Journal of Clinical Neonatology

ORIGINAL ARTICLE
Year
: 2022  |  Volume : 11  |  Issue : 2  |  Page : 71--78

Bacterial etiology and antibiotic sensitivity patterns in late-onset neonatal blood infection: A 6-year retrospective study


Ihab Hussein Elkadry, Chokkiyil Hafis Ibrahim Ponnambath 
 Department of Neonatology, Corniche Hospital, Abu-Dhabi, United Arab Emirates

Correspondence Address:
Ihab Hussein Elkadry
Department of Neonatology, Corniche Hospital, P O Box 3788, Abu-Dhabi
United Arab Emirates

Abstract

Introduction: The incidence and etiology of neonatal bloodstream infections vary globally. Early appropriate antibiotic therapy is crucial. An empiric antibiotic choice should be driven by accurate knowledge of the local spectrum of pathogens and susceptibilities. Methodology: A retrospective observational study was conducted on neonates born at local tertiary center from January 1, 2013, to December 31, 2018, with late-onset bloodstream infection (LBSI). Trends of causative organisms and antibiotic susceptibilities were analyzed. Results: A total of 696 LBSI occurred in 469 neonates. Overall incidence over the 6 years was 122/1000 admissions. The median time to LBSI was 13 days of life. Majority of infections occurred in infants <32 weeks. About 75.9% were caused by Gram-positive and the rest by Gram-negative bacteria. The most common organism was coagulase-negative staphylococcus (CoNS) which showed an increase in resistance to amikacin over time, but with stable sensitivity patterns to teicoplanin. Klebsiella and Escherichia coli were the most common Gram-negative organisms. There was improving sensitivity to cephalosporin in Klebsiella species. Sixteen percent of Gram-negative isolates were extended spectrum beta-lactamase (ESBL) producing. Majority of the Gram-negative bacteria including ESBL-producing strains remained sensitive to amikacin. An empiric antibiotic combination of teicoplanin and amikacin was appropriate to cover the majority of LBSIs. Conclusions: The majority of late-onset neonatal bloodstream infections in this study cohort were caused by Gram-positive organisms of which CoNS was the most common. The empiric antibiotic choices for LBSI on our unit seem appropriate based on the data. In units where the organism and susceptibility patterns are similar, the same antibiotic choices may be justified.



How to cite this article:
Elkadry IH, Ibrahim Ponnambath CH. Bacterial etiology and antibiotic sensitivity patterns in late-onset neonatal blood infection: A 6-year retrospective study.J Clin Neonatol 2022;11:71-78


How to cite this URL:
Elkadry IH, Ibrahim Ponnambath CH. Bacterial etiology and antibiotic sensitivity patterns in late-onset neonatal blood infection: A 6-year retrospective study. J Clin Neonatol [serial online] 2022 [cited 2022 Jul 2 ];11:71-78
Available from: https://www.jcnonweb.com/text.asp?2022/11/2/71/343422


Full Text



 Introduction



Neonatal sepsis is one of the leading causes of morbidity and mortality in the neonatal period worldwide.[1] The incidence and etiology of neonatal bloodstream infections vary widely amongst different regions of the world with incidence ranging from 1–170/1000 live births.[2]

Neonatal bloodstream infections can be either early-onset sepsis (EOS) or late-onset sepsis (LOS). EOS is defined as occurring within the first 72 h after birth and LOS after that period. Seventy-two hours after birth is considered a cut-off time point adequate to differentiate LOS from EOS in terms of the spectrum of causative pathogens. Early-onset infections are mostly acquired perinatally, whereas the majority of late-onset infections are nosocomial. In the last few decades, there has been a general reduction in EOS, attributed to improvements in perinatal care and the use of prophylactic antibiotics to prevent infections caused by Group B Streptococcus. Meanwhile, LOS has increased alongside improved neonatal care and survival, especially for premature infants.[3],[4]

The etiology of EOS and LOS in neonates are different in terms of the most common bacteria encountered. In most middle- and high-income countries, the most prevalent organisms in the neonatal intensive care unit (NICU) are coagulase-negative Staphylococcus followed by Gram-negative organisms, whereas in the low-income countries, it is a mixture of Staphylococcus aureus and Gram-negatives.[5],[6],[7],[8] A recent 2-year multicenter study from the middle east countries revealed a distribution of pathogens very similar to that published from the west.[9]

