Advertisement
Research Article| Volume 20, 101247, March 2023

Download started.

Ok

Epidemiology, risk factors and outcomes associated with candidaemia in very low birth weight infants at a tertiary South African Hospital over a 7-year period (2013–2019)

Open AccessPublished:January 31, 2023DOI:https://doi.org/10.1016/j.cegh.2023.101247

      Abstract

      Introduction

      Candidaemia is a significant problem in neonatal units and is associated with high morbidity, including long-term neurodevelopmental impairment in survivors, and high mortality of very low birth weight infants (VLBWI).

      Method

      A retrospective cohort study amongst VLBWI admitted to the neonatal unit at Charlotte Maxeke Johannesburg Academic Hospital (CMJAH), Johannesburg, South Africa, from 1 January 2013 to 31 December 2019. All VLBWI were born at the hospital or transferred to the neonatal unit from birth to day 28 of life with blood culture confirmed candidaemia.

      Results

      During the study period, 3414 VLBWI were admitted to the unit. Of these, 5.12% (n = 176) developed culture confirmed candidaemia. The incidence was 5.1 per 1000 admissions. The most common species, which persisted throughout the study period, was Candida parapsilosis, followed by Candida albicans. C. parapsilosis peaked in 2018 while C. albicans peaked in 2015. Emergence of C. auris occurred in 2019. Important risk factors associated with the development of candidaemia included necrotizing enterocolitis (p < 0.001, OR 4.63 [3.29–6.54]), surgery (p < 0.001 OR 7.02 [4.48–11.12]), conventional ventilation (p < 0.001, OR 6.23 [4.48–8.68]), patent ductus arteriosus (p < 0.001, OR 3.81 [2.67–5.44]), intraventricular haemorrhage (p < 0.001, OR 3.32 [2.99–5.44]) and prolonged hospital stay (p < 0.001). Mortality was not statistically different (p = 0.80 OR 0.95[0.68–1.31]) between the two groups.

      Conclusion

      There is a high incidence of candidaemia in the neonatal unit. Several modifiable risk factors including improved antifungal stewardship and prevention of candidaemia with oral or systemic antifungal prophylaxis may decrease the incidence of candidaemia, and associated morbidity.

      Keywords

      Abbreviations

      AFS
      antifungal stewardship
      AMS
      antimicrobial stewardship
      AFST
      antifungal susceptibility testing
      BDG
      Beta-D-glucan
      BC
      blood culture
      BPD
      bronchopulmonary hypoplasia
      CDC
      Centre of Disease Control
      CI
      confidence interval
      CLSI
      Clinical and Laboratory Standard Institute
      IDSA
      Infectious Disease Society of America
      HICS
      high income country setting
      IQR
      interquartile range
      IVH
      intraventricular haemorrhage
      LMICs
      low middle income country setting
      LOS
      late-onset sepsis
      MV
      mechanical ventilation
      NCPAP
      nasal continuous positive airway pressure
      NEC
      necrotizing enterocolitis
      OR
      odds ratio
      PDA
      patent ductus aterious
      PN
      parenteral nutrition
      REDCap
      Research Electronic Data Capture
      VLBWI
      very low birth weight infants

      1. Background

      Candidaemia is a significant problem in neonatal units and is associated with high morbidity, including long-term neurodevelopmental impairment in survivors, and high mortality of very low birth weight infants (VLBWI).
      • Rundjan L.
      • Wahyuningsih R.
      • Oeswadi C.A.
      • Marsogi M.
      • Purnamasari A.
      Oral nystatin prophylaxis to prevent systemic fungal infection in very low birth weight preterm infants: a randomized controlled trial.
      • Charoo B.
      • Ashraf Y.
      • Bhat J.
      • Qazi I.
      Systemic Candida infection in preterm babies: experience from a tertiary care hospital of North India.
      • Ballot D.E.
      • Bosman N.
      • Nana T.
      • Ramdin T.
      • Cooper P.A.
      Background changing patterns of neonatal fungal sepsis in a developing country.
      Globally, the mortality rate of candidaemia can be as high as 50%.
      • Charoo B.
      • Ashraf Y.
      • Bhat J.
      • Qazi I.
      Systemic Candida infection in preterm babies: experience from a tertiary care hospital of North India.
      ,

      Montagna MT, Caggiano G, Lovero G, De Giglio O, Coretti C, Cuna T, et al. Epidemiology of invasive fungal infections in the intensive care unit: results of a multicenter Italian survey (AURORA Project). Infection2013;41(3):645–653.

