Retrospective evaluation of the relationship of febrile neutropenia attacks and urinary system infection in children with leukemia

Objective: Febrile neutropenia is a common and important complication of cancer patients receiving chemotherapy. In febrile neutropenia attacks, the places where infections are seen intensely are the intestinal system, lungs, skin and soft tissue and urinary system. However there are not many studies showing the relationship between urinary system infections and febrile neutropenia in children with acute leukemia. This study was carried out to show the relationship between febrile neutropenia attacks and urinary tract infection in childhood leukemia patients. Materials and Methods: Febrile neutropenia attacks of leukemia patients followed in Manisa Celal Bayar University Medical Faculty Hospital Pediatric Hematology Clinic between 2013 and 2020 were retrospectively analyzed. Absolute neutrophil count, C-Reactive Protein level, blood and urine culture growth result and antibiotic sensitivity properties, urine analysis, antibiotic prophylaxis before febrile neutropenic attack and general treatment of febrile neutropenia attacks were investigated. Results: In the study, we analyzed 128 febrile neutropenia attacks (mean attack per patient 3.5) in 36 patients (72.2% male, 27.8% female) aged 0-17 years (mean age: 7,4 ±5,1).Of the patients, 29 (80.6%) were ALL, 5 (16.7%) AML and 1 (2.8%) JMML. During the study period, 5 patients (13.9%) died, the remaining 31 (86.1%) patients were alive. When we divide the febrile neutropenia attacks into groups according to the absolute neutrophil count, 63.3% are deep neutropenic (<100 m³/L), 36.2% are severe neutropenic (100-500 m³/L) and 0.2% are moderate neutropenic (>500m³/L) was found. When we look at the infection foci of febrile neutropenia attacks; 21.1% bacteremia, 10.9% culture positive urinary tract infection, 10.1% possible UTI and 17.9% neutropenic fever without focus were evaluated. The sum of definite UTI and probable UTI detected episodes (21%) was almost equal with the incidence of bacteremia (21.1%). When we look at the blood-urine culture growth rates, we found 21.1% positivity in blood cultures and 10.9% in urine cultures. Of the microorganisms grown in the blood culture, 77.8% were gram positive and 22.2% were gram negative. Cogulase-negative staphylococcus (55.6%) was the most isolated agent among gram-positive microorganisms and Klebsiella (11.1%) among gram-negative microorganisms. While the most isolated agents in urine culture were 50% Escherichia Coli, 2nd place was 43% Klebsiella, and 7% growth was detected from gram positive agents. As a result of the antibiotic sensitivity, it was observed that both bacteria that grew most in the urine culture were 100% resistant to Trimethoprim -Sulfamethoxazole and Ampicillin, 50% sensitive to 3rd Generation Cephalosporins, and 100% sensitive to Imipenem and nitrofurantoin. In the urine analysis of febrile neutropenic children, leukocyte-esterase, nitrite, and leukocyturia sensitivity were found to be limited (42.5%, 35.5%, 50%, respectively). At the same time, the sensitivity of localized findings related to UTI was found to be limited to 50%. In our study, when we looked at the urine samples of all febrile neutropenia episodes, 30 (23.5%) had findings in the urine analysis (at least one of the leukocytesterase, nitrite, leukocyturia findings were positive), but only 8 (26.6%) had urine culture growth. Although there were findings in the remaining 22 (73.4%) urine analysis, urine culture growth did not occur. In addition, urine culture positivity was found in only 5 (27.8%) of 18 (14%) febrile neutropenia episodes, which were thought to have possible urinary tract infection at the beginning (both there are findings in the urine analysis and there are clinical findings related to urinary tract infection). No urine culture growth was observed in 3 (72.2%) (when the growth threshold value was 105CFU/ml). Discussion and Conclusion: We observed that urinary tract infection was as common in the pediatric acute leukemia patient with febrile neutropenia as it was in other children. Therefore, even if there are no signs of urinary tract infection in febrile neutropenia, urine analysis (complete urinalysis and urine culture) should be taken. It can be considered that the pathogen growth limit of 105CFU/ml in midstream urine for urine culture positivity, recommended in the urinary tract infection protocol published by the American Academy of Pediatrics in 2011 and accepted in most laboratories, including our study, can be particularly high for neutropenic patients. Because the data in our study supports this, and if lower reproductive threshold values were taken, it can be thought that 10.1% febrile neutropenia attacks, which we consider as possible urinary tract infection, would decrease and definitive urinary tract infection cases would increase. A logistic regression model was created to evaluate the independent risks of different variables affecting the occurrence of urinary tract infection. The variables included in the model were patients' gender, diagnosis, absolute neutrophil counts, CRP values, urinary leukocyte-esterase, nitrite, leukocyturia, and dysuria. In the model, the risk of urinary tract infection with nitrite positivity in urine was found to be 15.27 times (95% CI 3.47-67.11) higher. In our study, we found that trimethoprim - sulfamethoxazole used in the prophylaxis of Pneumocystis Jiroveci in leukemias did not have a reducing effect on urinary tract infections, and that the breeding Escherichia coli were resistant to this antibiotic, according to the results of the urine culture study. Key words: Leukemia in children, febrile neutropenia in leukemia, urinary tract infection.