Approach to febrile neutropenia in the general paediatric setting

Preventing Infection in Pediatric Patients Receiving Chemotherapy: A Survey of Provider Recommendations

PURPOSE

Sepsis is the leading cause of mortality in patients with childhood cancer receiving cytotoxic chemotherapy. Pediatric hematology/oncology and transplant (PHOT) providers must counsel their patients on the safety of public activities and weigh the risk of infection exposure with the social and developmental benefits of in-person school and social outings. We hypothesize that there is significant variability in recommendations given by PHOT providers.

METHODS

An electronic anonymous survey was developed and piloted by a group of PHOT providers to assess current methods for educating patients and families on limiting infectious exposures. Five clinical vignettes were created by the study team to explore how providers balance the competing priorities of safety and health-related quality of life (HRQoL). The electronic survey was institutional review board-approved and disseminated via email to all PHOT providers affiliated with the Children's Oncology Group across the United States.

RESULTS

In total, 545 clinicians completed the survey. Most respondents were attending physicians (393, 72%), followed by fellows (61, 11%), advanced practice providers (APPs; 38, 7%), and nurses (37, 7%). On average, nurses and fellows made more conservative recommendations for avoiding infectious exposures compared with the recommendations from attending physicians and APPs (P < .0001). On average, providers with more years of clinical experience expressed less cautious recommendations, whereas those with less years of experience provided more cautious recommendations for avoiding infectious exposures (P = .0072).

CONCLUSION

This survey demonstrates the importance of collaboration between all members of the care team in defining priorities for balancing safety risk and HRQoL to provide consistent messaging to patients. The variations in survey responses highlight the need for universal guidelines to standardize physician recommendations for limiting infectious exposures in pediatric patients on chemotherapy.

Impact of a Pediatric Emergency Medicine Pharmacist, Institutional Guideline, and Electronic Order Set on Empiric Antimicrobial Use for Febrile Neutropenia

OBJECTIVES

This study evaluated the difference in appropriateness of antimicrobial selection in pediatric patients with febrile neutropenia (FN) after implementation of an institutional guideline, a dedicated pediatric emergency medicine (EM) pharmacist, and an electronic order set.

METHODS

This was a retrospective cohort study that included febrile patients aged younger than 18 years who were at risk of neutropenia, as defined by our institutional algorithm. Charts were evaluated for inclusion by searching for patients who presented to the emergency department (ED) between February 2018 and January 2022 who had International Classification of Diseases, Tenth Revision (ICD-10) codes for patients at risk of FN. Three independent groups were compared before, during, and after interventions. A historical control group (group 1), postdedicated EM pharmacist and institutional guideline cohort (group 2), and postdedicated EM pharmacist, institutional guideline, and electronic order set cohort (group 3) were compared. Secondary outcomes included time from registration in the ED to administration of the first dose of empiric antimicrobials, days to defervescence, pediatric intensive care unit length of stay, and hospital length of stay.

RESULTS

Seventy-eight charts were reviewed for inclusion. Among those included (n = 38), there was an increase in appropriate use of antimicrobials from 71% to 92% to 100% ( P = 0.1534) between group 1, group 2, and group 3, respectively. In addition, the interventions in this study lead to an overall decrease in the median time from registration to first dose of antibiotics from 142 minutes to 72 minutes ( P = 0.1370).

CONCLUSIONS

This study demonstrated the positive impact a pediatric EM pharmacist along with an institutional guideline and an electronic order set have on appropriate antimicrobial selection in pediatric FN patients. Institutions should consider multipronged approaches to improve the selection and time to administration of appropriate empiric antimicrobials in the ED.

La prise en charge de la neutropénie fébrile chez les enfants et les adolescents immunocompétents

La neutropénie fébrile est une manifestation clinique fréquente en pédiatrie, qui peut être associée à une infection bactérienne invasive. Cependant, le risque de ce type d'infection est faible chez les enfants et les adolescents autrement en santé qui font de la fièvre et présentent une neutropénie, la plupart des cas étant causés par une infection virale. Les enfants âgés de six mois à 18 ans qui ont l'air bien, ne souffrent pas d'un cancer, sont considérés comme immunocompétents et présentent un premier épisode de neutropénie, sans autres facteurs de risque, n'ont généralement pas besoin d'antibiotiques empiriques. Cependant, une évaluation approfondie est indiquée, y compris une anamnèse et un examen physique complets, de même qu'une hémoculture lorsque la numération absolue des neutrophiles est inférieure à 0,5 × 10⁹/L. Il est recommandé d'assurer un suivi étroit, de reprendre l'hémogramme et de donner des conseils préventifs stricts.

