A study on clinical profile of invasive fungal infection in neutropenic children in a tertiary care teaching hospital

INTRODUCTION

Febrile neutropenia is a common complication of anti-cancer chemotherapy. With the development of more aggressive chemotherapy regimen for malignancies, the survival of cancer patients has improved. According to infectious disease society of America (IDSA), febrile neutropenia has been defined as “single oral temperature of 38.3^°C (101^°F) or a sustained oral temperature of 38^°C (100.4^°F) for more than one hour and absolute neutrophil count less than 500 /mm³ or a predicted decline to less than 500 /mm³ over the next 48 hours.¹ Majority of the patients with febrile neutropenia have no documented source of infection. According to IDSA, incidence of clinically documented infections during febrile neutropenic episodes is 20-30%. Bacterial infections occur more commonly during early days of neutropenia whereas invasive fungal infections are found mostly after prolonged neutropenia.² Factors that appear to be associated with invasive fungal infection (IFI) in both children and adults in this severely immunosuppressed population are the underlying malignancy, presence of prolonged neutropenia, intensive chemotherapeutic regimen.^3-5 Previous studies showed that early treatment of fungal infections leads to better prognosis in neutropenic patients. Therefore, early diagnosis and treatment of these infections are of crucial importance. Recognizing IFI is often difficult because of non-specific signs and symptoms and fever may be the only symptom. Knowledge of predisposing condition, likely organisms, risk factors, relevant clinical signs and symptoms of IFI is important for enabling early diagnosis of IFI and improving disease outcomes. In 2002 and 2008 the European organization for research and treatment of cancer and mycoses study group (EORTC/MSG) criteria were laid down to be used to maintain uniformity at international level.⁶ Candida species are most common cause of invasive fungal infections both in children and adults. Several studies have shown regional differences in the distribution of Candida species. However, there are limited studies regarding invasive fungal infections in children in India. Hence, there was a need to identify the common causative organisms of fungal sepsis in children in our population and our primary objective was to diagnose invasive fungal infections in children of febrile neutropenia as per revised EORTC criteria and identify the common fungal etiological agents and study the clinical profile in detail among these infections.

METHODS

This study was done as an observational, cross sectional study from November 2018 to April 2020. The study was conducted at the department of microbiology, Lady Hardinge Medical College and department of pediatrics, New Kalawati Saran Children Hospital. All children less than 18 years of age with absolute neutrophil count (ANC) <500 /ml, with fever not responding to antibiotics for more than 3-4 days and clinically suspected to have invasive fungal infection were included in the study. Patients who were above 18 years of age, with normal eosinophil count, with no history of fever were excluded. According to the formula,

The sample size has been calculated, considering a prevalence of 10% of febrile neutropenia and with a minimum absolute error of 10% around 55 was the sample size and 55 children who fulfilled our criteria were included. Prior approval from institutional ethics committee was taken. Subjects were enrolled only after obtaining written informed voluntary consent and voluntary assent from the child after explaining nature of study in detail in their own language.

Primarily the children will be enrolled on the basis of inclusion criteria, demographic information, clinical signs, risk factors, investigational data etc. will be recorded in a predesigned protocol.1-3 ml of blood was collected, blood culture was carried out by automated BacT/alert system and biphasic brain heart infusion media. Relevant samples depending on organ system collected include Tissue biopsy, skin scraping, broncho-alveolar lavage, pleural fluid, sputum, CSF, urine etc. Microscopic examination was performed by doing KOH wet mount, gram staining, Giemsa Staining and India Ink. Identification of bacterial and fungal isolates was done as per routine procedure. Antigen detection for Aspergillus, Cryptococcus and Pneumocystis was done wherever possible on need basis. DNA extraction was performed using a commercial kit and RT-PCR against universal fungal primer and species-specific primers for Candida was done. Commonly used molecular techniques include conventional, semi-nested and nested PCR, PCR-enzyme immunoassay, real-time PCR, multiplex PCR followed by DNA sequencing or pyrosequencing. Compared to routine yeast cultures that require more than 48 hours, this technique had a turnaround time of 9 hours. The sensitivity, specificity, positive predictive value and negative predictive value of this PCR assay for detection of candidemia were 95%, 95%, 90% and 97%, respectively.⁷ Data was entered and analysed on SPSS version 16. Appropriate test was applied for both qualitative and quantitative data, p<0.05 was considered statistically significant.

