SciELO - Scientific Electronic Library Online

vol.90 número2Desarrollo de habilidades comunicativas en lactantes con síndrome de Down posterior a capacitaciones sistematizadas en comunicación gestualSíndrome de Klippel-Feil autosómico dominante: Una malformación de segmentación vertebral índice de autoresíndice de materiabúsqueda de artículos
Home Pagelista alfabética de revistas  

Servicios Personalizados




Links relacionados

  • En proceso de indezaciónCitado por Google
  • No hay articulos similaresSimilares en SciELO
  • En proceso de indezaciónSimilares en Google


Revista chilena de pediatría

versión impresa ISSN 0370-4106

Rev. chil. pediatr. vol.90 no.2 Santiago abr. 2019 


Risk factors and lethality associated with Neonatal Candidemia in a neonatal unit

Elizabeth Caparó Ingram1 

Manuel Vásquez Vega1 

Ximena Norero1 

Xavier Sáez-Llorens1 

Rodrigo DeAntonio2 

Edgar Rodríguez Barría3 

1 Hospital del Niño Dr. José Renán Esquivel, Panamá.

2 Instituto Nacional de Salud Pública, México.

3 Hospital Dr. Manuel Nieto, Darién, Panamá.



To identify the main risk factors associated with neonatal candidemia.

Patients and Method:

A retrospective paired case-control study was conducted from January 2014 to December 2016. The cases were patients with isolation in blood culture and/or cerebrospinal fluid of Candi da spp. after their first 48 hours in the hospital and the controls cases were neonates chosen from the statistical census of neonatology paired according to their admission date (30-day range), birth weight, gestational age, and discharge condition (alive or deceased). For each case, we select two controls. The risk factors evaluated were intrahospital stay over seven days, use of broad-spectrum antibiotics, mechanical ventilation, parenteral nutrition longer than five days, invasive procedures such as central venous access and abdominal and thoracic surgeries, necrotizing enterocolitis and growth of bacterial microorganisms in blood culture before candidemia.


During the study period, 141 patients developed candidemia. 49% of the cases corresponded to Candida parapsilosis with the highest associated lethality rate. The multivariate analysis identified as risk factors hospital stay longer than seven days (OR = 17.0, 95% CI = 2.36-122.4), use of umbilical lines (OR = 9.04, 95% CI = 1.55-52.5), abdominal and/or thoracic surgery (OR = 12.4, 95% CI = 1.76-87.3), and treatment with Meropenem (OR = 4.62, 95% CI = 1.34-15.9).


Prolonged intrahospital stay longer than seven days and thoracic and/or abdominal surgery were the most significant risk factors in this study for the development of neonatal candidemia.

Keywords: Neonate; Candidemia; Risk factors; Fluconazole; Panama


Candida infections are currently associated with increased mortality in the neonatal population; thusit is necessary to evaluate the main risk factors associated with this infectious process.

Patients in neonatal intensive care units (NICU) are colonized by Candida species shortly after birth. Colonization is found mainly in the gastrointestinal and respiratory tracts in the first two weeks of life due to vertical transmission associated with childbirth. After two weeks of life, the most frequent coloniza tion site is the skin, this can be related to the manipu lation of the patient by health personnel1,2.

Risk factors for candidemia in premature patients are associated with gastrointestinal tract immaturity, colonizing strain virulence, immunological deficien cies and immaturity, and incomplete epidermis development that weakens physical and immunological ba rriers facilitating invasion. After having access to mucous membranes or the bloodstream, Candida shows a predilection for the invasion of the central nervous system, kidneys, liver, spleen, heart, and retina, with complications compromising the function of these organs3,4.

Candida is the main cause of invasive fungal in fections in the NICU and is the third most commonly isolated microorganism in blood cultures in late neo natal sepsis. It is estimated to cause 2-4% of infections in early neonatal sepsis and 10-12% in late neonatal sepsis5.

Neonatal candidiasis significantly increases mor tality and morbidity, with a mortality rate between 15 and 60%, which is higher in extremely low-birth- weight (ELBW) infants (less than and equal to 1,000 grams)5.

The objective of this study was to determine the main risk factors associated with candidemia in neona tes seen at the Hospital del Niño Dr. José Renán Esquivel de Panamá (DRJRE).