Management of febrile neutropenia in immunocompetent children and youth

Febrile neutropenia is a common clinical presentation in children that can be associated with invasive bacterial infection (IBI). However, in otherwise healthy children and youth with fever and neutropenia, the risk for IBI is low, with most cases being caused by viral infections. Well-appearing, non-oncologic, and presumed immunocompetent children aged 6 months to 18 years experiencing a first episode of neutropenia, with no additional risk factors, typically do not require empiric antibiotics. However, a thorough assessment, including complete history and physical exam, is indicated, and a blood culture should be performed when the absolute neutrophil count is <0.5 × 10⁹/L. Close follow-up, a repeat complete blood count, and strong anticipatory guidance are recommended.

Value-Based Care for Healthy Children With First Episode of Febrile Neutropenia

OBJECTIVES

There is a lack of guidance on the management of febrile neutropenia in otherwise healthy children, including the need for hospitalization and antibiotic administration, leading to significant practice variation in management. The aim of this initiative was to decrease the number of unnecessary hospitalizations and empirical antibiotics prescribed by 50% over a 24-month period for well-appearing, previously healthy patients older than 6 months presenting to the emergency department with a first episode of febrile neutropenia.

METHODS

A multidisciplinary team of stakeholders was assembled to develop a multipronged intervention strategy using the Model for Improvement. A guideline for the management of healthy children with febrile neutropenia was created, coupled with education, targeted audit and feedback, and reminders. Statistical control process methods were used to analyze the primary outcome of the percentage of low-risk patients receiving empirical antibiotics and/or hospitalization. Balancing measures included missed serious bacterial infection, emergency department (ED) return visit, and a new hematologic diagnosis.

RESULTS

Over the 44-month study period, the mean percentage of low-risk patients hospitalized and/or who received antibiotics decreased from 73.3% to 12.9%. Importantly, there were no missed serious bacterial infections, no new hematologic diagnoses after ED discharge, and only 2 ED return visits within 72 hours without adverse outcomes.

CONCLUSIONS

A guideline for the standardized management of febrile neutropenia in low-risk patients increases value-based care through reduced hospitalizations and antibiotics. Education, targeted audit and feedback, and reminders supported sustainability of these improvements.

Brucellosis: A Rare Cause of Febrile Neutropenia in a Child

We report a case of brucellosis-induced severe neutropenia in a 2-year-old girl who presented with a 2-week history of fever. On clinical examination, the patient was febrile with mild aphthous stomatitis. However, her general condition was stable, and systemic examination did not show involvement of any other organ. Laboratory test results revealed severe neutropenia, mild anemia, and an elevated serum C-reactive protein level. Flow cytometry of peripheral blood leukocytes revealed no malignancy, and blood film morphology was unremarkable except for mild microcytosis and hypochromia. Antineutrophil antibody and Coombs test results were negative. We administered intravenous cefuroxime; however, therapy was switched to meropenem plus clarithromycin because fever persisted for 5 days, despite treatment. On the 10th day after admission, Brucella serology tests showed positive results, and trimethoprim-sulfamethoxazole plus rifampicin therapy was prescribed for 8 weeks. The fever defervesced, and the child was discharged in a good state of health. Neutropenia persisted for several months but gradually resolved. Neutropenia, defined as an absolute neutrophil count (ANC) < 1.5 cells × 10 9 /L beyond the first year of life, is a benign transient condition associated with an intercurrent infection (usually viral illnesses or infections) in immunocompetent children. However, severe neutropenia (ANC < 0.5 × 10 9 /L) associated with fever necessitates hospitalization and administration of broad-spectrum antibiotics to avoid the high risk of sepsis, particularly in children. Brucellosis is rarely associated with hematologic abnormalities such as neutropenia. Early diagnosis of hematologic complications of brucellosis is essential for prompt initiation of specific and aggressive treatment.

Risk Factors of Serious Bacterial Infection in Previously Healthy Children Older Than 90 Days Old With Fever and Neutropenia

BACKGROUND

The main objective was to determine the clinical or analytical factors that independently predict risk of serious bacterial infection (RSBI) in immunocompetent patients older than 90 days given a diagnosis of fever and for whom neutropenia was an incidental finding. The secondary objective was to describe the prevalence of serious bacterial infections (SBIs).

METHODS

This is a 3-year-long, multicenter, prospective analytical and observational study carried out at 6 pediatric emergency departments. Data for epidemiological, clinical, and analytical variables were collected.