Antibiotics are the most commonly used drugs in the NICU.[10] Antibiotic treatment is initiated immediately among infants with suspected infection based on symptoms and risk factors. The correct choice of empiric antibiotic therapy is important to improve the outcomes in neonatal sepsis. Empiric therapy is based on the knowledge of the most common organisms causing neonatal sepsis locally and their sensitivity patterns. Across the world resistance to antibiotics is on the rise-more so in the Asian subcontinent.[5],[6],[7],[8],[11]

There have been only a few peer-reviewed published studies from the middle east and gulf region in the last decade about the detailed antibiotic sensitivity patterns of bacteria causing early- and late-onset bacterial infections.[9],[10],[12],[13] It is important to review the available data periodically to ascertain that the choice of empiric antibiotic therapy is valid based on the most recently available data on the distribution of organisms and their sensitivities. In this article, we present the data on the late-onset infections in newborn infants cared for at our centre-Corniche Hospital, a large tertiary maternal and neonatal care center based in Abu Dhabi, United Arab Emirates (UAE). It hosts the largest neonatal service in the UAE with nearly 6000 deliveries per annum and 900 admissions to the neonatal unit. The choice of empirical antibiotic therapy on the neonatal unit has been guided by annual review of the local antibiogram. Currently, the first-line antibiotics for LOS are teicoplanin and amikacin, with therapy being modified according to final organism identification and sensitivity reports. The empiric antibiotic therapy at our hospital has remained the same for the last 8 years. This study aims to analyze the trends in etiology and antimicrobial sensitivity in culture-proven bloodstream infections at Corniche Hospital over 6 years to assess if the empiric therapy used on the unit is still effective and safe. Also publishing this data will be important for the wider region as the distribution of pathogens are likely to be similar.

 Methodology



A retrospective observational study was conducted on all neonates who were born at the Corniche Hospital or admitted to the NICU from January 1, 2013, to December 31, 2018, with a culture-positive LBSI. The blood culture data were extracted from the electronic database maintained by the microbiology department. Clinical and demographic data were collected from the electronic medical records.

A bloodstream infection with a pathogen which was not a skin commensal was considered a true infection. The most common skin commensal grown in blood cultures on NICUs are coagulase-negative Staphylococcus (CoNS). It is not routine to do paired cultures or repeat cultures during an episode of infection on our unit as defined in some of the international recommendations. For the purposes of the study, a positive blood culture with CoNS was considered a true infection if the infant was symptomatic and associated with C-reactive protein >10 before or within 48 h of obtaining the blood culture, or the infant received antibiotics for at least 5 days based on clinical assessment. Delayed mixed growth beyond 48 h of incubation and other skin commensals such as diphtheroids and micrococci were considered contaminants. If the same organism was cultured with the same sensitivity pattern during a single episode of sepsis, it was considered a duplicate and filtered.

Blood culture and antibiotic sensitivity testing

Blood cultures were processed using the BD Bactec® system and the antibiotics sensitivity testing was conducted on the VITEK®-2 platform. For the purposes of the study, bacteria showing intermediate sensitivity to a particular antibiotic was considered resistant.

Data were entered into a standard database and analyzed using Microsoft Excel®. The results are presented as summary statistics.

The study was approved by the institutional research ethics committee. The requirement for consent was waived as only minimal identifiable data which was enough to verify data accuracy was collected and there was no direct patient involvement or intervention.

 Results



A total of 696 confirmed late-onset bloodstream infections (LBSI) occurred in 469 newborn infants admitted to our NICU between January 1, 2013, and December 31, 2018 (some infants had more than one episode of LBSI). The overall incidence of infection over the 6-year period was 122 per 1000 admissions to NICU. There was decreasing trend over the course of the 6 years [Figure 1]. The median time of onset of the first episode of late-onset infection was 13 days of life (interquartile range 7–21 days). Majority of infections occurred in preterm infants (<37 weeks) with the major bulk in infants <32 weeks [Figure 2].{Figure 1}{Figure 2}

The distribution of organisms causing LBSI is as in [Table 1]. Of all the organisms, 528 (75.9%) were Gram-positive and 168 (24.1%) Gram-negative. The most common organism group consisted of coagulase-negative Staphylococcal species (CoNS), which contributed to 62% of all infections. Majority of Gram-negative infections were caused by Klebsiella sp. followed by Escherichia coli. Eighty-seven (12.5%) of the organisms were considered multidrug-resistant Staphylococcus aureus (MRSA and extended-spectrum beta-lactamase [ESBL]). The annual trends of causative organisms from 2013 to 18 are summarized in [Figure 3]. The absolute number of infections has fallen year on year due to improved infection control practices on the unit with CoNS infections nearly halving in number between 2013 and 2018. However, CoNS continued to be the major causative organism in late-onset bacterial bloodstream infections.{Figure 3}{Table 1}

Antibiotic sensitivity trends late-onset bloodstream infections

The antibiotic sensitivity over the 6-year study period of bacteria causing LBSI is summarised as cumulative sensitivities. The 6-year trends in resistance to the commonly used antibiotics on NICU for individual organisms causing >90% of the infections are summarized separately.