      • Barton M.
      • Shen A.
      • O'Brien K.
      • et al.
      Early-onset invasive candidiasis in extremely low birth weight infants: perinatal acquisition predicts poor outcome.
      • Aziz M.
      • Patel A.L.
      • Losavio J.
      • et al.
      Efficacy of fluconazole prophylaxis for prevention of invasive fungal infection in extremely low birth weight infants.
      In a study conducted in our centre in 2013, the mortality rate from candidaemia was recorded at 45.8%.
      • Ballot D.E.
      • Bosman N.
      • Nana T.
      • Ramdin T.
      • Cooper P.A.
      Background changing patterns of neonatal fungal sepsis in a developing country.
      Candidaemia is an important cause of late onset sepsis (LOS) in VLBWI and accounts for approximately 12% of all LOS.
      • Malunga C.
      • Nana T.
      • Ballot D.
      A case–control study of candidaemia in very low birthweight infants in a tertiary hospital in Johannesburg.
      An early and reliable diagnosis of candidaemia is challenging. This is related to its nonspecific clinical presentation (lethargy, glycaemic instability, increased oxygen requirements, and thrombocytopenia) and the poor sensitivity of Candida isolation in culture, all of which may result in a delayed initiation of appropriate therapy.
      • Rundjan L.
      • Wahyuningsih R.
      • Oeswadi C.A.
      • Marsogi M.
      • Purnamasari A.
      Oral nystatin prophylaxis to prevent systemic fungal infection in very low birth weight preterm infants: a randomized controlled trial.
      ,
      • Ballot D.E.
      • Bosman N.
      • Nana T.
      • Ramdin T.
      • Cooper P.A.
      Background changing patterns of neonatal fungal sepsis in a developing country.
      ,
      • Cleminson J.
      • Austin N.
      • McGuire W.
      Prophylactic systemic antifungal agents to prevent mortality and morbidity in very low birth weight infants.
      In addition to blood stream infection with Candida, infants may develop meningitis, urinary tract infection, ophthalmitis and endocarditis.
      • Hammoud M.S.
      • Al-Taiar A.
      • Fouad M.
      • Raina A.
      • Khan Z.
      Persistent candidemia in neonatal care units: risk factors and clinical significance.
      A high index of suspicion and the use of investigations including retinal examination, echocardiography, and abdominal ultrasonography, may be necessary to confirm the suspected diagnosis. Serum 1, 3 beta-D glucan (BDG), a rapid, non-culture-based broad fungal antigen assay, has assisted in the early initiation of antifungal therapy in at-risk patients. Its use, followed by close follow-up and discontinuation of antifungal therapy when invasive candidiasis is excluded, has shown to improve outcomes of patients.
      • Cleminson J.
      • Austin N.
      • McGuire W.
      Prophylactic systemic antifungal agents to prevent mortality and morbidity in very low birth weight infants.
      ,
      • Mackay C.A.
      • Ballot D.E.
      • Perovic O.
      Serum 1,3-βD-Glucan assay in the diagnosis of invasive fungal disease in neonates.
      ,
      • Sawai T.
      • Nakao T.
      • Yamaguchi S.
      • et al.
      Detection of high serum levels of β-D-Glucan in disseminated nocardial infection: a case report.
      VLBWI have a compromised immune system and mucosal integrity. Gastrointestinal tract colonization by Candida species, especially heavy colonization, may become a primary source of translocation through epithelial barriers and systemic dissemination in high-risk VLBWI.
      • Charoo B.
      • Ashraf Y.
      • Bhat J.
      • Qazi I.
      Systemic Candida infection in preterm babies: experience from a tertiary care hospital of North India.
      The high morbidity and mortality associated with candidaemia in VLBWI render prevention of this infection essential.
      Clinical implementation of antifungal stewardship (AFS) is recommended for acute-care hospitals.
      • Govender N.P.
      • Avenant T.
      • Brink A.
      • et al.
      Federation of Infectious Diseases Societies of Southern Africa guideline: recommendations for the detection, management, and prevention of healthcare-associated Candida auris colonisation and disease in South Africa.
      A multidisciplinary team including a clinical microbiologist is required to implement and monitor AFS interventions. Audits of candidaemia and antifungal therapy should be performed in every neonatal unit.
      Due to the paucity of data from the neonatal population, and specifically from the African continent, this study was undertaken to determine the change in the prevalence and epidemiology of candidaemia in our setting as well as to describe the associated risk factors and mortality in VLBWI with candidaemia.