RESULTS

One hundred forty patients with febrile neutropenia (60.7% mild, 39.3% moderate to severe) were recruited. Serious bacterial infection incidence was 15.0% (95% confidence interval [CI], 9-21): 1 Invasive Bacterial Infection (Staphylococcus epidermidis bacteremia), 10 urinary tract infections, 8 pneumonias, and 2 cellulitis. Median total neutrophil counts per microliter showed no statistically significant differences (P = 0.512; 1000 [750-1200] in SBI patients vs 1100 [800-1300] in non-SBI patients). Higher RSBI was observed in patients with neutrophils less than 20% relative to total leukocytes (SBI, 15, 26.3%) than in those with neutrophils of 20% or greater (SBI, 6, 7.2%) (odds ratio, 4.6; 95% CI, 1.7-12.7). In patients with greater than 5000 leukocytes/μL, a percentage of neutrophils less than 20% was related to a greater RSBI with a trend toward statistical significance (odds ratio, 6.1; 95% CI, 0.7-51.1; P = 0.066). The clinical variables did not show a significant association with RSBI.

CONCLUSIONS

None of the clinical or analytical variables assessed were associated with the RSBI. However, according to a post hoc analysis, in patients with greater than 5000 leukocytes/μL, a neutrophil percentage less than 20% could be an independent risk factor for SBI. A thorough physical examination and basic diagnostic tests (urinalysis and chest x-ray) may help to establish a diagnosis of SBI in the vast majority of cases.

Emergency Department Utilization and Serious Bacterial Infection Rates in Children With Renal Transplants

OBJECTIVES

Emergency department (ED) visits by children with solid organ transplants have increased significantly. Our objectives were to describe the common complaints, diagnosis, types, and rates of serious bacterial infection (SBI) in children with renal transplant (RT) who present to the hospital.

METHODS

We conducted a retrospective study from 2012 to 2016 of RT children up to 18 years who presented to the ED or were directly admitted. We excluded patients who presented for a procedure. We collected demographics, transplant type, immunosuppressive data, chief complaints, diagnostic testing with results, interventions performed, and final diagnosis.

RESULTS

We analyzed 131 visits in 29 patients during the study period. Most common chief complaints were infectious (34.4%) and gastrointestinal (26%). Infection was proven in 42.0% of visits with only 3.1% being organ rejection. Serious bacterial infection was diagnosed in 34 visits (26.0%) with urinary tract infection (UTI) being the most common (20.6%). Of the 33 visits for fever, SBI occurred in 16 (48.5%) patients with the most common SBI being UTI 10 (30.3%). Bacteremia occurred in 1 patient and hypotension in 4 patients. Antibiotic administration was the most common intervention performed (78; 59.5%). Significant interventions were uncommon (2 patients). Logistic regression revealed no factors to be associated with SBI.

CONCLUSIONS

Our cohort of children with RT presented most commonly with infections to the hospital with UTI being the most common SBI. Bacteremia and significant interventions were rare. Future studies are needed to identify subgroups of low-risk pediatric RT patients who can possibly be safely discharged home from the ED.

The etiologic, microbiologic, clinical and outcome characteristics of immunocompetent young children <2 years of age hospitalized with acute neutropenia

BACKGROUND

To describe the etiologic, microbiologic, clinical and outcome characteristics of acute neutropenia (absolute neutrophil count, ANC, <1.5 × 10⁹/L) in hospitalized immunocompetent children.

METHODS

Serious bacterial infections (SBI) were defined as culture-positive blood, urine, cerebrospinal fluid, articular fluid or stool infections, alveolar pneumonia, Brucellosis and Rickettsiosis.

RESULTS

431/671 (64.2%) healthy infants and children hospitalized with acute neutropenia were <2 years of age; 176 (40.8%), 167 (38.8%) and 88 (20.4%) patients were aged 0-3, 4-12 and 13-24 months, respectively. There were 19 (4.4%), 53 (12.3%), 140 (32.5%) and 209 (50.8%) patients with ANC count <200, 200-500, 501-1000 and 1001-1500 × 10⁹ cells/L, respectively. Severe neutropenia (<500 × 10⁹/L) was recorded in 72 (16.7%) patients. Fever >38 °C was present in 208/431 (48.3%) patients. Blood cultures were positive in 10 (2.3%), with Brucella melitensis, Staphylococcus aureus and Enterobacter spp. identified in 4, 3 and 2 patients, respectively; 5/10 patients with positive blood cultures were <3 months of age. Overall, 55/431 (12.7%) and 65/431 (15.1%) patients were diagnosed with SBIs and bacterial infections, respectively. Nasal washings-PCR for respiratory viruses was positive in 139/293 (47.4%) patients tested. An infectious etiology (bacterial and/or viral) was diagnosed in 190/431 (44.1%) patients. Three patients were diagnosed with acute lymphocytic leukemia. Resolution of neutropenia was achieved in 111/208 (53.4%) evaluable patients (63%, 50.6% and 48% of patients aged 0-3, 4-12 and >12 months, respectively and 56.8%, 53.5% and 52% of patients with severe, moderate and mild neutropenia, respectively).

