Secondary Logo

Journal Logo

Immunology and Host Response

Viridans Group Streptococci in Pediatric Leukemia and Stem Cell Transplant: Review of a Risk-stratified Guideline for Empiric Vancomycin in Febrile Neutropenia

Ruhayel, Sandra D. MBBS, DCH*; Foley, David A. MMED; Hamilton, Kate MMS; Ferguson, Patricia FRACP, PhD; Kotecha, Rishi S. MRCPCH, FRACP, PhD*,§,¶; Bowen, Asha C. FRACP, PhD†,∥,**; Yeoh, Daniel K. FRACP, MCHR†,††

Author Information
The Pediatric Infectious Disease Journal: September 2021 - Volume 40 - Issue 9 - p 832-834
doi: 10.1097/INF.0000000000003210


Viridans group streptococci (VGS) are an important cause of bacteremia in children with cancer.1,2 VGS bloodstream infection (VGS-BSI) can rapidly progress to cause a toxic shock-like syndrome characterized by hypotension and acute respiratory distress syndrome, with mortality of up to 23% in children.1,3 Groups at highest risk of VGS-BSI include children with acute myeloid leukemia (AML),1 relapsed acute lymphoblastic leukemia (ALL),4 infant ALL5 and those undergoing allogeneic hematopoietic stem cell transplant (HSCT).1

VGS-BSI-associated mortality, coupled with a rise in penicillin-resistance among VGS isolates,6 has led to advocacy for the addition of empiric vancomycin to treat febrile neutropenia in high-risk children.7 However, recommendations differ between guidelines and practice varies between centers.8,9 At our institution, the addition of empiric vancomycin for the initial management of febrile neutropenia is recommended in patients at the highest risk of VGS-BSI. We aimed to assess the efficacy of this risk-stratified approach by examining the epidemiology, microbiology and outcomes of VGS-BSI at our center.


We conducted a single-center prospective cohort study including all children (0–18 years) receiving chemotherapy for hematologic malignancy or undergoing allogeneic HSCT from November 1, 2016 until January 31, 2020 at Perth Children’s Hospital in Western Australia. All patients diagnosed with leukemia including relapsed disease, and those undergoing allogeneic HSCT during the study period were included. As per institutional guidelines, the addition of empiric vancomycin to standard febrile neutropenia therapy (piperacillin–tazobactam) is recommended in all patients at highest risk of VGS-BSI (those with AML, relapsed ALL, infant ALL or undergoing allogeneic HSCT), and any patient with clinical signs of shock. For patients not initially commenced on vancomycin, the addition of vancomycin was recommended in the event a Gram-positive organism was identified from blood culture.

The shock was defined as inadequate cardiac output with hypotension requiring intravascular volume expansion (fluid bolus ≥20 mL/kg) and/or inotropic support. Neutropenia was defined as an absolute neutrophil count of <0.5 × 109/L. Patients with a leukocyte count of <0.1 × 109/L where a neutrophil count was not performed, were assumed to have concurrent neutropenia (<0.5 × 109/L). Fever was defined as core body temperature ≥38.5°C on a single occasion or ≥38.0°C on 2 sequential occasions within a 24-hour period.

All VGS-BSI episodes were identified from the laboratory information database of blood culture isolates. A VGS-BSI episode was defined as isolation of Viridans group Streptococcus spp. from a blood culture. Episodes with concomitant growth of a Gram-positive commensal (eg, Micrococcus spp. or coagulase-negative Staphylococcus) were excluded as contaminants.

For each VGS-BSI episode, demographic and clinical data were collected from patient medical records including the presence of shock, the requirement for mechanical ventilation, admission to the intensive care unit (ICU) and survival at day 7 and day 30. Laboratory data included blood culture time-to-positivity and antibiotic susceptibility. Susceptibility testing was carried out using an E-test (Biomerieux, France) as per manufacturer specifications. Susceptibility to empiric B-lactam antibiotic (piperacillin–tazobactam) was inferred from penicillin minimum inhibitory concentration, with breakpoints defined as per the Clinical & Laboratory Standards Institute.10

Statistical analysis was performed using GraphPad QuickCalcs (GraphPad Software, Inc., CA). Fisher’s exact test was used to compare the characteristics of high-risk and non-high-risk patients for categorical data, and the Mann-Whitney U test for continuous variables. P values of <0.05 were considered statistically significant. National ethics approval was granted by the Western Sydney Local Health District Human Research Ethics Committee (HREC/16/WMEAD/281).