      2. Methods

      2.1 Study design

      We performed a retrospective cohort study amongst VLBWI admitted to the neonatal unit at Charlotte Maxeke Johannesburg Academic Hospital (CMJAH), Johannesburg, South Africa, from 1 January 2013 to 31 December 2019.
      • (Fungal sepsis in a small subset of this cohort was previously published).
        • Malunga C.
        • Nana T.
        • Ballot D.
        A case–control study of candidaemia in very low birthweight infants in a tertiary hospital in Johannesburg.

      2.2 Study setting

      Charlotte Maxeke Johannesburg Academic Hospital has a 90-bed high care and low care neonatal unit, including a 10-bed kangaroo mother care ward and a 14-bed combined paediatric and neonatal intensive care unit.
      In this unit, infants suspected of nosocomial fungal infections are treated with empiric Amphotericin B deoxycholate therapy while awaiting blood culture and BDG results. Empiric antifungal therapy is based on unit-specific prevalence and antifungal susceptibility patterns of specific Candida species. A high prevalence of azole resistant non-albicans Candida species, including C. parapsilosis, in South Africa renders the routine use of empiric fluconazole inappropriate.
      • Ballot D.E.
      • Bosman N.
      • Nana T.
      • Ramdin T.
      • Cooper P.A.
      Background changing patterns of neonatal fungal sepsis in a developing country.
      ,
      • Chibabhai V.
      Incidence of candidemia and prevalence of azole-resistant candidemia at a tertiary South African hospital – a retrospective laboratory analysis 2016–2020.

      2.3 Inclusion criteria

      All VLBWI were born at the hospital or transferred to the neonatal unit from birth to day 28 of life.

      2.4 Exclusion criteria

      Infants with incomplete or missing information were excluded from the study.

      2.5 Data collection

      Data was managed using Research Electronic Data Capture (REDCap), hosted by the University of the Witwatersrand.
      • Harris P.a.
      • Taylor R.
      • Thielke R.
      • Payne J.
      • Gonzalez N.
      • Conde J.G.
      Research Electronic Data Capture (REDCap) - a metadata driven methodology and workflow process for providing translational research informatict support.
      Maternal, labour room and neonatal variables were collected.
      Tabled 1Definitions
      Candidaemia:At least one blood culture obtained by peripheral venipuncture which cultured Candida species.
      Necrotizing enterocolitis:Modified Bell's criteria of stage 2 or more
      • Neu J.
      • Modi N.
      • Caplan M.
      Necrotizing enterocolitis comes in different forms: historical perspectives and defining the disease.
      Intraventricular Haemorrhage:diagnosed on cranial ultrasound and defined using Papiles's score of grade 3 and grade 4
      • Papile L.A.
      • Burstein J.
      • Burstein R.
      • Koffler H.
      Incidence and evolution of subependymal and intraventricular hemorrhage: a study of infants with birth weights less than 1,500 gm.
      Early onset sepsis:blood culture proven sepsis <72 h of life
      Late onset sepsis:blood culture proven sepsis >72 h of life
      Delivery room resuscitation:bag mask ventilation or CPR and adrenaline after delivery
      Bronchopulmonary dysplasia:requiring oxygen >28 days of life

      2.6 Microbiologic methods

      Blood cultures requested for fungal culture were incubated in the BacTAlert® automated blood culture system for 14 days. Once the bottle flagged positive, the bottle was removed, a Gram stain was performed, and the results thereof reported immediately to the attending clinician. Bottles with yeasts visible on Gram stain were plated out onto 5% sheep blood agar, chocolate agar, and Sabouraud dextrose agar. These were incubated at 37 °C for 72 h.
      Once yeast isolates were cultured, they were identified using the Vitek® MS MALDI-TOF (Matrix-assisted Laser Desorption Ionisation – Time of Flight) analyzer or the Vitek® 2 automated identification and antimicrobial susceptibility testing system. Antifungal susceptibility test (AFST) results were obtained using the Vitek® 2 or the gradient diffusion method for species where no clinical interpretive breakpoints were available. AFSTS were interpreted according to Clinical and Laboratory Standards Institute (CLSI) guidelines. Candida auris isolates were interpreted using the tentative Centers for Disease Control and Prevention (CDC) breakpoint.

      2.7 Statistical analysis

      Statistical analysis was performed using SPSS version 27 (IBM, USA). Continuous data with normal distribution were described using mean and standard deviation, and skewed data were described using median and interquartile range (IQR). Categorical data were described using frequencies and percentages. VLBWI with candidaemia were compared to VLBWI without candidaemia. Potential risk factors associated with candidemia were identified using logistic regression analysis. Variables with 2-tailed P-value <0.05 were defined as statistically significant. Odds ratios (ORs) along with 95% confidence intervals (CIs) were used to assess the strength of any association.