CONCLUSION

Acute neutropenia is common in immunocompetent children <2 years of age and is frequently associated with viral infections. We showed a substantial involvement of bacterial infections and particularly SBIs in the etiology of acute neutropenia. After a 1-month follow-up, resolution of neutropenia occurred in half of the patients, without association with age subgroups and with neutropenia severity.

The Management of Young Children With a Likely Infectious Condition Presenting Moderate to Severe Neutropenia

OBJECTIVE

The objective of this study was to describe the outcome of healthy children presenting with newly-diagnosed neutropenia in an infectious context.

RESULTS

A total of 184 episodes of neutropenia were included in children 3 months to 5 years of age. There were 118 (64%) episodes of moderate neutropenia and 66 (36%) of severe neutropenia (SN). SN episodes were more likely related to intensification of antibiotic regimen used and further investigations. The median duration of neutropenia was 8.5 days. Chronic benign neutropenia occurred in 7 (4%) patients.

CONCLUSION

SN led to intensification of antibiotic therapy, but no children encountered an unfavorable outcome and the neutropenia episodes were short-lived.

Comparison of the etiologic, microbiologic, clinical and outcome characteristics of febrile vs. non-febrile neutropenia in hospitalized immunocompetent children

We compared the etiologic, microbiologic, clinical, and outcome picture among febrile and non-febrile immunocompetent children hospitalized during 2013–2015 with acute neutropenia (absolute neutrophil count < 1.5 × 10⁹/L). Serious bacterial infections (SBI) were defined as culture-positive blood, urine, cerebrospinal fluid, articular fluid or stool infections, pneumonia, brucellosis, and rickettsiosis. Overall, 664 children < 18 years of age were enrolled; 407 (62.2%) had fever > 38.0 °C and 247 (37.8%) were non-febrile at admission. There were 425 (64.0%), 125 (18.8%), 48 (7.2%), and 66 (9.9%) patients aged 0–24 months, 2-6, 7–12, and > 12 years, respectively. No differences were recorded in the distribution of febrile vs. non-febrile patients among the age groups nor among the 3 neutropenia severity groups (< 0.5, 0.5–1.0 and 1.0–1.5 × 10⁹/L). SBI infections were diagnosed in 98 (14.8%) patients, with higher rates among febrile patients vs. non-febrile patients (16.8% vs. 11.5%, P = 0.06). Brucellosis and rickettsiosis were diagnosed in 15.4% and 23.1% tests performed, respectively. 295/688 (42.9%) virologic examinations returned positive. Among patients < 24 months, more febrile ones had viral infectious compared with afebrile patients (P = 0.025). Acute leukemia was diagnosed in 6 patients. Neutropenia resolved in 163/323 (50.5%) patients during a 1-month follow-up. No differences were recorded in neutropenia resolution between febrile and non-febrile children among all 3 severity groups. Severe neutropenia was rare and occurred mainly in very young patients. SBIs were more common among febrile patients compared with non-febrile patients, but there was no association between severity of neutropenia or its resolution and the presence or absence of fever at diagnosis.

Clinical Profile and Short Term Outcome of Children with Neutropenia

OBJECTIVES

To study the clinical profile and short term outcome of neutropenia in children during hospital stay.

METHODS

An observational study was carried out at a Children's Hospital. The study population comprised of 77 admitted children aged 1 mo to 18 y with a varied clinical profile and neutropenia; defined as absolute neutrophil count less than 1500/mm³. Patients known to have HIV, immunodeficiency, malignancy, aplastic anemia or chronic systemic illness were not enrolled. Necessary investigations were done to identify etiology as per clinical features. They were treated and followed up for a period of 4 wk or discharge; whichever was earlier. The primary outcome of duration of neutropenia and secondary outcomes of hospital stay duration, association with thrombocytopenia, incidence of complications and finally discharge/death were analyzed.

RESULTS

Acute transient neutropenia was seen, the median duration being 3 d in younger patients. Dengue fever was the commonest etiology. The median duration of hospital stay was 8 d. Fifty three (68.8%) patients had associated thrombocytopenia. Three children developed complications like nosocomial sepsis and shock. Seventy two (93.5%) were discharged, 1 died, 3 left against medical advice and one patient was followed up for 4 wk.

