**3. Neonatal** *Candida* **infections**

Candida species are correlated to invasive fungal infections among at-risk groups as neonatal patients admitted NICU in and have been ranked third to seventh as a cause of nosocomial bloodstream infection, defined as candidemia, depending on geographical patterns [25–28]. Studies on invasive candidiasis infections and candidemia are frequently focused on specific diagnoses and/or specific populations. In all published studies, ICU was the most frequent localization of the patients, even with different frequencies [29].

using a retrospective matched case-control study conducted in the NICUs of a teaching hospital from July 2003 to June 2006. A total of 164 infants with culture-proven bloodstream infections were identified and the common etiologic pathogens included coagulase-negative staphylococci (28.7%), *Staphylococcus aureus* (16.5%), *Klebsiella pneumoniae* (14.6%) and *Candida* species accounting for 11 (6.7%) episodes. According to these authors, parenteral nutrition was a significant and independent risk of late-onset neonatal sepsis, including those caused by *Candida* species. This risk should be considered when implementing early parenteral nutri-

Fungal Infections in Neonatal Intensive Care http://dx.doi.org/10.5772/intechopen.70302 113

The collected potential risk factors consisted of: (1) prenatal and maternal history such as toxemia, multiple gestation, intra-uterine growth retardation and perinatal infections; (2) perinatal history such as premature rupture of membrane greater than 18 hours and delay in initial crying; (3) invasive procedures such as instrument insertion and its duration (e.g., placement of nasogastric tubes, endotracheal tubes, mechanical ventilation, peripherally inserted central catheters, chest tubes, blood transfusion or exchanged blood transfusion and lumbar puncture); (4) the concomitant use of medications such as parenteral nutrition and intravenous lipid, antibiotics and steroids and (5) comorbidities such as meconium aspiration syndrome (MAS), persistent pulmonary hypertension of the newborn (PPHN), intraventricular hemorrhage (IVH), necrotizing enterocolitis (NEC), respiratory distress syndrome (RDS), inborn error of metabolism and cardiac anomalies (except for patent ductus arteriosus and secundum type of atrial septal defect). However, transfusion, antimicrobial treatment, use of steroids and the presence of other comorbidities were not associated with *Candida* infections

Invasive neonatal candidiasis presented an overall mortality rate of 35% during a study in Los Angeles (USA) in a neonatal intensive care unit. In general, every infant used a central venous catheter (CVC), required mechanical ventilation and previous administration of antibacterial agents. According to the authors, delayed institution of antifungal therapy was associated with increased mortality as well as length of hospitalization and the duration of prior anti-

Frequently, *C. albicans* is the most fungal clinical isolate; however, the incidence of bloodstream infections caused by *Candida* non-*albicans*, mainly *C. parapsilosis* complex and *C. glabrata*, has increased over the past 15 years. The current high rate of *Candida parapsilosis* infections may be attributed the capacity of this isolate to form biofilms and contaminate solutions, as those

In recent Italian study, *C. albicans* was the most frequently identified strain, but nearly 20% of infections were due to *Candida* non-*albicans*, mainly *C. krusei* and *C. glabrata* [10]. Since both these strains can be resistant to fluconazole, that is the antifungal drug with the best urinary penetration [37], treatment of these patients could be challenging, despite a recent report that showed effective concentrations of micafungin in the urinary tract [38]. In these search, fungemia was the second most frequent diagnosis and was more frequent in children with malignancy/hematopoietic stem cell transplantation, those undergoing abdominal surgery and in low birth weight neonates, also in this case, confirming other recent pediatric data such as Ota et al.

tion in NICUs.

in neonates [35].

bacterial therapy [36].

used in parenteral nutrition [8].

[39] and Steinbach et al. [40].

Candidemia is associated with high rates of illness and death and has an attributable mortality rate that varies widely in the literature, ranging from 29 to 76%, both in adult and pediatric patients [30, 31]. Furthermore, *Candida* species are common gastrointestinal flora that causes a wide range of severe manifestations when disseminated into the bloodstream. Thus, candidemia has been described as the most common manifestation of invasive candidiasis [32].

These yeasts are less frequent than those infections caused by Gram-positive or Gram-negative bacteria; nonetheless, they are higher rates of morbidity and mortality. Particularly, among newborn with extremely low weight, 10% may to develop candidemia that has until 30% mortality in this patient group. Among infants who survive these infections, several long-term neurological impairments such as cerebral palsy, blindness, hearing and cognitive deficits and periventricular leukomalacia may occur [9, 27].