Patients and Method

A retrospective paired case-control study was con ducted between January 2014 and December 2016. Due to the previously known high prevalence of candidemia in the institution’s neonatal unit, a period of three years was chosen for this study.

The cases were all neonates who had blood culture and/or cerebrospinal fluid with isolation of Candida after 48 hours of hospitalization in the neonatal unit of the Hospital del Niño DRJRE, admitted during the stu dy period. Controls were taken from the statistical cen sus of the neonatal unit matched by admission date, birth weight, gestational age, and discharge condition. For multiple pregnancies, only one of the children was selected as the control group. Clinical records were then evaluated in the archives of Hospital del Niño DRJRE.

The sample was calculated using OpenEpi program having a 95% confidence level, 80% detection probability, with two controls per case, a hypothetical propor tion of controls with exposure of 4%, a hypothetical proportion of cases with exposure of 13%, and odds ratios of 3.59 obtaining a minimum size of 76 cases according to Kelsey and 89 according to Fleiss, and a minimum size of 152 control samples according to Kelsey and 178 according to Fleiss, making a total of 267 patients.

Reference population and study place

The study was conducted in the neonatal unit of the Hospital del Niño DRJRE, with capacity for 90 patients (two intensive care units for 14 patients each, three intermediate care units for 20 to 24 patients each, and an isolation unit for two patients), to which 6,755 neonates were admitted during the study period.

The extremely low-birth-weight (ELBW) is de fined as a newborn weighing less than 1,000 grams, the very-low-birth-weight (VLBW) as a newborn weighing less than 1,500 grams, and the low birth weight (LBW) as a newborn weighing less than 2,500 grams6.

During the study, all patients under 1,250 grams received a dose of fluconazole prophylaxis at 3mg/kg using the intravenous route every 48 hours, the first dose given on the third day of life and continuing prophylaxis for six weeks. This protocol has been applied in the institution since 2009.7

Microbiological Method

Blood specimens were inoculated into BacT/AlerT PF Plus pediatric aerobic bottles (Biomérieux, Fran ce), then incubated in the BacT/AlerT 3D automa ted system for five days. The blood volume extracted for blood culture was determined according to the weight of the patient: 1 ml for weights less than or equal to 1000 grams, 2 ml for weights from 1001 to 3000 grams, and 4 ml for weights more than 3001 grams. Bottles that were positive were Gram stained, and each sample with microscopic yeasts eviden ce was subcultured in sabouraud agar medium and incubated at 35 +/- 2 ° C for 24 hours. Phenotypic identification and susceptibility testing for each iso lation were performed using the Vitek®2 system (Bio mérieux, France) with YST, AST and YBC cards. The cerebrospinal fluid sample was inoculated in thio- glycolate broth and incubated at 37°C for 72 hours. As yeasts grew in the solid medium, the identification process and susceptibility testing were carried out in the same way as blood culture.

Analysis plan

Initially, a descriptive analysis of the cases and con trols characteristics was carried out. Variables propo sed as risk factors in the literature were selected, and the presence or absence of these associations was established in the neonatal unit. Through the comparison between cases and controls, a univariate analysis was performed estimating the Odds Ratio (OR) as the ex tent to which these risk factors predispose patients to develop candidemia, defining as statistically significant a p-value < 0.05. All variables with a p-value < 0.05 were selected for inclusion in the multivariate logistic regression model. The analysis was performed using the statistical software STATA 14.

The development of this study is guided by the in ternational ethical guidelines for epidemiological stu dies developed by the World Health Organization (Fe bruary 2008)8. The study was approved by the bioethics committee of the institution and by the bioethics com mittee of the Ministry of Health. Written permission was obtained from both committees to review clinical records.


A total of 6755 newborns were admitted to the neonatal unit from January 2014 to December 2016 of which 2801 (41.5%) were admitted to the NICU and 3954 (58.5%) to the intermediate care unit.