      2.8 Ethical considerations

      Permission to perform this study was granted by the Human Research Ethics Committee (HREC) of the University of the Witwatersrand (M180625).

      3. Results

      During the study period, 3414 VLBWI were admitted to the unit of these, 5.12% (n = 176) developed cultures confirmed candidaemia. The incidence was 5.1 per 1000 admissions.
      The most common species, which persisted throughout the study period, was Candida parapsilosis, followed by Candida albicans (see Fig. 1.) C. parapsilosis peaked in 2018 while C. albicans peaked in 2015. Emergence of C. auris occurred in 2019. The data from 2013 to 2014 had missing species names (n = 84; 34.15%); hence, these were omitted in Fig. 1.
      Fig. 1
      Fig. 1Annual percentage distribution of Candidaemia episodes according to the 5 most common isolated Candida species.
      The total percentages for 5-year period (excluding 2013 and 2014) were 60.49% for C. parapsilosis, 28.40% for C. albicans, 6.17% for other Candida species, 3.09% for C.auris, and 1.85% for C. glabarata.
      Birth weight, gestational age and length of hospital stay associated with candidaemia are shown in Table 1.
      Table 1Continuous variables associated with candidaemia in VLBWI at a tertiary hospital in Johannesburg, South Africa.
      VariableCategoryFungal sepsisP-valueOdds Ratio (95%CI)
      No N (%)Yes N (%)
      Birth weight (g)Median (IQR)1130 (920–1310)1050 (910–1230)0.011.00(0.99–1.00)
      Gestational age (weeks)Mean (SD)29.00 (2.65)28.49 (2.27)<0.0010.85(0.77–0.94)
      Length of hospital stay (days)Median (IQR)25 (9–43)50 (26–72)<0.0011.03(1.02–1.03)
      The maternal and neonatal factors associated with candidemia in VLBWI are summarized in Table 2 and Table 3.
      Table 2Maternal variables associated with candidemia in VLBWI at a tertiary hospital in Johannesburg, South Africa.
      VariableCategoryFungal sepsisP-value*Odds Ratio (95%CI)
      No N (%)Yes N (%)
      Maternal HIV statusPositive942 (29.2)60 (34.3)0.091.26(0.92–1.74)
      N = 3399Negative2 281 (70.8)116 (65.7)
      Mode of deliveryVaginal1 342 (43.2)68 (44.7)0.380.94(0.68–1.30)
      N = 3236Caesarean section1 742 (56.8)84 (54.1)
      Antenatal careYes2 399 (79.4)118 (76.6)0.420.85(0.58–1.25)
      N = 3175No622 (20.6)36 (23.5)
      Antenatal steroidsYes1 376 (49.0)73 (52.1)0.491.13(0.81–1.59)
      N = 2946No1 430 (51.0)67 (47.9)
      ChorioamnionitisYes85 (2.8)9 (5.1)0.061.99(0.99–4.03)
      N = 3383No3 123 (97.4)166 (94.9)
      Table 3Neonatal variables associated with candidemia in VLBWI at a tertiary hospital in Johannesburg, South Africa.
      VariableCategoryFungal sepsisOdds Ratio (95%CI)
      No N (%)Yes N (%)
      Place of birthOutborn628 (18.4)57 (32.6)<0.0012.01(1.45–2.79)
      N = 3414Inborn2 611 (80.6)118 (67.4)
      Initial resuscitation in the delivery roomYes2 186 (69,0)115 (68.0)0.870.96(0.69–1.34
      N = 3338No983 (31.0)54 (32.0)
      Necrotising enterocolitisYes2 921 (91.1)119 (68.8)<0.0014.64 (3.30–6.54)
      N = 3380No286 (8.9)54 (31.2)
      Other surgeryYes84 (2.8)28 (16.8)<0.0017.02 (4.43–11.12)
      N = 3177No2 926 (97.2)139 (83.2)
      Parenteral nutritionYes105 (20.3)15 (48.9)<0.0013.46 (1.67–7.14)
      N = 548No411 (79.7)17 (53.1)
      Patent ductus arteriosusYes287 (8.89)47 (27.2)<0.0013.81 (2.67–5.44)
      N = 3394No2 932 (91.1)128 (72.8)
      Intraventricular haemorrhageNo3 091 (95.4)151 (86.3)<0.0013.319 (2.09–5.27)
      N = 3414Yes148 (4.6)24 (13.7)
      Conventional VentilationNo2 263 (75.7)56 (33.3)<0.0016.23 (4.48–8.68)
      N = 3157Yes726 (24.3)112 (66.7)
      NCPAP ventilationYes2 108 (69.5)139 (84.3)<0.0012.44 (1.58–3.76)
      N = 3197No925 (30.5)25 (15.2)
      Early onset of bacterial sepsisNo3 077 (95.5)170 (97.1)0.340.62(0.25–1.54)
      N = 3398Yes145 (4.5)6 (2.9)
      Late onset of bacterial sepsisNo2 327 (72.3)0<0.0010.27 (0.26–0.29)
      N = 3393Yes891 (27.7)175 (100)
      Breastfeeding methodBreast & formula113 (5.6)7 (7.1)0.022.85 (2.05–3.96)
      N = 2103Breastmilk only971 (48.5)34 (34.3)
      Formula only920 (45.958 (58.6)
      Death outcomeNo2 260 (69.8)120 (68.6)0.800.95(0.68–1.31)
      N = 3405Yes970 (30.2)55 (31.4)
      *Chi-2.
      p-values for categorical variables.
      The univariate analysis of clinical factors and fungal sepsis in Table 1, Table 2 found many of the variables to be statistically significant, namely, necrotizing enterocolitis (NEC), surgery, parenteral nutrition (PN), conventional ventilation, nasal continuous positive airway pressure (NCPAP) ventilation, patent ductus arteriosus (PDA), intraventricular haemorrhage (IVH), late onset sepsis, bronchopulmonary dysplasia (BPD) and prolonged hospital stay.
      Mortality was not statistically significant (p = 0.80 OR 0.95[ 0.68–1.31]) between the two groups.