Antibiotic sensitivity trends for Gram-positive late-onset bloodstream infections

The cumulative antibiotic sensitivity of all Gram-positive organisms during the 6-year study period is summarized in [Table 2]. The cumulative sensitivities to individual antibiotics were included only if at least 50% of the isolates were tested.{Table 2}

The yearly trends in resistance of the bacteria causing the majority of the infections (CoNS, S. aureus, and Enterococcus sp.) to commonly used antibiotics in newborns is summarised in the following graph [Figure 4].{Figure 4}

Apart from a slight increasing trend in amikacin resistance among CoNS infections, there were no other discernible trends. There seemed to have been a spike in bloodstream infections caused by MRSA between 2015 and 2017 which seemed to have settled down to zero in 2018. Overall 26% (11/42) of the Staphylococcal isolates were methicillin resistant.

Antibiotic sensitivity trends for Gram-negative late-onset bloodstream infections

The cumulative antibiotic sensitivity of all Gram-negative organisms during the 6-year study period is summarized in [Table 3]. As with Gram-positive organisms, the sensitivities to antibiotics were included only if at least 50% of the isolates were tested.{Table 3}

The yearly trends in resistance of the bacteria causing majority of the gram-negative infections (Klebsiella, E. coli, Enterobacter, Serratia) to commonly used antibiotics targeting Gram-negative organisms in newborns are summarized in the following graph [Figure 5].{Figure 5}

There was decreasing a trend in resistance to third (CAZ, CTX) and fourth-generation cephalosporins (FEP) among the Klebsiella species. E. Coli and Enterobacter isolates showed an increasing trend in resistance to fluoroquinolones (CIP). Twenty-seven (16.1%) of all Gram-negative isolated were ESBL producers. If considering only the Enterobacteriaceae the proportion was 19% (27/143). All of them were either E. coli (43% of all isolates) or Klebsiella Sp.(17%). The yearly trends of ESBL-positive isolates are shown in [Figure 6]. The ESBL rate among the Klebsiella species seemed to have decreasing trend over time whereas it increased for the E. coli isolates after a dip in the 2nd year of the study. Of all the ESBL isolates 88.5% were sensitive to amikacin (Klebsiella-100%, E. coli 81%).{Figure 6}

No other discernible trends were noted among the Gram-negative bacteria. Carbapenem resistance among the Gram-negative bacteria was low (3%) and only seen in a proportion (17%) of the Pseudomonas isolates.

 Discussion



We have described the trends in bacterial pathogens causing neonatal LBSIs and their sensitivity patterns on our local NICU. Similar to studies published from other centers, the distribution of organisms is similar with respect to the fact that CoNS and Enterobacteriaceae cause the majority of the infections.[4],[5],[6],[7],[8],[9],[11],[12],[14] However unlike reports from the west were E. coli seemed to be the predominant Gram-negative organism, the data from our center are similar to the data from the wider Gulf region, Indian subcontinent, and East Asia, where Klebsiella sp. seem to be the predominant Gram-negative organism.[4],[5],[9],[11],[13],[14],[15],[16] This pattern is quite different from middle and lower-income countries where Gram-negative bacteria seem to be predominant in the LOS isolates.[17],[18],[19] One distinct difference between our data and that from other gulf countries is the higher proportion of CoNS infection (62% v/s 35%).[9] This may partly be explained by the definitions we used to consider CoNS as a true pathogen which may have overestimated the incidence of CoNS. However, the definitions we used are not markedly different from that used in other published studies and hence, the overestimation probably had only a small effect on the difference seen.