RESULTS

In our study which included 55 students, most children belonged to an age group of 3-9 years of which the majority (34.5%) were between 3 and 5 years of age followed by those between 5 to 7 years (18.18%) and 7 to 9 years (14.54%). The male:female ratio was 2.93:1, male patients in most age groups being considerably more than female patients.

Most patients (76.36%) were from the general wards and also hematology and oncology wards while 21.82% were from the PICU (pediatric intensive care unit) and 1.82% were from the NICU. Out of the 28 children under the age of 5 years, 15 (53.57%) children had moderate/severe malnutrition and 13 (46.42%) were normal weight for their age. Based on WHO growth charts for BMI, out of 27 children aged more than 5 years, thinness was present in 7 (25.92%) children, severe thinness in 1 (3.70%) patient while 19 (70.37%) children had normal BMI for age. The patients had a varied diagnosis, B-cell ALL (acute lymphoblastic leukemia) (67.27%), followed by T-cell ALL (18.18%), Aplastic anemia and ALCL (anaplastic large cell lymphoma) in 12.73% and 1.82% patients respectively. Hospitalization ≥7 days (100%) was the most common associated risk factor, followed by neutropenia (100%), immunodeficiency (100%), cytotoxic drugs (87.27%) and malignancies (87.27%). Other less commonly associated risk factors were hyperalimentation (23.64%), residence near a construction site (23.64%), presence of urinary catheters (10.91%) and HIV positivity (9.09%). None of the patients had any association with diabetes mellitus or burns.

The most common presenting complaint was fever in 54 patients (98.18%). Cough (72.73%) was the second most common complaint followed by coryza (29.09%). Vomiting and nasal/ear/oral discharge were noted in 9.09% followed by abdominal pain (7.27%), breathlessness (5.45%) and burning micturition (3.64%) while headache and loose stools were seen in 1.82% patients. Around 48 (87.27%) patients had taken cytotoxic drugs/steroids followed by 18 (32.73%) patients who had received antimetabolites. 17 (35.41%) patients were in induction phase of chemotherapy followed by 21 (43.75%) in consolidation phase and 10 (20.88%) in the maintenance phase. All 7 (12.72%) patients with severe aplastic anemia received cyclosporine and 3 (5.45%) out of them received Antithymocyte globulin (ATG). Neutropenia was the most common hematological finding seen in 100% patients, followed by anemia seen in 81.81% and thrombocytopenia in 54.54% patients. Leukocytosis and leukopenia were seen in 18.18% and 9.09% patients respectively. Among the biochemical parameters, hypernatremia was seen in 3.64% and potassium levels were increased in 7.41% of patients. Urea was increased in 7.41%.

Chest X-ray was the most common radiologic investigation done in all patients. In these atients, B/L infiltrates (49.09%) was the most common finding, followed by non-homogenous opacities (16.36%), consolidations (7.27%) and cystic lesions (3.64%) diffuse hyperintensity, pleural effusion and reticular shadows were seen in (1.82%) each. 10 (18.18%) patients had normal findings.

Ultrasound examination of the abdomen was done in 22 patients. In these patients, hepatomegaly (31.18%) was the most common finding, followed by hepatosplenomegaly (27.27%), liver abscess and neutropenic enterocolitis was seen in 4.54% patients each and 5 (22.72%) patients had normal findings. Ultrasound chest showed pericardial and pleural effusion in 4.54% patients each. On CT, Halo Sign was the most common finding noted in 5 (29.41%) patients followed by lobar consolidation in 2 (11.76%) patients and Ground glass opacities, pleural effusion and tree in bud appearance in 1 (5.88%) patient each. CT/MRI brain showed cerebral granuloma in 2 (11.76%) patients. Blocked osteomeatal complex and Pansinusitis was noted in CT/MRI paranasal sinus (PNS) in 1 (5.88%) patient each. The other findings in CT/MRI abdomen were hepatic and subcapsular abscess (11.76%) and neutropenic enterocolitis (5.88%).