      4. Discussion

      There has been a significant increase in the incidence of candidaemia in VLBWI in our neonatal unit. In 2013 the incidence was 1.2 per 1 000 admissions, 2014 it was 1.5 per 1 000 admissions and 5.1 per 1 000 admissions in 2019.
      • Ballot D.E.
      • Bosman N.
      • Nana T.
      • Ramdin T.
      • Cooper P.A.
      Background changing patterns of neonatal fungal sepsis in a developing country.
      ,
      • Malunga C.
      • Nana T.
      • Ballot D.
      A case–control study of candidaemia in very low birthweight infants in a tertiary hospital in Johannesburg.
      However, this is based on an imperfect gold standard because blood cultures lack sensitivity for fungal sepsis.
      • Mackay C.A.
      • Ballot D.E.
      • Perovic O.
      Serum 1,3-βD-Glucan assay in the diagnosis of invasive fungal disease in neonates.
      ,
      • Chibabhai V.
      • Fadana V.
      • Bosman N.
      • Nana T.
      Comparative sensitivity of 1,3 beta-D-glucan for common causes of candidaemia in South Africa.
      The incidence of candidaemia was comparable to other studies in low and middle income countries (LMICs), but significantly higher than studies conducted in high income countries (HICS).
      • Charoo B.
      • Ashraf Y.
      • Bhat J.
      • Qazi I.
      Systemic Candida infection in preterm babies: experience from a tertiary care hospital of North India.
      ,
      • Ballot D.E.
      • Bosman N.
      • Nana T.
      • Ramdin T.
      • Cooper P.A.
      Background changing patterns of neonatal fungal sepsis in a developing country.
      ,
      • Govender N.P.
      • Avenant T.
      • Brink A.
      • et al.
      Federation of Infectious Diseases Societies of Southern Africa guideline: recommendations for the detection, management, and prevention of healthcare-associated Candida auris colonisation and disease in South Africa.
      This is likely related to overcrowding, poor infection control measures and inappropriate antibiotic use in the neonatal unit.
      During the study period, our neonatal unit observed a change from a predominance of C. albicans to C. parapsilosis candidaemia and the emergence of C. auris in 2019.
      • Ballot D.E.
      • Bosman N.
      • Nana T.
      • Ramdin T.
      • Cooper P.A.
      Background changing patterns of neonatal fungal sepsis in a developing country.
      This is consistent with national surveillance data which has seen a rise in C. auris since 2014.
      • Govender N.P.
      • Avenant T.
      • Brink A.
      • et al.
      Federation of Infectious Diseases Societies of Southern Africa guideline: recommendations for the detection, management, and prevention of healthcare-associated Candida auris colonisation and disease in South Africa.
      ,
      • Govender N.P.
      • Patel J.
      • Magobo R.E.
      • et al.
      Emergence of azole-resistant Candida parapsilosis causing bloodstream infection: results from laboratory-based sentinel surveillance in South Africa.
      ,
      • Cantey J.B.
      • Milstone A.M.
      Bloodstream infections: epidemiology and resistance.
      Globally,C. auris now accounts for one in 10 cases of candidemia.
      • Van Schalkwyk E.
      • Mpembe R.S.
      • Thomas J.
      • et al.
      Epidemiologic shift in Candidemia driven by Candida auris, South Africa, 2016-2017.
      This presents a very serious global health threat. This emphasises the importance of C. auris identification on blood culture and screening of patients when transmission or colonization of C. auris is suspected to prevent and control the spread in neonatal units. C. auris is difficult to eradicate in the neonatal unit and is resistant to treat. Similarly, to a study by V. Chibabhai, our study, further highlights the high prevalence of azole resistant C. parapsilosis and C.auris due to prior exposure to broad spectrum antibiotics
      • Chibabhai V.
      Incidence of candidemia and prevalence of azole-resistant candidemia at a tertiary South African hospital – a retrospective laboratory analysis 2016–2020.
      The Federation of Infectious Diseases Societies of Southern Africa has recently published guidelines pertaining to the detection, management, and prevention of healthcare-associated C. auris infection.
      • Van Schalkwyk E.
      • Mpembe R.S.
      • Thomas J.
      • et al.
      Epidemiologic shift in Candidemia driven by Candida auris, South Africa, 2016-2017.
      Factors such as the presence of central venous catheters and endotracheal tubes, ventilation, delayed enteral feeding, necrotizing enterocolitis and surgery increase fungal colonization and candidaemia.
      • Charoo B.
      • Ashraf Y.
      • Bhat J.
      • Qazi I.
      Systemic Candida infection in preterm babies: experience from a tertiary care hospital of North India.
      ,
      • Malunga C.
      • Nana T.
      • Ballot D.
      A case–control study of candidaemia in very low birthweight infants in a tertiary hospital in Johannesburg.
      Prolonged use of parenteral nutrition (PN) and the use of certain medications, such as broad-spectrum antibiotics, corticosteroids, histamine type 2-receptor blockers, and theophylline, also increase the risk of candidaemia.