A total of 836 positive blood cultures were reported, of which 328 episodes (39.2%) were Gram-negative microorganisms, 263 (31.5%) fungi, and 245 (29.3%) Gram-positive ones. Out of the 263 blood cultures reported with fungal growth, 141 patients were identified with episodes of candidemia, however, seven patients were excluded as they presented Candida growth prior to 48 hours of hospitalization, making a total of 134 patients with candidemia in the group of cases. Three patients (2%) presented Candida isolation in cerebrospinal fluid. The control group consisted of 268 patients.

The candidemia incidence was 20 per 1000 admis sions, with the highest rate in ELBW patients (406.6 per 1000 ELBW neonates). The increase was from 15 per 1000 admissions in 2014 to 30 per 1000 admissions in 2015, having a decrease by 2016 to 18 per 1000 ad missions.

When evaluating the persistence of candidemia (defined as blood culture by Candida six days after starting antifungal therapy)5 versus non-persistence, it was found that there was no relationship between persistence and mortality, which was 59% in persis tent candidemia versus 63% in the non-persistent one.

General characteristics of cases and controls

(Table 1) shows the gender, gestational age, birth weight, and discharge status characteristics which were similar in cases and controls.

Table 1 Clinical characteristics of cases and controls. 

Out of 273 infants with ELBW and VLBW (< 1500 g), 91 (33%) developed candidemia, and of 68 infants with LBW (< 2500 g), 24 (35%) developed candidemia. During the study, 191 (100%) of patients under 1250 grams received fluconazole prophylaxis, developing candidemia in 80 patients (42%).

An overall 62% candidemia fatality rate was found. (Table 2) describes the fatality by species.

Table 2 Fatality rate of candidemia by species. 

Candida parapsilosis caused 66 (49%) episodes of candidemia, second is Candida albicans with 44 (33%) episodes, followed by Candida tropicalis with 11 (8%) episodes, Candida with 7 (5%) episodes, other species caused 9 (7%) episodes, and Candida guilliermondii, Candida haemulonii and Candida lusitaniae caused 3 (2%), 2 (1.4%), 2 (1.4%) episodes respectively.

Univariate Analysis

Initially, the variables were evaluated indepen dently, highlighting possible risk factors for can didemia on an individual basis (Table 3). When performing the univariate paired analysis, it was observed that the same variables remained as risk factors, statistically significant for candidemia, identifying inpatient stay longer than seven days, the use of broad-spectrum antibiotics, admission to NICU, and exposure to mechanical ventilation, the use of parenteral nutrition for more than five days, inva sive procedures such as placement of central venous access and abdominal and thoracic surgeries, necro tizing enterocolitis, and bacterial microorganism growth in blood culture prior to candidemia. Out of the statistically significant risk factors, those that represent a higher risk are length of stay longer than seven days, use of parenteral nutrition for more than five days, and exposure to abdominal surgeries (Ta ble 4).

Table 3 Risk Factors for Candidemia in neonates hospitalized in the Hospital del Niño, January 2014 to December 2016. Unpaired Bivariate Analysis. 

Table 4 Risk Factors for Candidemia in neonates hospitalized in the Hospital del Niño, January 2014 to December 2016. Matched Bivariate Analysis. 

Multivariate Analysis

In the multivariate analysis, risk factors identified were inpatient stay longer than seven days, the use of umbilical catheters, surgery, and treatment with meropenem (Table 5).

Table 5 Risk Factors for Candidemia in neonates hospitalized in the Hospital del Niño, January 2014 to December 2016. Matched Multivariate Analysis. 


The study was able to evaluate the most important risk factors associated with candidemia in patients hospitalized in the neonatal unit: inpatient stay longer than seven days, the use of umbilical catheters, sur gery, and treatment with meropenem. Although there are limitations in assessing risk factors, we decided to conduct a paired case-control study to reduce possible bias.

The hospital incidence rate of Candida in the stu dy was 20 per 1000 admissions, which is higher when comparing with other studies such as Feja et al.6 in New York, Fu et al.9 and Chen et al.10 in China, and Pinhat et al.11 in Brazil, being 15.9, 13.6, 10.6 and 9.0 per 1000 admissions respectively. Among the possible reasons associated with the high incidence of candi- demia in our institution, we can mention daily over crowding (census above capacity), the small number of doctors and nurses (numerical ratio per patient), the pathologies high complexity, to be a national reference center, and the indiscriminate use of broad-spectrum antibiotics. According to the obtained results, we re commend reducing the use of broad-spectrum anti biotics, especially carbapenem antibiotics, improving aseptic techniques when placing and using umbilical catheters, and trying as much as possible to reduce the inpatient stay to reduce the incidence of candidemia in the institution.

