 168 patients with suspected sepsis and positive blood culture for CoNS were recruited. Patients were subdivided into sepsis (29%) and probable contamination (71%). Logistic regression analyses were performed to evaluate different risk factors and clinical signs and symptoms associated with sepsis.

Results: TTP cut-off value that best discriminated sepsis from contamination was found to be 18 hours. Regression analysis revealed that TTP ≤18h, gestational age ≤32 weeks, taquicardia/bradicardia and hypoactivity/lethargy were independent predictors of sepsis.

Conclusion: usefulness of TTP to distinguish sepsis from contamination in isolation of CoNS, especially in neonates with other associated risk factors (lower gestational age or being a CVC carrier). The clinical signs that most increase the discriminatory power of TTP are the presence of tachycardia/bradycardia or hypoactivity.

Health sciences/Diseases/Infectious diseases/Bacterial infection

Health sciences/Biomarkers/Diagnostic markers

Neonatal sepsis is a clinical challenge in neonatal intensive care units (NICU). The unspecificity of clinical signs, especially in very preterm newborn infants, and the risks associated with delaying the initiation of treatment are the main reasons for the overuse of empirical antibiotherapy in neonatal units¹. Even after a negative late onset sepsis evaluation, the median antibiotic course length is around 4 days.² This overuse is associated with increased neonatal morbidities, mortality, or significant neurodevelopmental impairment, but also with higher risk allergic diseases, obesity, and inflammatory bowel disease in later life³.

Reducing the duration of empirical treatment should be a priority in neonatal units. Accurate diagnosis is essential for the early discontinuation of antibiotics. The gold standard for sepsis diagnosis requires isolation of a microorganism in blood (or sterile body fluid). However, when coagulase-negative staphylococci (CoNS) are isolated, which is the main causative agent of late-onset sepsis in very low birthweight infants, distinguishing infection from contamination is challenging⁴. CoNS, particularly Staphylococcus epidermidis, are among the most numerous commensal bacteria on the skin and mucous membranes. The Centers for Disease Prevention and Control clinical criteria for true positives require the detection of CoNS from two or more blood cultures within 48 hours and accompanying clinical signs in the patient^5,6. This recommendation is far from the usual practice in neonatal units, since the fragility of the patients makes it necessary to minimize both the volume of the extractions and the painful procedures.

Widespread use of automated blood culture detection systems has shortened the time in which most pathogens are detected, and time to positivity (TTP) has been reported useful to shorten and reduce empirical antibiotic treatment days⁷. In this sense, and giving value to blood cultures (BCs), we wonder if the TTP is useful to distinguish between sepsis and contamination in CoNS isolations.

Our objectives are to assess the usefulness of the TTP of bacterial growth in blood culture to distinguish between sepsis and contamination in CoNS isolates and to analyze which associated clinical variables can increase its accuracy.

2.1. Patients and specimens

This unicentric retrospective observational study included 168 cases with suspected sepsis and positive blood culture for CoNS, between the years 2018 and 2023. Multiple bacterial species other than CoNS had the potential for growth that affected TTP. To avoid this effect, 16 patients with bacteremia caused by multiple bacterial species were excluded. We collected data of 152 non-consecutive patients admitted to the Neonatal Unit (87 males and 65 females; median age 12 days; range, 0–80 days) from electronic medical records.

Positive BCs were scrutinized and subdivided into sepsis (group 1) and probable contamination (group 2), based on analytical criteria that included C-reactive protein (CRP), leukocytes, neutrophils, and platelet values; the information related to the TPP was blinded for classification purposes.

All patients included in the study met one or more clinical criteria for suspected sepsis (temperature instability (> 38ºC/<36.5ºC), heart rate > 200 bpm or < 80 bpm, apneas, slow capillary refill, increased respiratory support or FiO2, hyperglycemia, digestive intolerance, lethargy/hypoactivity). For classification in the sepsis group, infants also had to meet one or more analytical criteria (leukopenia < 4000/mm3, neutropenia < 1500/mm3, thrombocytopenia < 100000/mm3, CRP > 20 mg/L), according to the NeoKiss definitions of CNS-related sepsis⁸.

Medical history and laboratory data were retrospectively reviewed. The current study was approved by the Research Ethics Committee of Hospital Clinico Universitario of Valencia, Spain (14/03/2024 ref 2024/069).

