Re-thinking treatment strategies for febrile neutropenia in paediatric oncology population: the perspective from a developing country

Abstract

Background This study was conducted to evaluate the microbiological profile of bacterial isolates in febrile neutropenia in a pediatric oncology unit, thereby, reviewing the use of restricted antibiotics and need for aggressive medical treatment accordingly. Methods A prospective observational study was conducted in a paediatric haemat-oncology division of a tertiary care teaching hospital in southern India from September 2014 to August 2016. One hundred and thirty children with febrile neutropenia were enrolled in the study. Blood cultures were performed using automated system. Cultures from other sites were obtained if needed, based on the clinical profile. Standard antibiotic susceptibility testing was done. Statistical analysis was done using SPSS. Results One hundred and thirty children were enrolled for the study. Two hundred and fifty episodes of febrile neutropenia were studied. Three hundred and eighty four cultures were sent and 92 (24%) cultures were positive. There were 48 (52.2%) Gram negative isolates followed by 33 (35.8%) Gram positive isolates, six (6.5%) fungal isolates and five (5.5%) poly-microbial cultures. Lactose fermenting Gram negative bacilli (20 isolates, 31.5%) were the most frequently isolated in the Gram negative group, with Escherichia coli being the most common organism (19 isolates, 20.6%). Amongst the Gram positive coagulase negative staphylococcus was the most common (twenty seven isolates, 29%). Escherichia coli and Non lactose fermenting gram negative bacteria (NFGNB) had only 36, 25% sensitivity to ceftazidime, respectively. Most Gram negative bacilli were found to have better sensitivity to amikacin (mean: 57%). There was a higher prevalence of extended spectrum beta lactamase producing organisms. Pan drug resistance, Extreme drug resistance and Multi drug resistance was found in three, twenty and thirteen Gram negative isolates respectively.Escherichia coli and Klebsiella were often drug resistant. Significantly higher mortality was associated with Gram negative isolates (eight deaths out of the thirteen deaths, 61.5%). Conclusions Our results show the importance of surveillance, monitoring resistance frequencies and identifying risk factors specific to each region. Given that significant mortality is attributed to drug resistant Gram negative bacilli, early initiation of appropriate antibiotics to cover for drug resistance is required while formulating empirical antibiotic policies for febrile neutropenia in the oncology units in the developing world.

Comparison Between Peripheral and Central Blood Cultures in Pediatric Oncology Patients With Blood Stream Infections

“…However, in clinical settings, when the amount of blood drawn simultaneously from peripheral and CVC culture is not equal and semiquantitative and quantitative cultures are not routinely performed, this information is not available and the decision on whether or not to remove the catheter is mostly based on patient's clinical condition and response to therapy. 23 Most of the pathogens isolated in our cohort (51.4%) were gram-negative bacteria similar to recent reports of fever and neutropenia in adult and pediatric patients outside the United States (east Asia and Europe) 24,25 but in contrast to studies from the United States and France in which most of the isolates were gram-positive mainly viridans group streptococci and CONS. 26,27 In addition, the growing resistance (especially among gram-negative bacteria) in oncology children as seen in our cohort (35.1% of ESBL phenotype among Enterobacteriaceae spp.)…”

“…Most of the pathogens isolated in our cohort (51.4%) were gram-negative bacteria similar to recent reports of fever and neutropenia in adult and pediatric patients outside the United States (east Asia and Europe) 24,25 but in contrast to studies from the United States and France in which most of the isolates were gram-positive mainly viridans group streptococci and CONS 26,27 . In addition, the growing resistance (especially among gram-negative bacteria) in oncology children as seen in our cohort (35.1% of ESBL phenotype among Enterobacteriaceae spp.)…”

Gaps in diagnosing suspected infection in immunocompromised children with cancer: A systematic review

“…A summary of the study characteristics is shown in Figure 2. The included studies were from diverse regions globally, with six (18%) conducted in the United States 2,11–15 and 28 (82%) studies in other countries (Africa, n = 1 [3%]; Asia, n = 15 [44%]; Oceania, n = 2 [6%]; Europe, n = 4 [12%]; Middle East, n = 4 [12%]; Central America, n = 1 [3%]; South America, n = 1 [3%]) 16–43 (Figure S2). Majority ( n = 21, 62%) 2,11,12,16,18,19,21,25,26,28–35,38,39,42,43 of studies utilized standard culture methods, while a smaller number ( n = 8, 24%) 13,14,20,22–24,27,41 of studies reported on newer methods such as metagenomic sequencing and PCR diagnostics (i.e., qPCR [including singleplex and multiplex], 22,23 panfungal real‐time PCR, 27 16S rRNA gene PCR, 23,24 respiratory viral panel PCR, 13,20,41 and Pneumocystis jirovecii PCR 20 ).…”

“…The included studies were from diverse regions globally, with six (18%) conducted in the United States 2,11–15 and 28 (82%) studies in other countries (Africa, n = 1 [3%]; Asia, n = 15 [44%]; Oceania, n = 2 [6%]; Europe, n = 4 [12%]; Middle East, n = 4 [12%]; Central America, n = 1 [3%]; South America, n = 1 [3%]) 16–43 (Figure S2). Majority ( n = 21, 62%) 2,11,12,16,18,19,21,25,26,28–35,38,39,42,43 of studies utilized standard culture methods, while a smaller number ( n = 8, 24%) 13,14,20,22–24,27,41 of studies reported on newer methods such as metagenomic sequencing and PCR diagnostics (i.e., qPCR [including singleplex and multiplex], 22,23 panfungal real‐time PCR, 27 16S rRNA gene PCR, 23,24 respiratory viral panel PCR, 13,20,41 and Pneumocystis jirovecii PCR 20 ). Other assessed tests or studies included histopathology examination, 15 Aspergillus galactomannan antigen assay (GM), 15,17,20,27 (1‐3)‐β‐D‐glucan assay (BDG), 15,27 and procalcitonin and C‐reactive protein biomarkers 37 .…”

“…Twenty‐four studies reported on the clinical course, including antimicrobial therapy 2,11,12,14–21,23,25,29,30,33,35–41,43 . The diagnostic impact on antimicrobial use was reported in only three studies.…”