A total of 49% of candidemia cases were caused by Candida parapsilosis similar to that exposed in the studies of Yu et al.12, Rodríguez et al.13, Neu et al.14, Celebi et al.15, and Caggiano et al.16. In contrast, other studies such as Feja et al.6, J. Fu et al.9, Chen et al.10, Saiman et al.17, Robinson et al.18, Juyal et al.19, Hsu et al.20, and Márquez et al.21, the predominant species was Candida albicans.

The overall mortality rate in the neonatal unit of our institution, in patients weighing less than 2500 grams, is 10% (source: neonatal database). During the study period, the overall fatality rate due to candide mia was 62%. Candida parapsilosis was responsible for the highest number of episodes and fatality due to candidemia.

The Candida parapsilosis virulence factors include biochemical characteristics and metabolic processes such as the production of phospholipases, the synthe sis and storage of fatty acids, great affinity for synthetic surfaces (greater than Candida albicans), and differen ces in the structure and composition of biofilms, which contribute to the increase in the number of cases ob served in recent years22,23,24. In addition, lower susceptibility to different antifungals, such as amphotericin B, fluconazole, and caspofungin, along with the emergen ce of multi-resistant strains, contributes to the higher incidence of this species compared to other species of Candida25.

No relationship was found in the study between persistence of candidemia and mortality, similar to that obtained by Robinson et al.18, Levy et al.26, and Benjamin et al.21.

Multivariate analysis identified the following risk factors: inpatient stay, use of umbilical catheters, sur gery, and treatment with meropenem. These risk fac tors appear as statistically significant in the literature. Surgical intervention as a risk factor has been pre viously mentioned in the study by Ávila-Agüero et al.28) based on the skin and epithelial barrier disruption for Candida invasion of the bloodstream. Inpatient stay and the use of meropenem have been identified as risk factors in studies by Fu et al.9, Chen et al.10, Yu, et al.12, Saiman et al.17, Stoll et al.29, Benjamin et al.30, and Hsu et al.31. It has been found that carbapenem antibiotics (Meropenem) may contribute to the proliferation of opportunistic microorganisms such as Candida by reducing the competitive pressure exerted by normal bacterial flora.9 A study by Kaufman et al.32 showed that decreased use of carbapenem antibiotics is asso ciated with decreased incidence of Candida infections.

In the study, the main risk factor was inpatient stay since it enables Candida to colonize patients. These results coincide with those found in other neonatal units33,34.

In addition, the umbilical catheters placement was identified as a risk factor for the development of candidemia since it facilitates the Candida introduction and attachment to foreign material, therefore, it forms bio films that lead to its horizontal transmission35,36,37.

It is noteworthy to highlight the high incidence of candidemia observed in patients weighing less than 1250 grams despite fluconazole prophylaxis. Possible explanations for this finding include a high degree of horizontal transmission due to a failure to comply with biosecurity standards, overcrowding in the neo natal units, and complexity of the neonatal population in the region. Future studies are needed to assess these possibilities.

The strengths of the study include adequate selec tion and definition of cases and controls, good access to clinical records, and control of biases during statis tical analysis.

The importance of identifying factors that increase the likelihood of developing candidemia is highlighted when considering the impact they have on morbidity and mortality in the neonatal population. It is there fore mandatory to make a timely diagnosis and start treatment early.

Ethical Responsibilities

Human Beings and animals protection: Disclosure the authors state that the procedures were followed according to the Declaration of Helsinki and the World Medical Association regarding human experimenta tion developed for the medical community.

Data confidentiality: The authors state that they have followed the protocols of their Center and Local regulations on the publication of patient data.

Rights to privacy and informed consent: The authors have obtained the informed consent of the patients and/or subjects referred to in the article. This docu ment is in the possession of the correspondence author.

Financial Disclosure: Authors state that no economic support has been asso ciated with the present study.