Maximum positivity of fungal elements was seen in sputum samples in both microscopy (54.54%) and culture (54.54%) followed by gastric aspirate having a positivity rate of 56.52% in microscopy and 52.17% in culture. Oral swab, urine and ET aspirate were other samples having high fungal positivity.

Around 54.54% of sputum samples were positive for fungal culture. A lower positivity was seen in blood culture (14.54%). Oral swabs, gastric aspirate and urine showed a positivity of 80%, 52.17% and 20% each. BAL, ET aspirate, central line, ear swab, throat swab and stool samples all were positive in 1 sample each. Overall, C. albicans (35.29%) and C. tropicalis (29.41%) were the most common fungal isolates. C. glabrata (9.80%), C. parapsilosis (9.80%) and A. flavus (7.84%) were the other common isolates while A. fumigatus (3.92%), C. krusei (1.96%), and Penicillium spp. (1.96%) were found in lesser proportion respectively. As far as the bacterial culture was concerned, of 55 blood samples, 2 (3.63%) yielded Acinetobacterbaumanii, and two yielded coagulase negative Staphylococcus. Klebsiella spp. (1.81%) Pseudomonasspp., and Methicillin resistant staphylococcus aureus grew in one patient each. Out of 33 sputum samples, one grew Acinetobacterbaumanii. Out of 23 patients in whom gastric aspirate culture was done, one grew Klebsiella spp. and one grew Pseudomonas spp. Out of 15 urine samples, one yielded coagulase negative Staphylococcus. Pneumocystis jirovecii was detected by direct immunofluorescence testing in sputum of 13 (32.50%) patients out of 40 suspected patients for pneumocystis infection. Cryptococcal antigen was detected by latex agglutination testing in CSF of one (25.00%) patient out of 4 patients suspected for cryptococcosis and galactomannan antigen testing for Aspergillus could only be done in one patient and was positive for the same. Invasive fungal infections were diagnosed in 47 (85.45%) patients based on the revision and update of the consensus definitions of invasive fungal disease from the EORTC/MSG consensus group, 2019. A total of 23 (48.93%) proven, 3 (6.38%) probable and 21 (44.68%) possible fungal infections were diagnosed.

DISCUSSION

The present study was a prospective observational study, the majority of patients in our study population were between 3-5 years and the picture was similar to the studies from abroad (Ozdemir et al and Prasad et al significant male preponderance was noted in the present study which was similar to the studies by (Ozdemir et al, Bakhshi et al).^8-10 Maybe because male children are still seen to seek better or more healthcare compared to females in many of our societies. Among the 55 patients 37 (67.27%) had B cell acute lymphoblastic leukemia, 10 (18.18%) had T cell ALL, 7 (12.73%) had severe aplastic anemia and 1 (1.82%) had anaplastic large cell lymphoma. The study by Ozdemir et al included children with both ALL and AML while study by Bakhshi et al included children with ALL only.^5,11 In the present study, only 25.92% had ‘thinness’ while 3.70% were found to be ‘severely thin.’ This is similar to the observations made by Linga et al. (186) (50%), Jain et al (56.8%) and Kumar et al (52%).^11,12 This reflects high prevalence of malnutrition among children in general. Fever (98.18%) was the most common symptom in our study, followed by cough (72.73%) and coryza (29.09%). In addition to these children in our study also presented with vomiting (9.09%), nasal/ear/oral discharge (9.09%), abdominal pain (7.27%), breathlessness (5.45%), burning micturition (3.64%), loose stools (1.82%) and headache (1.82%). Duration of fever >7 days, breathlessness and burning micturition were found to be statistically significant in the present study (p<0.05). Similarly, a study in 2013 of 132 cases of neonatal candidemia by Juyal et al it was noted that failure to thrive (71.21%), feed intolerance (65.91%), respiratory distress (63.64%), abdominal distension (59.09%), lethargy (55.30%), poor perfusion (28.79%) and convulsions (20.45%) were the common clinical presentations in these patients.¹³