2.2 Blood culture processing

A total of 152 neonatal blood samples were collected upon clinical signs of sepsis. Blood samples were collected in a pediatric blood culture bottle (BacT/ALERT ®, Biomérieux Inc., Craponne, France) and incubated in an automated continuous monitoring blood culturing instrument (BACTEC FX; Becton Dickinson, S.A. ). Recommended blood volume collected for culture, at least 1 ml⁹, was checked in each sample by weighing using a calibrated balance, before incubation. They were incubated for up to 5 days and classified as positive if bacterial growth during this time. The time to positivity (TTP) was defined as the time from the start of incubation to a positive signal in automated BC systems. When a bottle tests positive for any organism, information is transmitted to the electronic medical record so that clinicians are informed in real time. Positive BCs were processed for direct MALDI-TOF MS identification as follows: 1 mL of positive BCs was centrifuged (13,000 rpm, 3 min), then the pellet was washed with sterile distilled water, vortexed and further centrifuged (13,000 rpm, 3 min). The pellet was smeared onto a MALDI Biotyper (Bruker Biotyper, Bruker Daltonics, Billerica, MA, USA) plate, left to dry at room temperature, followed by the addition of one microliter of matrix solution (α-Cyano-4-hydroxycinnamic acid-HCCA). After drying the target plate was placed in the MALDI Biotyper and analyzed. Extraction of proteins with formic acid was skipped. Positive BCs were processed 24 h/7 days a week. In parallel, positive BCs were subcultured on chocolate agar, incubated 24–48 h and isolated colonies were identified by MALDI-TOF MS. Criteria for successful identification were met when the spectral score of at least one out of the three spots was ≥ 2.0 (species level) and ≥ 1.7 (genus level), as recommended by the manufacturer.Antimicrobial susceptibility testing was performed by microdilution using appropriate MicroScan panels (Beckman Coulter; Brea, CA, USA).

2.3 Laboratory measurements

Clinical laboratory tests included leukocytes, neutrophils, platelets values and serum levels of CRP.

2.4 Statistical methods

Quantitative data were shown as the median with interquartile range (IQR) and differences between medians were compared using the Mann-Whitney U-test. Categorical variables were analyzed using the Chi-square statistic. The quantitative variable of TTP (hours) of bacterial growth in blood culture was transformed into a categorical variable for the logistic regression analysis, establishing the cut-off point with a better combination of sensitivity and specificity using a ROC curve.

Logistic regression analyses were performed to evaluate different risk factors associated with sepsis. Variables with P values < 0.1 in univariate models were included in multivariate models. Two-sided exact P-values were reported. A P-value < 0.05 was considered statistically significant. The analyses were performed using SPSS version 20.0 (SPSS, Chicago, IL, USA).

3.1 Characteristics of Study Population

All 152 patients in this cohort had suspected sepsis and 143 (94%) required hospitalization in the NICU. Of the positive cultures, 108 (71%) were classified as contaminants. Group 1 (confirmed sepsis) included 44 patients (25 males and 19 females; median age 12 days; range, 4–80 days) and group 2 (contaminant cultures) included 108 patients (62 males and 46 females; median age 11.5 days; range, 0–59 days). Both groups were comparable regarding sex, chronological age and antibiotic therapy prior to the episode (P = 0.91, P = 0.3 and P = 0.6, respectively). We refer to a previous course of antibiotic if it was administered within the last 72 hours before sample collection and CoNS was susceptible to the empirically administered antibiotic. In contrast, birth weight (BW) and gestational age (GA) were significantly lower in the sepsis group (P < 0.001) (Table 1).

Table 1

Characteristics of Infants n = 152
	Sepsis (n = 44)	Contamination (n = 108)	P-value
Chronological age (IQR), days	12 (8–18)	11,5 (7–18)	0,44
Sex (Male/female)	25/19	62/46	0,94
Previous antibiotic therapy*	10/34	22/86	0,76
Median birth weight (IQR), g	1090 (890–1361)	2100 (908–3355)	< 0,001
Gestational age (IQR), weeks	30 (27–31)	35 (28–39)	< 0,001
* within 72 h prior to sampling

3.2 Time to positivity (TPP) of blood cultures/TTP comparison

We compared the TTP of contaminant cultures to that of pathogens. Contaminants had longer TTP (median 20.20 hours; IQR, 18.20–23.10) than sepsis group (median 18.65 hours; IQR, 16.15-22-25) P < 0.02; (Fig. 1).

ROC curve analyses were next performed to determine optimal TTP cut-off values that best discriminated between patients who had sepsis from those whose cultures were classified as contaminants; optimal threshold was found to be 18 hours displaying a sensitivity of 77.8% and a specificity of 43.2%.

3.3 Risk factors associated with sepsis

To examine whether TTP was a good predictor of whether a positive CoNS blood culture can be considered a real bloodstream infection we performed binary logistic regression analysis with different classical risk factors and clinical signs associated with sepsis (Table 2).