      • Charoo B.
      • Ashraf Y.
      • Bhat J.
      • Qazi I.
      Systemic Candida infection in preterm babies: experience from a tertiary care hospital of North India.
      ,
      • Manzoni P.
      • Mostert M.
      • Castagnola E.
      Update on the management of Candida infections in preterm neonates.
      Similarly,in our study risk factors identified included NEC, surgery, PN, ventilation, LOS, PDA, IVH and BPD. However, this suggests critically ill VLBWI were more susceptible to candidaemia as well as prolonged hospital admission. These risk factors could be incorporated in a Candida score to implement early initiation of antifungal treatment in neonatal units to decrease morbidity and mortality.
      • Mellinghoff S.C.
      • Hoenigl M.
      • Koehler P.
      • et al.
      EQUAL Candida Score: an ECMM score derived from current guidelines to measure QUAlity of Clinical Candidaemia Management.
      In the current study, mortality was not significantly different in the two groups. However, the unit has a high rate of bacterial LOS and other complications of prematurity, which contributes to the overall high mortality rate in the nonfungal sepsis group of this study cohort.
      Based on the findings of our study, we recommend that Amphotericin B continue to be used as empiric therapy based on the predominance of non-albicans Candida species in VLBWI presenting with suspected LOS and risk factors including NEC, surgery, PN. Babies with evidence of PDA, IVH, and those with prolonged hospital stay should also be considered for empiric therapy when LOS is suspected. There is little evidence that echinocandins penetrate the blood-brain barrier and should be used with caution in neonates or avoided, since meningo-encephalitis cannot easily be identified and excluded on lumbar punctures.
      • Caudle K.E.
      • Inger A.G.
      • Butler D.R.
      • David Rogers P.
      Uso de Equinocandinas en la Unidad de Cuidados Intensivos Neonatal.
      Using AFS principles, antifungal therapy should be started empirically only in critically ill VLBWIs with thrombocytopenia after collection of blood cultures including fungal blood cultures and BDG or other non-culture based diagnostic assays (depending on availability). In cases where the blood culture is negative for Candida species, the BDG is not suggestive, and an alternative diagnosis is found, the antifungal agent should be discontinued. When azole susceptible species have been identified, de-escalation of azoles is recommended once repeat blood cultures are confirmed to be negative and adequate source control obtained.
      • Chibabhai V.
      • Fadana V.
      • Bosman N.
      • Nana T.
      Comparative sensitivity of 1,3 beta-D-glucan for common causes of candidaemia in South Africa.
      ,
      • Pappas P.G.
      • Kauffman C.A.
      • Andes D.R.
      • et al.
      Clinical practice guideline for the management of candidiasis: 2016 update by the infectious Diseases society of America.
      The Infectious Diseases Society of America (IDSA) guidelines recommend antifungal prophylaxis in neonatal units with fungal sepsis incidence rates >10%.
      • Pappas P.G.
      • Kauffman C.A.
      • Andes D.R.
      • et al.
      Clinical practice guideline for the management of candidiasis: 2016 update by the infectious Diseases society of America.
      This should be considered in settings with high-risk neonates. A recent Cochrane review concluded that prophylactic systemic antifungal therapy reduces the incidence of invasive fungal infection in VLBWI.
      • Cleminson J.
      • Austin N.
      • McGuire W.
      Prophylactic systemic antifungal agents to prevent mortality and morbidity in very low birth weight infants.
      However, the results should be interpreted cautiously due to the high incidence of invasive fungal infection in the control groups of many of the included trials. Additionally, there was no statistically significant decrease in mortality and there is only limited data on the long-term neurodevelopmental consequences for infants exposed to this intervention.
      • Rundjan L.
      • Wahyuningsih R.
      • Oeswadi C.A.
      • Marsogi M.
      • Purnamasari A.
      Oral nystatin prophylaxis to prevent systemic fungal infection in very low birth weight preterm infants: a randomized controlled trial.
      ,
      • Cleminson J.
      • Austin N.
      • McGuire W.
      Prophylactic systemic antifungal agents to prevent mortality and morbidity in very low birth weight infants.
      The emergence of antifungal resistance remains an important concern when using prophylactic antifungal therapy.
      • Ballot D.E.
      • Bosman N.
      • Nana T.
      • Ramdin T.
      • Cooper P.A.
      Background changing patterns of neonatal fungal sepsis in a developing country.
      ,
      • Cleminson J.
      • Austin N.
      • McGuire W.
      Prophylactic systemic antifungal agents to prevent mortality and morbidity in very low birth weight infants.
      In settings with high rates of non-albicans candidaemia, oral nystatin prophylaxis is easily available, affordable, and easily administered.