Neutropenia, prolonged hospitalization, indwelling urinary catheters, hyperalimentation, immunosuppressive or cytotoxic therapy, immunodeficiency, malignancy, presence of venous access device in situ are some of the well-known risk factors associated with invasive fungal infections. When we analysed the duration of stay in hospital, hospitalization ≥7 days was the most common and statistically significant risk factor (p=0.011) in comparison with lesser days of stay and was seen in all our patients. 13 (23.64%) of them were ICU patients with ICU stay ≥5 days. In study done by Nucci et al¹⁴. candidemia at the pediatric ICU of a tertiary care hospital, they found that 40.9% of the patients who developed candidemia had a PICU stay >15 days.¹⁵ Most UTI due to Candida species occur in hospitalized patients with indwelling urinary catheters. In the present study 10.91% of patients had indwelling urinary catheters and it was found to be a statistically significant risk factor (p=0.002). Hyperalimentation or total parenteral nutrition is another risk factor associated with invasive fungal infection. This along with the use of broad- spectrum antibiotics leads to overgrowth of Candida that results in colonization and can lead to infection. Hyperalimentation was present in 23.64% of the patients and was found to have a statistically significant association with IFI. (p=0.022). This higher incidence was also noted in the Australian study by Blyth et al.¹⁶ Broad spectrum antibiotics are another well-known risk factor for invasive Candida infections. All the patients in the present study had been treated with multiple antibiotics. Moreover, the duration of antibiotics >14 days used also has an influence on the development of candidemia and was found to be statistically significant in our study. (p=0.033). Feja et al concluded that the use of all these antibiotics was found to be a significant risk factor for candidemia by univariate analysis.¹⁷ In our study, 17 (35.41%) were in the induction phase, 21 (43.75%) were in the consolidation phase and 10 (20.88%) were in the maintenance phase of chemotherapy. However, this was different from the study by Tufekci et al where the number of patients was higher in the consolidation phase.¹⁸ The duration (≥7 days) and severity (<100 ANC) of neutropenia directly correlates with the total incidence of all infections and of those infections that are life threatening and both were found to be statistically significant in our study (p<0.05). This was similar to a study by Lin et al where prolonged neutropenia was significantly associated with invasive fungal infections.¹⁹

Neutropenia (100%) was the most common hematological abnormality in the patients in our study followed by anemia (81.82%) which is more likely to be related to the overall poor nutritional status of the children in the study rather than being directly connected to invasive fungal infection. Lymphocytosis (18.18%) and leukopenia (9.09%) were found to have a statistically significant association (p≤0.05).