Table 2

Logistic regression analysis of risk factors and clinical signs of sepsis
	UNIVARIATE		MULTIVARIATE
Variable	OR (IC 95%)	P-value	OR (IC 95%)	P-value
Time to positivity (TTP): ≤ 18h vs > 18h	2.44 (1.12–5.35)	0.02	4.61(1.33–15.90)	0.015
Birth weight: ≤ 1500g vs > 1500g	7.01 (2.86–17.13)	< 0.001
Gestational age: ≤ 32 vs > 32 weeks	7.86 (3.21–19.26)	< 0.001	10.2 (2.88–36.03)	< 0.001
Central venous catheter (CVC)	15.11 (5.04–45.30)	< 0.001
Peripheral venous catheter (PVC)	0.98 (0.37–2.57)	0.98
Parenteral nutrition	7.35 (3.19–16.94)	< 0.001
Temperature instability	1.61 (0.76–3.41)	0.21
Tachycardia or bradycardia	20.4 (8.02–51.96)	< 0.001	10.87 (3.82–30.94)	< 0.001
Poor Peripheral Perfusion	2.94 (1.29–6.67)	0.01
Apnea	12.28 (5.34–28.25)	< 0.001
Increase in respiratory support	6.69 (3.07–14.58)	< 0.001
Hyperglycemia	4.33 (1.43–13.03)	0.009
Feeding Intolerance	2.85 (1.09–7.45)	0.03
Hypoactivity/Lethargy	5.41 (2.48–11.79)	< 0.001	2.72 (0.93–7.89)	0.06

Univariate analysis revealed that TTP ≤ 18 hours, neonatal weight ≤ 1500 g, gestational age ≤ 32 weeks, central venous catheter carriers, parenteral nutrition, tachycardia/bradycardia, peripheral perfusion impairment, apnea, respiratory support increase, hyperglycemia, feeding intolerance and hypoactivity/lethargy were associated with sepsis.

Multivariate regression analysis revealed that TTP ≤ 18h (OR 4.61; 95% CI 1.33–15.90; P = 0.01), gestational age ≤ 32 weeks (OR 10.20; 95% CI 2.88–36.03; P < 0.001), taquicardia/bradicardia (OR 10.87; 95% CI 3.82–30.94; P < 0.001) and hypoactivity or lethargy (OR 2.72; 95% CI 0.93–7.89; P = 0.06), were independent predictors of sepsis.

The performance of the final model was assessed using a confusion matrix, yielding a positive predictive value (PPV) of 77.3% and a negative predictive value (NPV) of 86.8%. The sensitivity and specificity of the model were 70.8% and 90.2%, respectively.

Establishing the diagnosis of a CoNS sepsis is challenging, due to the difficulty in

discriminating against culture contamination from true infection⁴. Furthermore, CoNS-related sepsis accounts for 50–60% of late-onset sepsis in most NICUs and is a major health care and economic burden¹⁰. However, there is no diagnostic consensus and antibiotics are often maintained unnecessarily.

Our objective was to determine the usefulness of TTP in CoNS-positive BCs to distinguish between sepsis and contamination. Although prior studies have evaluated TTP in the pediatric population, most have included evaluations for both early-onset sepsis and late-onset sepsis, and have included all microorganisms^7,11,12. To date, this is the first study concerning TTP in CoNS-related late-onset neonatal sepsis.

In our study cohort and considering the NeoKiss diagnostic criteria⁸, only 29% of our cohort should be considered as true sepsis. When using the NeoKiss analytical criteria to classify patients as sepsis due to CoNS, it is unlikely that in the sepsis group we are faced with any contamination, which could justify the relatively short TTP. In previous studies, referring to a longer TTP, classification criteria include the neonatologist's decision to continue antibiotic treatment for at least 5 days^7,13. The interpretation of what constitutes sepsis by one clinician may differ from another, which necessarily introduces a degree of uncertainty and bias in classification groups.

The sample is representative of the type of patients admitted to the NICU: it includes all gestational ages and birth weights in patients in a tertiary hospital with a level IIIb NICU. We found, in line with previous literature, that patients with lower gestational age and lower birth weight were more likely to present sepsis due to CoNS¹⁴. In this group of infants, the recommendation to draw paired blood cultures⁵ becomes more complicated, so having more parameters to help us differentiate between sepsis and contamination is crucial. We explored TTP in both groups and statistically significantly found differences between both groups, with lower TTP in those patients classified as sepsis. In the ROC curve, the cut-off that best discriminated one group from another was 18 hours, a cut-off that we subsequently used for logistic regression analyses.