A total of 107 patients received chemotherapy for a hematologic malignancy or underwent allogeneic HSCT during the study period. Seventy-four patients (69.2%) had a diagnosis of ALL (1 with relapsed disease and 2 with infant ALL); 18 (16.8%) had AML and 15 (14%) underwent allogeneic HSCT. A total of 34 patients were identified as high-risk of VGS-BSI on the basis of underlying diagnosis or treatment with HSCT.

Of 84 total BSI episodes that occurred during the study period, 19 (22.6%) were due to VGS-BSI (Table 1). Among patients with VGS-BSI, median age was 6.7 (interquartile range [IQR] 2.3–10.7) years and 63.2% (12/19) of episodes occurred in female patients. The majority (57.9%, 11/19) of VGS-BSI episodes occurred in high-risk patients; 7 with AML, 3 post-HSCT and in 1 child with ALL and clinical evidence of shock. Of these patients classified as at high-risk of VGS-BSI, 81.9% (9/11) had been commenced on initial empiric vancomycin therapy in accordance with institutional guidelines.

TABLE 1. - Comparison of Diagnosis, Demographics, Clinical Data and Treatment, Microbiologic Data Between High-Risk and Non-High-Risk Patients With VGS-BSI (n = 19)
High-risk (n = 11) N (%) Non-high-risk (n = 8) N (%) P Total (n = 19) N (%)
Underlying diagnosis
 AML 7 (63.6) 7 (36.8)
 HSCT 3 (27.3) 3 (15.8)
 ALL 1 (9.1) 8 (100) 9 (47.4)
 Relapsed ALL 0 (0) 0 (0)
 Infant ALL 0 (0) 0 (0)
 Male 2 (18.2) 5 (62.5) 0.0739 7 (36.8)
 Median age in years (IQR) 4.8 (1.7–12.9) 7.3 (4.3–9.3) 0.5353 6.7 (2.3–10.7)
Clinical data and treatment
 Neutropenia 10 (90.9) 7 (87.5) 1.00 17 (89.5)
 Shock present 1 (9.1) 0 (0) 1.00 1 (5.3)
 Empiric vancomycin given 9 (81.8) 1 (12.5) 0.0055 10 (52.6)
 Vancomycin added once BC positive 1 (9.1) 6 (75) 0.0063 7 (36.8)
 ICU admission 1 (9.1) 0 (0) 1.00 1 (5.3)
 Death 1 (9.1) 0 (0) 1.00 1 (5.3)
 VGS-BSI related death 0 (0) 0 (0) 1.00 0 (0)
Microbiologic data
 Non-piperacillin/tazobactam susceptible isolate* 7 (63.6) 4 (50) 0.6577 11 (57.9)
 Piperacillin/tazobactam susceptible isolate* 4 (36.4) 4 (50) 0.6577 8 (42.1)
 Isolate susceptible to recommended empiric antibiotics 11 (100) 4 (50) 15 (78.9)
*Piperacillin/tazobactam susceptibility was inferred from penicillin susceptibility results.
AML indicates acute myeloid leukemia; ALL, acute lymphoblastic leukemia; BC, blood culture; HSCT, hematopoietic stem cell transplant; IQR, interquartile range; ICU, intensive care unit; VGS-BSI, viridian group streptococci-bloodstream infection.

VGS species isolated included Streptococcus mitits (12 episodes), Streptococcus spp. – viridans group (2 episodes), Streptococcus parasanguinis (1 episode), Streptococcus salivarius (1 episode), Streptococcus oralis (1 episode), Streptococcus gordonii (1 episode) and Streptococcus cristatus (1 episode).

Of the VGS isolates, 21.1% (4/19) were resistant to piperacillin/tazobactam (inferred from penicillin susceptibility testing), and 36.8% (7/19) had reduced susceptibility. The median blood culture time-to-positivity was 11.7 (IQR 9.1–13.8) hours; 100% flagged positive within 17 hours (see Figure, Supplemental Digital Content 1; VGS isolates were susceptible to recommended risk-stratified empiric antibiotics in 100% (11/11) of episodes in high-risk patients and 78.9% (15/19) of all episodes. In all non-high-risk, clinically stable patients with a piperacillin/tazobactam-resistant VGS isolate (4/19), vancomycin was commenced once a positive blood culture was detected, all within 24 hours of fever onset.