In our study increased blood urea levels were seen in 7.41%, while serum creatinine was raised in 1.85%. Lin et al reported that increased creatinine values to be associated with invasive fungal infection.¹⁹ This can be explained by a probable underlying renal disorder (p=0.014). Chest X-ray was the most common radiologic investigation and it was done in all patients. B/L infiltrates (49.09%) was the most common finding. Non-homogenous opacities were seen in (16.36%) of patients followed by consolidations (7.27%) and cystic lesions (3.64%). Diffuse hyperintensity, pleural effusion and reticular shadows were seen in (1.82%) patients each. 10 (18.18%) patients had normal findings. Consolidation was found to have a statistically significant association with IFI (p=0.008). Disseminated candidiasis can lead to fungal abscesses formation in the CNS, kidneys, liver, spleen, skin, bowel, and peritoneum. In our study, liver abscess was seen in 4.54% patients. Hepatomegaly (31.18%) was the most common finding, followed by hepatosplenomegaly (27.27%) and neutropenic enterocolitis in 4.54% patients. CT/MRI of chest was done in 17 patients, of which majority (29.41%) had Halo sign followed by consolidation seen in 11.76% patients. Other findings were ground glass opacities, pleural effusion and tree in bud appearance seen in 5.88% each. This was similar to a study by Gulhan et al where Halo sign was the most common finding followed by Lobar consolidation.²⁰ Maximum positivity of fungal elements was seen in sputum samples in both microscopy (54.54%) and culture (54.54%) followed by gastric aspirate which is difficult to corroborate for pathogenic role as Candida is a common organism grown in sputum. Overall, C. albicans (35.29%) and C. tropicalis (29.41%) were the most common fungal isolates in comparison to other species. We observed that Gram negative organisms were more commonly isolated (63.63%) than gram positive (36.36%) organisms. A similar pattern of isolation of predominantly gram-negative organisms was observed in studies from other tertiary cancer centres in India. Out of 55 patients in the study, candidemia confirmed by blood culture was seen in 8(14.54%) patients. All the candidemia strains, C. tropicalis (10.90%), C. parapsilosis (1.81%) and C. albicans (1.81%) isolated were found to be sensitive to fluconazole, voriconazole, amphotericin B, flucytosine, caspofungin and micafungin. Panfungal and Candida multiplex RT-qPCR identified 10 samples as positive for Candida DNA (18.18%), comprising C. tropicalis (N=4), C. parapsilosis (N=3) and C. albicans (N=2) and a mixed infection of C. tropicalis and C. parapsilosis in one sample. Candida multiplex PCR detected five blood culturepositive samples out of eight (growing Candida spp.). C. tropicalis (40%) was the most common Candida bloodstream isolated on PCR followed by C. parapsilosis (30%) and C. albicans (20%). Mixed infection (C. parapsilosis and C. tropicalis) was seen in 10% of patients. This is similar to a study on nosocomial pediatric candidemia conducted by Awasthi et al where among 323 cases of septicemia blood for fungal culture was positive in 7.4% of the patients.²¹ The study also reported a higher prevalence of candidemia due to C. tropicalis (71.5%) among pediatric patients. Different candida species (other than C. albicans) are being isolated over the years and in different places and their role in pathogenicity needs to be further studied upon. Pneumocystis antigen was detected in 13 out of 40 patients, Cryptococcus antigen was detected in 1 out of 4 patients and Galactomannan antigen of Aspergillus was done only in 1 (100%) patient and was positive for the same. Wang et al had included 33 patients and their baseline lumbar punctures were performed from which CSF cryptococcal antigen test results were positive in all 20 patients.²² The high incidence of PCP infection found in the present study, suggested a local outbreak of infection which was a focus of study indicates to identify high-risk patients and timely intervention in improving outcome.

Limitations

Limitations of current study were; relatively smaller sample size which was on account of limited duration during which the study was required to be completed.

CONCLUSION

Invasive fungal infections increasingly have been recognized as a common threat in patients with febrile neutropenia especially in the setting of prolonged neutropenia. Candida spp., Aspergillus spp. and Pneumocystis remain the primary fungal pathogens in this patient population. Keeping in mind the non-specific clinical presentation of invasive fungal infections, a high index of clinical suspicion is needed in all ill patients with sepsis that is not responsive to antibacterial therapy. Early clinical suspicion followed by prompt sampling and evaluation for fungal infections may aid in timely diagnosis and reduction of mortality of these patients.