CONCLUSIONS

Acute febrile illnesses like dengue, enteric fever, malaria are the predominant causes of neutropenia. Mild neutropenia was seen in over 50% children requiring a short duration of admission (5-8 d); without any complications.

Outcomes and infectious etiologies of febrile neutropenia in non-immunocompromised children who present in an emergency department

There are no unified protocols governing the management of healthy children with febrile neutropenia in the emergency department (ED). Conservative management is the norm, with admission and empirical broad-spectrum antibiotics prescribed, although viral infections are considered the most frequent etiology. The aim of this study was to describe the clinical outcomes and identified etiologies of unsuspected neutropenia in febrile immunocompetent children assessed in the ED. This was a retrospective study: well-appearing healthy children <18 years old with febrile moderate [absolute neutrophil count (ANC) 500-999 neutrophils ×10(9)/l] or severe (ANC <500 neutrophils ×10(9)/l) neutropenia diagnosed in ED between 2005 and 2013 were included. Patients newly diagnosed with hematologic or oncologic disease were excluded. We included 190 patients: 158 (83.2 %) with moderate and 32(16.8 %) with severe neutropenia. One hundred and one (53.2 %) were admitted; 48(47.5 %) with broad-spectrum antibiotics. The median length of stay was 3 days (IQR 3-5) and the median duration of neutropenia was 6 days (IQR 3-12). An infectious agent was identified in 23(12.1 %); 21 (91.3 %) were viruses. Four (2.1 %) children had a serious bacterial infection (SBI): urinary tract infection and lobar pneumonia (two cases each). All blood cultures performed (144; 75.8 %) were negative. Over the 1-year follow-up, one or several blood tests were performed on 167 patients (87.9 %); two (1.2 %) were diagnosed with autoimmune chronic neutropenia. Previously healthy children with moderate or severe febrile neutropenia have a low risk of SBI and a favorable clinical outcome. Less aggressive management could be carried out in most of them. Although chronic hematological diseases are infrequently diagnosed, serial ANC are necessary to detect them.

The Risk of Serious Bacterial Infection in Neutropenic Immunocompetent Febrile Children

Only few reports have looked into the risk of invasive bacterial infection in children with neutropenia that is not malignancy related. The objective of the current study was to determine the clinical significance of neutropenia as a predictor of serious bacterial infection (SBI) in immunocompetent children. We conducted a retrospective case-control study including children 3 months to 18 years of age with fever ≥ 38°C hospitalized or presenting to the emergency department. Patients who had neutropenia ≤ 1000 ANC/μL and had a blood culture taken were matched for age with the consecutive febrile patients for whom a blood culture was taken. The main outcome was the rate of SBI. SBIs were more prevalent among the control group than in the group of children with neutropenia, 19/71 and 6/71, respectively (P = 0.0005). More children were treated with antibiotics among the control group than in the group of children with neutropenia, 39/71 and 20/71, respectively (P < 0.0001). Acute-phase reactants including CRP and platelets were higher in the control group. We concluded that immunocompetent patients with fever and moderate neutropenia do not carry a higher risk for SBIs compared with patients with fever who do not have neutropenia.

Infectious events prior to chemotherapy initiation in children with acute myeloid leukemia

BACKGROUND

The primary objective was to describe infectious complications in children with acute myeloid leukemia from presentation to the healthcare system to initiation of chemotherapy and to describe how these infections differ depending on neutropenia.

METHODS

We conducted a retrospective, population-based cohort study that included children and adolescents with acute myeloid leukemia diagnosed and treated at 15 Canadian centers. We evaluated infections that occurred between presentation to the healthcare system (for symptoms that led to the diagnosis of acute myeloid leukemia) until initiation of chemotherapy.

RESULTS

Among 328 children, 92 (28.0%) were neutropenic at presentation. Eleven (3.4%) had sterile-site microbiologically documented infection and four had bacteremia (only one Gram negative). Infection rate was not influenced by neutropenia. No child died from an infectious cause prior to chemotherapy initiation.

CONCLUSION

It may be reasonable to withhold empiric antibiotics in febrile non-neutropenic children with newly diagnosed acute myeloid leukemia until initiation of chemotherapy as long as they appear well without a clinical focus of infection. Future work could examine biomarkers or a clinical score to identify children presenting with leukemia and fever who are more likely to have an invasive infection.

Low risk of bacteremia in otherwise healthy children presenting with fever and severe neutropenia

We examined 785 placentas, including 51 from documented cases of congenital toxoplasmosis. Toxoplasma was detected in 16 placentas,including 1 in which congenital toxoplasmosis was ruled out. Placental screening had poor sensitivity (25%) but good specificity (99%), positive predictive value (93%), and negative predictive value (95%).