All VGS-BSI episodes were associated with fever, and 89.5% (17/19) of patients had neutropenia. One patient developed intracranial hemorrhage secondary to a leukemia-related coagulopathy and died in ICU shortly after, unrelated to the VGS-BSI. All other patients recovered without requiring ICU admission; there were no VGS-BSI-related deaths. No patients with VGS-BSI had vancomycin-associated nephrotoxicity.


VGS-BSI can cause life-threatening complications in children with cancer and early initiation of appropriate antibiotic therapy is essential. With increasing penicillin-resistance among VGS isolates,6 empiric febrile neutropenia monotherapy with an extended-spectrum B-lactam agent may be suboptimal in patients at high-risk of VGS-BSI.7 However, the universal addition of vancomycin to empiric febrile neutropenia therapy has not shown benefit, increases nephrotoxicity,11 may increase antimicrobial resistance,12 and is not routinely recommended.8,9 A risk-stratified approach limiting the addition of empiric vancomycin to empiric febrile neutropenia therapy to high-risk and clinically unstable patients, maximizes utility while minimizing the potential for harm. This report demonstrates that this risk-stratified approach is effective, guiding appropriate therapy and supporting good clinical outcomes in VGS-BSI.

The proportion of non-penicillin susceptible VGS isolates in our study (57.9%) was comparable with previous pediatric studies (29.6%–67.5%).2,4 Reassuringly, there were no VGS-BSI-related deaths and all VGS isolates in high-risk patients were susceptible to the recommended empiric antibiotics. Although 4 non-high-risk patients had a non-penicillin susceptible VGS-BSI, vancomycin was added within 24 hours and all recovered without incident. The clinical stability observed in these patients, despite the slight delay in vancomycin therapy, may attest to the lower pathogenicity of VGS in lower-risk populations. Ideally, individual institutional febrile neutropenia empiric antibiotic recommendations should be informed by local epidemiologic data on antibiotic resistance.

Modern laboratory methods allow continuous assessment of blood culture samples, allowing early notification and time-to-positivity assessment. In this study, all blood cultures were detected as positive for VGS-BSI within 17 hours, indicating that empiric vancomycin targeting VGS can be safely discontinued in stable patients at 24 hours if cultures remain negative.

This study is limited by the single-center design and relatively small number of VGS-BSI episodes, but given the limited data assessing similar risk-stratified approaches, the findings remain important. Moreover, the safety of our risk-stratified approach was demonstrated over the 3-year study period in a cohort comprising more than 100 pediatric patients. Second, we employed a pragmatic definition of BSI including all single VGS isolates, which may have led to an overestimation of VGS-BSI episodes. The definition of VGS-BSI is not standardized and while some previous surveillance studies required isolation of VGS from 2 samples,1,4 several other studies have defined VGS-BSI as any single isolation of VGS,13 as this would result in treatment initiation from a clinical perspective. Notably, all VGS-BSI episodes in this study were associated with fever and were treated as true episodes of bacteremia. Finally, regarding the assessment of safety, the overall use of empiric vancomycin in episodes of febrile neutropenia (including the majority, where VGS was not isolated) was not specifically assessed. In a separate institutional audit assessing vancomycin-associated nephrotoxicity over a concurrent 12-month period, only 1 episode of nephrotoxicity associated with initial empiric vancomycin use occurred among hemato-oncology patients.14 Although the safety of this approach was not formally assessed in this study, these audit data together with ongoing monitoring for vancomycin-associated nephrotoxicity by the local antimicrobial stewardship serve to suggest that this approach is safe.

In conclusion, this data demonstrates the efficacy of a risk-stratified approach targeting the addition of empiric vancomycin to febrile neutropenia therapy in children with cancer at high-risk of VGS-BSI. The majority of patients received the appropriate antibiotic cover for VGS-BSI and no episodes progressed to VGS-BSI-related shock requiring ICU admission; there were no VGS-BSI-associated deaths. All VGS isolates flagged positive within 24 hours indicating cessation of empiric vancomycin can be considered in clinically stable patients with no growth on blood culture at 24 hours. A broader assessment of this risk-stratified approach is warranted.