      4.1 Limitations

      Our study has several limitations. This was a retrospective study with some missing data. There was a difference in data collection methodology before 2015, meaning species were not recorded. Missing species data in 2013 and 2014 may have skewed the species distribution. The study was based on blood culture positive cases only. Fungal sepsis diagnosed using non-culture-based techniques such as BDG and clinical suspicion were not included and may be the reason for the lower prevalence documented in this study. This was a single-centre study, and the findings may not be broadly applicable in other settings.

      5. Conclusion

      The study found that there is a high incidence risk of candidaemia within the neonatal unit. Significant risk factors for candidaemia included NEC, surgery, mechanical ventilation, and prolonged hospital stay. In 2019, there was an emergence of C auris. Several modifiable factors including improved antifungal stewardship and prevention of candidaemia with oral or systemic antifungal prophylaxis may decrease the mortality from candidaemia.

      Funding

      None.

      Declaration of competing interest

      None.

      Acknowledgements

      The authors acknowledge the late Mr. L Rapola for his assistance in data capture.

      References

        • Rundjan L.
        • Wahyuningsih R.
        • Oeswadi C.A.
        • Marsogi M.
        • Purnamasari A.
        Oral nystatin prophylaxis to prevent systemic fungal infection in very low birth weight preterm infants: a randomized controlled trial.
        BMC Pediatr. 2020; 20
        • Charoo B.
        • Ashraf Y.
        • Bhat J.
        • Qazi I.
        Systemic Candida infection in preterm babies: experience from a tertiary care hospital of North India.
        J Clin Neonatol. 2019; 8: 151
        • Ballot D.E.
        • Bosman N.
        • Nana T.
        • Ramdin T.
        • Cooper P.A.
        Background changing patterns of neonatal fungal sepsis in a developing country.
        J Trop Pediatr. 2013; 59: 460-464
      1. Montagna MT, Caggiano G, Lovero G, De Giglio O, Coretti C, Cuna T, et al. Epidemiology of invasive fungal infections in the intensive care unit: results of a multicenter Italian survey (AURORA Project). Infection2013;41(3):645–653.