One of the primary aims of this study was to assess the appropriateness of the empiric antibiotics used as a first line–teicoplanin and amikacin. Over the 6-year period 98% of the Gram-positive isolates where sensitive to teicoplanin with no significant adverse trends. Two percent of the CoNS had intermediate sensitivity to the antibiotic. Teicoplanin is also justified over conventionally used beta-lactam antibiotics like flucloxacillin because the sensitivity of CoNS is low to beta-lactams and up to 26% of the S. aureus isolates were methicillin resistant. Ninety-two percent of the Gram-negative isolates were sensitive to amikacin overall with very little resistance seen in the recent years with 100% of the isolates tested in the last year of the study being sensitive. Because of the historical resistance of some of the isolates to amikacin, the practice on our unit is to add in either a carbapenem (meropenem) or piperacillin-tazobactam combination when we get an initial Gram-negative report is received and de-escalate if appropriate when the final results become available. However, the recent trend suggests that continuing empiric therapy with amikacin, pending final sensitivity reporting, might be safe unless meningitis is strongly suspected.

Increasing antibiotic resistance is an established and alarming global healthcare threat.[20] It has been a problem in the middle east as well.[9],[13] The resistance patterns in the gram-positive organisms have been stable in our population over the course of the study period. Invasive infections from MRSA contributed to 26% of the Staphylococcal infections-but the trend has remained stable with no notable increase in recent years. We have not encountered any vancomycin resistance enterococcus amongst Enterococci which were the third-most common Gram-positive organisms causing LBSIs.

Although there is significant resistance to conventionally used antibiotics in the Gram-negative organisms, namely third-generation cephalosporin (cefotaxime) and gentamicin-there was no increasing trend apart from an increasing resistance to cefotaxime in E. coli. There was slight increase in resistance to fluoroquinolones amongst E. coli and Enterobacter isolates, but they are not frequently used in the neonatal population due to the long-term side effects. Increasing incidence of ESBL-producing organisms is a concern worldwide, more so in the middle- and lower-income countries.[7],[20] In our cohort, ESBL was seen in E. coli and Klebsiella Sp. There was an increasing trend in ESBL positive isolated amongst E. coli with up to 50% of the isolates testing positive in the last year of the study and 43% overall. This is higher than what has been reported from the public health surveillance systems for all isolates in the emirate of Abu Dhabi (both community and hospital-acquired from all sources-26% in 2019) and similar to that reported recently from the Indian subcontinent and is worrying.[7],[21]

Carbapenems are one of the last resort antibiotics currently available in the antimicrobial armamentarium. However due to increasing resistance to other antibiotic classes, the use of this group of antibiotics is increasing. Resistance to meropenem was seen in 3% of the Gram-negative isolates and was restricted to Pseudomonas isolates seen in years 2 and 3 of the study. Although there have been none seen in the second half of the study, the emergence of these organisms reinforces the need for good antibiotic stewardship programs and limits the use of broad-spectrum antibiotics.

 Conclusions



This study has described the bacteriological spectrum of late-on-onset bloodstream infections on a large tertiary unit in the Gulf region and the antimicrobial resistance trends. The antibiotic choices on our unit seem appropriate based on the review of data. In units where the organism and susceptibility patterns are similar to what we have seen locally, the same antibiotic choices may be justified. However, it is important that every neonatal service has access to local data on the distribution of pathogens and susceptibility data to periodically review the choice of empiric therapy. It is also important to keep empiric therapy to the narrowest spectrum possible to reduce the emergence of resistant organisms.

Human research statement

This research was conducted in accordance with the ethical standards of all applicable national and institutional committees and the World Medical Association's Helsinki Declaration. The study was approved by the Corniche Hospital Research Ethics Committee.

Acknowledgments

The authors are grateful to Dr. Stefan Weber, consultant microbiologist, Reference Laboratory for Infectious Diseases, Sheikh Khalifa Medical City, Abu Dhabi, United Arab Emirates for provision of the raw antibiotic sensitivity data and advice during the study and manuscript preparation.

Financial support and sponsorship

Nil

Conflicts of interest

There are no conflicts of interest.