All risk factors and clinical signs or symptoms analyzed were predictors of sepsis in the univariate analysis, except for being a carrier of PVC or having temperature instability. The fact that being a PVC carrier was not significant in our sample should be interpreted with caution, since the number of patients who were PVC carriers was very low. In the multivariate analysis, independent predictors of sepsis risk were TTP < 18 h, GA less than or equal to 32 GWA, presenting tachycardia/bradycardia or hyporeactivity/lethargy as a symptom, coinciding with previous studies on the clinical picture associated with CoNS infection⁴. In contrast to what Freeman et al demonstrated in a case cohort study, in which 56.6% of all the cases of nosocomial bacteremia could be attributed to lipid administration, parenteral nutrition was not a predictor of sepsis in our cohort¹⁵.

The sensitivity and specificity of TTP alone is low (as expected) but improves considerably in combination with risk factors and clinical signs. As far as we know, this is the first study examining all these variables together to achieve most accuracy in diagnosis of CoNS nosocomial sepsis, facilitating the difficult decision to discontinue antibiotics in newborns.

The most important determinant of the validity of a blood culture is that the sample volume is > 1 mL¹⁶. It is important to highlight that in our microbiology laboratory the volume of the sample is checked before being processed, which allows us to guarantee an optimal performance of the blood culture, unlike other studies where this data is uncertain^7,17. We also believe that this may have contributed to the fact that TTP values were lower than those previously published.

Another strength of the study is that the time from incubation until the pathogen begins to grow has been used to assess the TTP in all isolates. The pre-analytical time (time from when the blood culture is extracted until it is introduced into the system) in our sample is homogeneous and short in all cases (less than 1 hour), since we have an onsite microbiology laboratory operating 24 hours a day. Guidelines recommend inoculation of a blood culture between 2 and 4 hours¹⁸. The fact that our TTP is defined as the time from incubation to growth (as well as the strict sepsis classification criteria) could also justify lower TTPs compared to other series¹¹ where TTP is referred to from when blood is inoculated into the blood culture bottle. Studies defining TTP as incubation-to-growth report a shorter median TTP, as illustrated in the studies by Garcia-Prats¹⁹ and Jardine²⁰. According to our results, if the inoculum volume is adequate, we could establish lower TTP values in CoNS than those currently considered in clinical practice. The likely pathogenicity of organisms growing in BC depending on the “time to positivity” (TTP) is well established in previous studies²¹ and could lead to more accurate and timely clinical decision-making, potentially improving patient outcomes and optimizing antimicrobial therapy.

The limitations of the study include the small sample size and the retrospective nature of the study. Nevertheless, we consider that the results obtained represent a line of work that merits further study.

The results of this study support the usefulness of TTP to distinguish sepsis from contamination in isolation of CoNS, especially in those neonates with other known associated risk factors, such as gestational age (< 32 WGA) or being a CVC carrier. The clinical signs that most increase the discriminatory power of TTP to distinguish sepsis from contamination are the presence of tachycardia, bradycardia or hypoactivity. Considering clinical judgment, risk factors and clinical signs, a TTP above 24 hours can assist physicians in the decision to discontinue antibiotherapy, reducing antibiotic exposure and its consequences in newborn infants with a positive BCs for CoNS.

Declaration of conflict of interest: the authors have no conflicts of interest to disclose.

Funding: the authors did not receive any external funding relevant to this study.

Data availability: The data that support the findings of this study are available from the corresponding author, SC-S, upon reasonable request.