        • Barton M.
        • Shen A.
        • O'Brien K.
        • et al.
        Early-onset invasive candidiasis in extremely low birth weight infants: perinatal acquisition predicts poor outcome.
        Clin Infect Dis. 2017; 64: 921-927
        • Aziz M.
        • Patel A.L.
        • Losavio J.
        • et al.
        Efficacy of fluconazole prophylaxis for prevention of invasive fungal infection in extremely low birth weight infants.
        Pediatr Infect Dis J. 2010; 29: 352-356
        • Malunga C.
        • Nana T.
        • Ballot D.
        A case–control study of candidaemia in very low birthweight infants in a tertiary hospital in Johannesburg.
        Wits J Clin Med. 2020; 2: 25
        • Cleminson J.
        • Austin N.
        • McGuire W.
        Prophylactic systemic antifungal agents to prevent mortality and morbidity in very low birth weight infants.
        Cochrane Database Syst Rev. 2015; 10
        • Hammoud M.S.
        • Al-Taiar A.
        • Fouad M.
        • Raina A.
        • Khan Z.
        Persistent candidemia in neonatal care units: risk factors and clinical significance.
        Int J Infect Dis. 2013; 17
        • Mackay C.A.
        • Ballot D.E.
        • Perovic O.
        Serum 1,3-βD-Glucan assay in the diagnosis of invasive fungal disease in neonates.
        Pediatr Rep. 2011; 3
        • Sawai T.
        • Nakao T.
        • Yamaguchi S.
        • et al.
        Detection of high serum levels of β-D-Glucan in disseminated nocardial infection: a case report.
        BMC Infect Dis. 2017; 17
        • Govender N.P.
        • Avenant T.
        • Brink A.
        • et al.
        Federation of Infectious Diseases Societies of Southern Africa guideline: recommendations for the detection, management, and prevention of healthcare-associated Candida auris colonisation and disease in South Africa.
        S Afr J Infect Dis. 2019; 34
        • Chibabhai V.
        Incidence of candidemia and prevalence of azole-resistant candidemia at a tertiary South African hospital – a retrospective laboratory analysis 2016–2020.
        S Afr J Infect Dis. 2022; 37
        • Harris P.a.
        • Taylor R.
        • Thielke R.
        • Payne J.
        • Gonzalez N.
        • Conde J.G.
        Research Electronic Data Capture (REDCap) - a metadata driven methodology and workflow process for providing translational research informatict support.
        J Biomed Inform [Internet. 2009; 42 (Available from:): 377-381
        • Neu J.
        • Modi N.
        • Caplan M.
        Necrotizing enterocolitis comes in different forms: historical perspectives and defining the disease.
        Semin Fetal Neonatal Med. 2018; 23: 370-373
        • Papile L.A.
        • Burstein J.
        • Burstein R.
        • Koffler H.
        Incidence and evolution of subependymal and intraventricular hemorrhage: a study of infants with birth weights less than 1,500 gm.
        J Pediatr. 1978; 92: 529-534
        • Chibabhai V.
        • Fadana V.
        • Bosman N.
        • Nana T.
        Comparative sensitivity of 1,3 beta-D-glucan for common causes of candidaemia in South Africa.
        Mycoses. 2019; 62: 1023-1028
        • Govender N.P.
        • Patel J.
        • Magobo R.E.
        • et al.
        Emergence of azole-resistant Candida parapsilosis causing bloodstream infection: results from laboratory-based sentinel surveillance in South Africa.
        J Antimicrob Chemother. 2016; 71: 1994-2004
        • Cantey J.B.
        • Milstone A.M.
        Bloodstream infections: epidemiology and resistance.
        Clin Perinatol. 2015; 42: 1-16
        • Van Schalkwyk E.
        • Mpembe R.S.
        • Thomas J.
        • et al.
        Epidemiologic shift in Candidemia driven by Candida auris, South Africa, 2016-2017.
        Emerg Infect Dis. 2019; 25: 1698-1707
        • Manzoni P.
        • Mostert M.
        • Castagnola E.
        Update on the management of Candida infections in preterm neonates.
        Arch Dis Child Fetal Neonatal Ed. 2015; 100: F454-F459
        • Mellinghoff S.C.
        • Hoenigl M.
        • Koehler P.
        • et al.
        EQUAL Candida Score: an ECMM score derived from current guidelines to measure QUAlity of Clinical Candidaemia Management.
        Mycoses. 2018; 61: 326-330
        • Caudle K.E.
        • Inger A.G.
        • Butler D.R.
        • David Rogers P.
        Uso de Equinocandinas en la Unidad de Cuidados Intensivos Neonatal.
        Ann Pharmacother. 2012; 46: 108-116
        • Pappas P.G.
        • Kauffman C.A.
        • Andes D.R.
        • et al.
        Clinical practice guideline for the management of candidiasis: 2016 update by the infectious Diseases society of America.
        Clin Infect Dis. 2015; 62: e1-e50