References

1Lawn JE, Blencowe H, Oza S, You D, Lee AC, Waiswa P, et al. Every newborn: Progress, priorities, and potential beyond survival. Lancet 2014;384:189-205.
2Cortese F, Scicchitano P, Gesualdo M, Filaninno A, De Giorgi E, Schettini F, et al. Early and late infections in newborns: Where do we stand? A review. Pediatr Neonatol 2016;57:265-73.
3Shah BA, Padbury JF. Neonatal sepsis: An old problem with new insights. Virulence 2014;5:170-8.
4Bizzarro MJ, Raskind C, Baltimore RS, Gallagher PG. Seventy-five years of neonatal sepsis at Yale: 1928-2003. Pediatrics 2005;116:595-602.
5Cailes B, Kortsalioudaki C, Buttery J, Pattnayak S, Greenough A, Matthes J, et al. Epidemiology of UK neonatal infections: The neonIN infection surveillance network. Arch Dis Child Fetal Neonatal Ed 2018;103:F547-53.
6Bhat YR, Lewis LE, Vandana KE. Bacterial isolates of early-onset neonatal sepsis and their antibiotic susceptibility pattern between 1998 and 2004: An audit from a center in India. Ital J Pediatr 2011;37:32.
7Roy MP, Bhatt M, Maurya V, Arya S, Gaind R, Chellani HK. Changing trend in bacterial etiology and antibiotic resistance in sepsis of intramural neonates at a tertiary care hospital. J Postgrad Med 2017;63:162-8.
8Yadav NS, Sharma S, Chaudhary DK, Panthi P, Pokhrel P, Shrestha A, et al. Bacteriological profile of neonatal sepsis and antibiotic susceptibility pattern of isolates admitted at Kanti Children's Hospital, Kathmandu, Nepal. BMC Res Notes 2018;11:301.
9Hammoud MS, Al-Taiar A, Al-Abdi SY, Bozaid H, Khan A, AlMuhairi LM, et al. Late-onset neonatal sepsis in Arab states in the Gulf region: Two-year prospective study. Int J Infect Dis 2017;55:125-30.
10Hsieh EM, Hornik CP, Clark RH, Laughon MM, Benjamin DK Jr., Smith PB, et al. Medication use in the neonatal intensive care unit. Am J Perinatol 2014;31:811-21.
11Shim GH, Kim SD, Kim HS, Kim ES, Lee HJ, Lee JA, et al. Trends in epidemiology of neonatal sepsis in a tertiary center in Korea: A 26-year longitudinal analysis, 1980-2005. J Korean Med Sci 2011;26:284-9.
12Hammoud MS, Al-Taiar A, Al-Abdi SY, Bozaid H, Khan A, AlMuhairi LM, et al. Culture-proven early-onset neonatal sepsis in Arab states in the Gulf region: Two-year prospective study. Int J Infect Dis 2017;55:11-5.
13Hammoud MS, Al-Taiar A, Thalib L, Al-Sweih N, Pathan S, Isaacs D. Incidence, aetiology and resistance of late-onset neonatal sepsis: A five-year prospective study. J Paediatr Child Health 2012;48:604-9.
14Al-Taiar A, Hammoud MS, Cuiqing L, Lee JK, Lui KM, Nakwan N, et al. Neonatal infections in China, Malaysia, Hong Kong and Thailand. Arch Dis Child Fetal Neonatal Ed 2013;98:F249-55.
15Stoll BJ, Hansen N, Fanaroff AA, Wright LL, Carlo WA, Ehrenkranz RA, et al. Late-onset sepsis in very low birth weight neonates: The experience of the NICHD Neonatal Research Network. Pediatrics 2002;110:285-91.
16Investigators of the Delhi Neonatal Infection Study (DeNIS) Collaboration. Characterisation and antimicrobial resistance of sepsis pathogens in neonates born in tertiary care centres in Delhi, India: A cohort study. Lancet Glob Health 2016;4:e752-60.
17Kayange N, Kamugisha E, Mwizamholya DL, Jeremiah S, Mshana SE. Predictors of positive blood culture and deaths among neonates with suspected neonatal sepsis in a tertiary hospital, Mwanza-Tanzania. BMC Pediatr 2010;10:39.
18Aurangzeb B, Hameed A. Neonatal sepsis in hospital-born babies: Bacterial isolates and antibiotic susceptibility patterns. J Coll Physicians Surg Pak 2003;13:629-32.
19Pokhrel B, Koirala T, Shah G, Joshi S, Baral P. Bacteriological profile and antibiotic susceptibility of neonatal sepsis in neonatal intensive care unit of a tertiary hospital in Nepal. BMC Pediatr 2018;18:208.
20Folgori L, Bielicki J. Future challenges in pediatric and neonatal sepsis: Emerging pathogens and antimicrobial resistance. J Pediatr Intensive Care 2019;8:17-24.
21Department of Health. Communicable Diseases Bulletin, Quarterly Summary Report, Second Quarter. Vol. 10. Abu-Dhabi, UAE: Department of Heath; 2019. p. 7-8.