Conflicts of Interest: The authors declare not having any conflict of inter ests. Rodrigo DeAntonio is an employee of GSK Vacci nes. Own opinions are expressed in this article and do not reflect opinions or views from GSK Vaccines.

Aknowledgments: Julio Nieto Ramos for providing the statistics of re ported candidemia cases in the neonatal nursery ran ging from January 2014 until December 2016. Dr. Joel Méndez for the support with the analysis of the case study results.


1. Waggoner-Fountain L, Walker M, Hollis R, et al. Vertical and horizontal transmission of unique Candida species to premature newborns. Clin Infect Dis. 1996;22(5):803-8. [ Links ]

2. Bendel C. Colonization and epithelial adhesion in the pathogenesis of neonatal candidiasis. Semin Perinatol. 2003;27(5):357-64. [ Links ]

3. Chapman R, Faix R. Persistently positive cultures and outcome in invasive neonatal candidiasis. Pediatr Infect Dis J. 2000;19(9):822-7. [ Links ]

4. Benjamin D Jr, Poole C, Steinbach W, et al. Neonatal candidemia and end- organ damage: a critical appraisal of the literature using meta-analytic techniques. Pediatrics. 2003;112(3 Pt1):634-40. [ Links ]

5. Hammoud M, Al-Tajar A, Fouad M, et al. Persistent candidemia in neonatal care units: risk factors and clinical significance. Int J Infect Dis. 2013;17(8):624-8. [ Links ]

6. Feja K, Wu F, Roberts K, et al. Risk factors for candidemia in critically ill infants: a matched case-control study. J Pediatr. 2005;147(2):156-61. [ Links ]

7. Rueda K, Moreno M, Espinosa M, et al. Impact of routine fluconazole prophylaxis for premature infants with birth weights of less than 1250 grams in a developing country. Pediatr Infect Dis J. 2010;29(11):1050-2. [ Links ]

8. CIOMS. International Ethical Guidelines for Epidemiological Studies - CIOMS: Geneva, February 2008. [ Links ]

9. Fu J, Wang X, Wei B, et al. Risk factors and clinical analysis of candidemia in very-low-birth-weight neonates. Am J Infect Control. 2016;44(11):1321-5. [ Links ]

10. Chen J, Jiang Y, Wei B, et al. Epidemiology of and risk factors for neonatal candidemia at a tertiary care hospital in western China. BMC Infect Dis. 2016;16:700. [ Links ]

11. Pinhat E, Borba M, Ferreira M, et al. Fungal colonization in newborn babies of very low birth weight: a cohort study. J Pediatr (Rio J). 2012;88(3):211-6. [ Links ]

12. Yu Y, Du L, Yuan T, et al. Risk factors and clinical analysis for invasive fungal infection in neonatal intensive care unit patients. Am J Perinatol. 2013;30(07):589-94. [ Links ]

13. Rodríguez D, Almirante B, Park B, et al. Candidemia in neonatal intensive care units: Barcelona, Spain. Pediatr Infect Dis J. 2006;25(3):224-9. [ Links ]

14. Neu N, Malik M, Lunding A, et al. Epidemiology of candidemia at a children’s hospital, 2002 to 2006. Pediatr Infect Dis J. 2009;28(9):806-9. [ Links ]

15. Celebi S, Hacimustafaoglu M, Koksal N, et al. Neonatal candidiasis: Results of an 8 year study. Pediatr Int. 2012;54(3):341-9. [ Links ]

16. Caggiano G, Lovero G, De Giglio O, et al. Candidemia in the Neonatal Intensive Care Unit: A Retrospective, Observational Survey and Analysis of Literature Data. Biomed Res Int. 2017;2017:7901763. [ Links ]

17. Saiman L, Ludington E, Dawson J, et al. Risk factors for Candida species colonization of neonatal intensive care unit patients. Pediatr Infect Dis J. 2001;20(12):1119-24. [ Links ]

18. Robinson J, Pham H, Bloom B, et al. Risk factors for persistent candidemia infection in a neonatal intensive care unit and its effect on mortality and length of hospitalization. J Perinatolol. 2012;32(8)621-5. [ Links ]

19. Juyal D, Adekhandi S, Negi V, et al. An outbreak of neonatal candidemia due to non-albicans Candida species in a resource constrained setting of Uttarakhand State, India. J Clin Neonatol. 2013;2(4):183-6. [ Links ]