We would like to acknowledge the Children’s Antimicrobial Stewardship Program at Perth Children’s Hospital and the Nurse Consultants of the Infection Prevention and Control team (in particular Dallas Sewell and Stacey Fitzgerald). We would like to thank the larger team behind the “Bacteraemia in Malignant Haematology and Haematopoietic Stem Cell Transplantation: A Multicentre Registry” through the ASID Clinical Research Network.


1. Gassas A, Grant R, Richardson S, et al. Predictors of viridans streptococcal shock syndrome in bacteremic children with cancer and stem-cell transplant recipients. J Clin Oncol. 2004;22:1222–1227.
2. Ahmed R, Hassall T, Morland B, et al. Viridans Streptococcus bacteremia in children on chemotherapy for cancer: an under-estimated problem. Pediatr Haematol Oncol. 2003;20:439–444.
3. Huang WT, Chang LY, Hsueh PR, et al. Clinical features and complications of viridans streptococci bloodstream infection in pediatric hemato-oncology patients. J Microbiol Immunol Infect. 2007;40:349–354.
4. Nielsen MJ, Claxton S, Pizer B, et al. Viridans group streptococcal infections in children after chemotherapy or stem cell transplantation: a 10-year review from a tertiary pediatric hospital. Medicine (Baltimore). 2016;95:e2952.
5. Salzer W, Dinndorf P, Dreyer Z, et al. Analysis of infectious complications in infants with acute lymphoblastic leukemia treated on the Children’s Cancer Group Protocol 1953: a report from the Children’s Oncology Group. J Pediatr Hematol Oncol. 2009;31:398–405.
6. Bruckner L, Gigliotti F. Viridans group streptococcal infections among children with cancer and the importance of emerging antibiotic resistance. Semin Pediatr Infect Dis. 2006;17:153–160.
7. Adcock KG, Akins RL, Farrington EA. Evaluation of empiric vancomycin therapy in children with fever and neutropenia. Pharmacotherapy. 1999;19:1315–1320.
8. Lehrnbecher T, Robinson P, Fisher B, et al. Guideline for the management of fever and neutropenia in children with cancer and hematopoietic stem-cell transplantation recipients: 2017 update. J Clin Oncol. 2017;35:2082–2094.
9. Freifeld AG, Bow EJ, Sepkowitz KA, et al.; Infectious Diseases Society of America. Clinical practice guideline for the use of antimicrobial agents in neutropenic patients with cancer: 2010 Update by the Infectious Diseases Society of America. Clin Infect Dis. 2011;52:427–431.
10. Institute, C.a.L.S. M100 Performance Standards for Antimicrobial Susceptibility Testing. 2020 [cited 2020 01/09/2020]; 30th Edition.
11. Beyar-Katz O, Dickstein Y, Borok S, et al. Empirical antibiotics targeting gram-positive bacteria for the treatment of febrile neutropenic patients with cancer. Cochrane Database Syst Rev. 2017;6:CD003914.
12. Karandikar MV, Milliren CE, Zaboulian R, et al. Limiting vancomycin exposure in pediatric oncology patients with febrile neutropenia may be associated with decreased vancomycin-resistant enterococcus incidence. J Pediatric Infect Dis Soc. 2020;9:428–436.
13. Brunet AS, Ploton C, Galambrun C, et al. Low incidence of sepsis due to viridans streptococci in a ten-year retrospective study of pediatric acute myeloid leukemia. Pediatr Blood Cancer. 2006;47:765–772.
14. Nettleton M. Vancomycin Stewardship Project: Audit of Vancomycin Use and Rates of Nephrotoxicity in Haematology/Oncology Patients over a 12-Month Period (from January 1 2018 to 31 December 2018). Perth Children’s Hospital; 2020. Unpublished data.

viridans group streptococci; vancomycin; febrile neutropenia; leukemia; hematopoietic stem cell transplant

Supplemental Digital Content

Copyright © 2021 Wolters Kluwer Health, Inc. All rights reserved.