Shane AL, Sánchez PJ, Stoll BJ. Neonatal sepsis. The Lancet. 2017;390(10104):1770–1780.
Speier RL, Cotten CM, Benjamin DK, et al. Late-Onset Sepsis Evaluation and Empiric Therapy in Extremely Low Gestational Age Newborns. J Pediatric Infect Dis Soc. 2023;12:S37–S43.
Cantey JB, Pyle AK, Wozniak PS, et al. Early Antibiotic Exposure and Adverse Outcomes in Preterm, Very Low Birth Weight Infants. Journal of Pediatrics. 2018;203:62–67.
Healy CM, Baker CJ, Palazzi DL, et al. Distinguishing true coagulase-negative Staphylococcus infections from contaminants in the neonatal intensive care unit. Journal of Perinatology. 2013;33(1):52–58.
Struthers S, Underhill H, Albersheim S, et al. A comparison of two versus one blood culture in the diagnosis and treatment of coagulase-negative staphylococcus in the neonatal intensive care unit. Journal of Perinatology. 2002;22(7):547–549.
López Sastre JB, Coto Cotallo D, Fernández Colomer B, et al. Neonatal sepsis of nosocomial origin: an epidemiological study from the “Grupo de Hospitales Castrillo”. J Perinat Med. 2002;30(2):149–157.
Mukhopadhyay S, Briker SM, Flannery DD, et al. Time to positivity of blood cultures in neonatal late-onset bacteraemia. Arch Dis Child Fetal Neonatal Ed. 2022;107(6):583–588.
Gastmeier P, Geffers C, Schwab F, et al. Development of a surveillance system for nosocomial infections: The component for neonatal intensive care units in Germany. Journal of Hospital Infection. 2004;57(2):126–131.
Cantey JB, Baird SD. Ending the Culture of Culture-Negative Sepsis in the Neonatal ICU. Pediatrics. 2017;140(4).
Fernandez Colomer B, Cernada Badia M, Coto Cotallo D, et al. The Spanish National Network Grupo Castrillo: 22 Years of Nationwide Neonatal Infection Surveillance. Am J Perinatol. 2020;37(4):S71–S75.
Kuzniewicz MW, Mukhopadhyay S, Li S, et al. Time to Positivity of Neonatal Blood Cultures for Early-onset Sepsis. Pediatric Infectious Disease Journal. 2020;39(7):634–640.
Arias-Felipe A, Ramírez-Berrios J, Recio-Martinez R, et al. Determining Time to Positivity of Blood Cultures in a Neonatal Unit. J Pediatric Infect Dis Soc. 2022;11(11):510–513.
Craft A, Finer N. Nosocomial coagulase negative staphylococcal (CoNS) catheter-related sepsis in preterm infants: definition, diagnosis, prophylaxis, and prevention. J Perinatol. 2001;21(3):186–192.
Flannery DD, Edwards EM, Coggins SA, et al. Late-Onset Sepsis Among Very Preterm Infants. Pediatrics. 2022;150(6).
Freeman J, Goldmann DA, Smith NE, et al. Association of Intravenous Lipid Emulsion and Coagulase-Negative Staphylococcal Bacteremia in Neonatal Intensive Care Units. New England Journal of Medicine. 1990;323(5):301–308.
Schelonka RL, Chai MK, Yoder BA, et al. Volume of blood required to detect common neonatal pathogens. J Pediatr. 1996;129(2):275–278.
Connell TG, Rele M, Cowley D, et al. How reliable is a negative blood culture result? Volume of blood submitted for culture in routine practice in a children’s hospital. Pediatrics. 2007;119(5):891–896.
Miller JM, Binnicker MJ, Campbell S, et al. A Guide to Utilization of the Microbiology Laboratory for Diagnosis of Infectious Diseases: 2018 Update by the Infectious Diseases Society of America and the American Society for Microbiology. Clinical Infectious Diseases. 2018;67(6):e1–e94.
Garcia-Prats JA, Cooper TR, Schneider VF, et al. Rapid Detection of Microorganisms in Blood Cultures of Newborn Infants Utilizing an Automated Blood Culture System. Pediatrics. 2000;105(3):523–527.
Jardine L, Davies MW, Faoagali J. Incubation time required for neonatal blood cultures to become positive. J Paediatr Child Health. 2006;42(12):797–802.
Huggard D, Powell J, Kirkham C, et al. Time to positivity (TTP) of neonatal blood cultures: a trend analysis over a decade from Ireland. Journal of Maternal-Fetal and Neonatal Medicine. 2021;34(5):780–786.

There is NO conflict of interest to disclose.

Download PDF

Editorial decision: revise
08 Aug, 2024
Review #2 received at journal
06 Aug, 2024
Review #1 received at journal
29 Jul, 2024
Reviewer #2 agreed at journal
23 Jul, 2024
Reviewer #1 agreed at journal
22 Jul, 2024
Reviewers invited by journal
16 Jul, 2024
Submission checks completed at journal
15 Jul, 2024
First submitted to journal
12 Jul, 2024
Unknown event
08 Jul, 2024
Editor assigned by journal
07 Jul, 2024

You are reading this latest preprint version

Time to Positivity of Coagulase Negative Staphylococcus in Neonatal Blood Cultures as a Useful Tool to Discriminate Between Sepsis and Contamination

Status:

Version 1

Abstract

Figures

1. INTRODUCTION

2. MATERIALS AND METHODS

2.1. Patients and specimens

2.2 Blood culture processing

2.3 Laboratory measurements

2.4 Statistical methods

3. RESULTS

3.1 Characteristics of Study Population

3.2 Time to positivity (TPP) of blood cultures/TTP comparison

3.3 Risk factors associated with sepsis

4. DISCUSSION

5. CONCLUSIONS

Declarations

References

Additional Declarations

Status:

Version 1