20. Hsu J, Lai M, Lee C, et al. Comparison of the incidence, clinical features and outcomes of invasive candidiasis in children and neonates. BMC Infectious Diseases. 2018;18:194. [ Links ]

21. Márquez F, Iturrieta I, Calvo M, et al. Epidemiología y susceptibilidad antifungica de especies causantes de candidemia en las ciudad de Valdivia, Chile. Rev Chil infectol. 2017;34(5):441-6. [ Links ]

22. Chow B, Linden J, Bliss J. Candida parapsilosis and the neonate: epidemiology, virulence and host defense in a unique patient setting. Expert Rev Anti Infect Ther 2012;10(8):935-46. [ Links ]

23. Toth R, Toth A, Vagvolgyi C, et al. Candida parapsilosis Secreted Lipase as an Important Virulence Factor. Curr Protein Pept Sci. 2017;18(10):1043-9. [ Links ]

24. Pharkjaksu S, Chongtrakool P, Suwannakarn K, et al. Species distribution, virulence factors, and antifungal susceptibility among Candida parapsilosis complex isolates from clinical specimens at Siriraj Hospital, Thailand, from 2011 to 2015. Med Mycol. 2018;56(4):426-33. [ Links ]

25. Prigent G, Ait-Ammar N, Levesque E, et al. Echinocandin Resistance in Candida Species Isolates from Liver Transplant Recipients. Antimicrob Agents Chemother. 2017;61(2):e01229-16. [ Links ]

26. Levy I, Shalit I, Askenazi S, et al. Duration and outcome of persistent candidemia in newborn infants. Mycoses. 2006;49(3):197-201. [ Links ]

27. Benjamin D, Hudak M, Duara S, et al. Effect of fluconazole prophylaxis on candidiasis and mortality in premature infants: a randomized clinical trial. JAMA. 2014;311(17):1742-9. [ Links ]

28. Ávila-Agüero M, Canas-Coto A, Ulloa Gutiérrez R, et al. Risk factors for Candida infections in a neonatal intensive care unit in Costa Rica. Int J Infect Dis. 2005;9(2):90-5. [ Links ]

29. Stoll B, Hansen N, Fanaroff A, et al. Late-onset sepsis in very low birth weight neonates: the experience of the NICHD Neonatal Research Network. J Pediatr. 2002;110(2):285-91. [ Links ]

30. Benjamin D Jr, DeLong E, Steinbach W, et al. Empirical therapy for neonatal candidemia in very low birth weight infants. J Pediatr. 2003;112(3):543-7. [ Links ]

31. Hsu J, Lai M, Lee C, et al. Comparison of the incidence, clinical features and outcomes of invasive candidiasis in children and neonates. BMC Infectious Diseases. 2018;18:194. [ Links ]

32. Kaufman D. Challenging issues in neonatal candidiasis. Curr Med Res Opin. 2010; 26(7):1769-78. [ Links ]

33. Basu S, Kumar R, Tilak R, et al. Candida Blood Stream Infection in Neonates: Experience from A Tertiary Care Teaching Hospital of Central India. Indian Pediatr. 2017;54(7):556-9. [ Links ]

34. James M, Anuja J, Ninan P. Clinical profile of neonatal candidiasis in newborn nursery. Int J Contemp Pediatr. 2018;5(2):334-7. [ Links ]

35. Bouza E, Guinea J, Guembe M. The Role of Antifungals against Candida biofilm in catheter-related candidemia. Antibiotics (Basel). 2015;4(1):1-17. [ Links ]

36. Lee J. Catheter-related bloodstream infections in neonatal intensive care units. Korean J Pediatr. 2011;54(9):363-7. [ Links ]

37. Pammi M, Zhong D, Johnson Y, et al. Polymicrobial bloodstream infections in the neonatal intensive care unit are associated with increased mortality: a case-control study. BMC Infect Dis. 2014;14:390. [ Links ]

Received: May 09, 2018; Accepted: November 27, 2018

Correspondence: Elizabeth caparó Ingram. E-mail:

Creative Commons License Este es un artículo publicado en acceso abierto bajo una licencia Creative Commons