Funding: No funding sources

Conflict of interest: None declared

Ethical approval: The study was approved by the Institutional Ethics Committee

REFERENCES

1. 1. Freifeld AG, Bow EJ, Sepkowitz KA, Boeckh MJ, Ito JI, Mullen CA, et al. 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(4):e56-93.
2. 2. Freifeld AG, Bow EJ, Sepkowitz KA. 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(4):e56-93.
3. 3. Creutzig U, Zimmermann M, Reinhardt D, Dworzak M, Stary J, Lehrnbecher T. Early deaths and treatment-related mortality in children undergoing therapy for acute myeloid leukemia: analysis of the multicenter clinical trials AML-BFM 93 and AML-BFM 98. J Clin Oncol. 2004;22(21):4384-93.
4. 4. Basu SK, Fernandez ID, Fisher SG, Asselin BL, Lyman GH. Length of stay and mortality associated with febrile neutropenia among children with cancer. J Clin Oncol. 2005;23(31):7958-66.
5. 5. Hakim H, Flynn PM, Knapp KM, Srivastava DK, Gaur AH. Etiology and clinical course of febrile neutropenia in children with cancer. J Pediatr Hematol Oncol. 2009;31(9):623-9.
6. 6. Hoenigl M, Strenger V, Buzina W, Valentin T, Koidl C, Wölfler A, et al. European organization for the research and treatment of cancer/mycoses study group (EORTC/MSG) host factors and invasive fungal infections in patients with haematological malignancies. J Antimicrob Chemother. 2012;67(8): 2029-33.
7. 7. Giusiano G, Mangiaterra M, Saito VG, Rojas F, Gómez V, Díaz MC. Etiology of fungaemia and catheter colonisation in Argentinean paediatric patients. Mycoses. 2006;49(1):49-54.
8. 8. Özdemir N, Tüysüz G, Çelik N, Yantri L, Erginöz E, Apak H, et al. Febrile neutropenia in children with acute lymphoblastic leukemia: single center experience. Turk Arch Pediatr Arş. 2016;51(2):79-86
9. 9. Prasad M, Chinnaswamy G, Arora B, Vora T, Hawaldar R, Banavali S. Risk predictors for adverse outcome in pediatric febrile neutropenia: Single center experience from a low and middle-income country. Indian J Cancer. 2014;51(4):432-7.
10. 10. Bakhshi S, Padmanjali KS, Arya LS. Infections in childhood acute lymphoblastic leukemia: an analysis of 222 febrile neutropenic episodes. Pediatr Hematol Oncol. 2008;25(5):385-92.
11. 11. Jain V, Dubey AP, Gupta SK. Nutritional parameters in children with malignancy. Indian Pediatr. 2003;40: 976-84.
12. 12. Kumar R, Marwaha RK, Bhalla AK, Gulati M. Protein energey malnutrition and skeletal muscle wasting in childhood acute lymphoblastic lukemia. Indian Pediatr. 2000;37(7):720-6.
13. 13. Juyal D, Sharma M, Pal S, Rathaur VK, Sharma N. Emergence of non-albicans Candida species in neonatal candidemia. North Am J Med Sci. 2013;5(9): 541-5.
14. 14. Nucci M, Barreiros G, Guimarães LF, Deriquehem VAS, Castiñeiras AC, Nouér SA. Increased incidence of candidemia in a tertiary care hospital with the COVID-19 pandemic. Mycoses. 2021;64(2):152-6.
15. 15. Becerra MR, Tantaleán JA, Suárez VJ, Alvarado MC, Candela JL, Urcia FC. Epidemiologic surveillance of nosocomial infections in a pediatric intensive care unit of a developing country. BMC Pediatr. 2010;10: 66.
16. 16. Blyth CC, Chen SC, Slavin MA, Serena C, Nguyen Q, Marriott D, et al. Not just little adults: candidemia epidemiology, molecular characterization, and antifungal susceptibility in neonatal and pediatric patients. Pediatrics. 2009;123(5):1360-8.
17. 17. Feja KN, Wu F, Roberts K, Loughrey M, Nesin M, Larson E, et al. Risk factors for candidemia in critically ill infants. J Pediatr. 2005;147(2):156-61.
18. 18. Tüfekçi Ö, Yılmaz Bengoa Ş, Demir Yenigürbüz F, Şimşek E, Karapınar TH, İrken G, et al. Management of invasive fungal infections in pediatric acute leukemia and the appropriate time for restarting chemotherapy. Turk J Hematol. 2015;32(4):329-37.
19. 19. Lin GL, Chang HH, Lu CY. Clinical characteristics and outcome of invasive fungal infections in pediatric acute myeloid leukemia patients in a medical center in Taiwan. J Microbiol Immunol Infect. 2016;51(2):251-9.
20. 20. Gülhan B, Kanik-Yuksek S, Ozkaya PA, Yarali N, Tezer H, Özbek N. Invasive fungal infection in children with hematologic malignancy. Turk J Pediatr. 2019;61:159.
21. 21. Awasthi AK, Jain A, Awasthi S, Ambast A, Singh K, Mishra V. Epidemiology and microbiology of nosocomial pediatric candidemia at a northern Indian tertiary care hospital. Mycopathologia. 2011;172(4): 269-77.
22. 22. Wang R, Chen Y, Wu J, Wang X, Cao Y, Zhao H, et al. Cryptococcosis in patients with hematological diseases: a 14-year retrospective clinical analysis in a Chinese tertiary hospital. BMC Infect Dis. 2017; 17(1):463.