Neonatal candidemia during the first week of life is less common and less well described than the later onset of this group of infections. According to Barton et al. [33], risk factors for candidemia among neonates had not been studied before their research, in early onset disease (EOD, ≤7 days) or compared to late onset disease (LOD, >7 days). After a 2-year study, the authors concluded that risk factors such as birthweight <750 g, gestation <25 weeks, chorioamnionitis and vaginal delivery were strongly associated with EOD. Infection with *Candida albicans*, disseminated disease, pneumonia and cardiovascular disease were significantly more common in EOD than in LOD. Also, neurodevelopmental impairment and mortality were also higher than controls.

Extremely low birth weight is considered a risk factor related to a poor prognosis in EOD. Also, the role of perinatal transmission is supported by its association with chorioamnionitis, vaginal delivery and pneumonia. Dissemination and cardiovascular involvement are common, and affected infants often die. Empiric treatment should be considered in these risk situations [33, 34].

Pereira et al. [34] developed a retrospective observational study to investigate the risks for sepsis in neonates, including *Candida* infections, and verified an association between healthcare-associated sepsis, antibiotic therapy in day 1, the duration of parenteral nutrition and the use of central vascular catheter. For each extra week on gestational age, the risks declined in 20%, and for each day of parenteral nutrition, the risk increased 22%.

Kung et al. [35] affirm that infants in NICU have a higher incidence of *Candida* infections than any other pediatric or adult population. The predisposing factors were evaluated in Taiwan using a retrospective matched case-control study conducted in the NICUs of a teaching hospital from July 2003 to June 2006. A total of 164 infants with culture-proven bloodstream infections were identified and the common etiologic pathogens included coagulase-negative staphylococci (28.7%), *Staphylococcus aureus* (16.5%), *Klebsiella pneumoniae* (14.6%) and *Candida* species accounting for 11 (6.7%) episodes. According to these authors, parenteral nutrition was a significant and independent risk of late-onset neonatal sepsis, including those caused by *Candida* species. This risk should be considered when implementing early parenteral nutrition in NICUs.

**3. Neonatal** *Candida* **infections**

112 Selected Topics in Neonatal Care

localization of the patients, even with different frequencies [29].

and periventricular leukomalacia may occur [9, 27].

were also higher than controls.

situations [33, 34].

Candida species are correlated to invasive fungal infections among at-risk groups as neonatal patients admitted NICU in and have been ranked third to seventh as a cause of nosocomial bloodstream infection, defined as candidemia, depending on geographical patterns [25–28]. Studies on invasive candidiasis infections and candidemia are frequently focused on specific diagnoses and/or specific populations. In all published studies, ICU was the most frequent

Candidemia is associated with high rates of illness and death and has an attributable mortality rate that varies widely in the literature, ranging from 29 to 76%, both in adult and pediatric patients [30, 31]. Furthermore, *Candida* species are common gastrointestinal flora that causes a wide range of severe manifestations when disseminated into the bloodstream. Thus, candidemia has been described as the most common manifestation of invasive candidiasis [32].

These yeasts are less frequent than those infections caused by Gram-positive or Gram-negative bacteria; nonetheless, they are higher rates of morbidity and mortality. Particularly, among newborn with extremely low weight, 10% may to develop candidemia that has until 30% mortality in this patient group. Among infants who survive these infections, several long-term neurological impairments such as cerebral palsy, blindness, hearing and cognitive deficits

Neonatal candidemia during the first week of life is less common and less well described than the later onset of this group of infections. According to Barton et al. [33], risk factors for candidemia among neonates had not been studied before their research, in early onset disease (EOD, ≤7 days) or compared to late onset disease (LOD, >7 days). After a 2-year study, the authors concluded that risk factors such as birthweight <750 g, gestation <25 weeks, chorioamnionitis and vaginal delivery were strongly associated with EOD. Infection with *Candida albicans*, disseminated disease, pneumonia and cardiovascular disease were significantly more common in EOD than in LOD. Also, neurodevelopmental impairment and mortality

Extremely low birth weight is considered a risk factor related to a poor prognosis in EOD. Also, the role of perinatal transmission is supported by its association with chorioamnionitis, vaginal delivery and pneumonia. Dissemination and cardiovascular involvement are common, and affected infants often die. Empiric treatment should be considered in these risk

Pereira et al. [34] developed a retrospective observational study to investigate the risks for sepsis in neonates, including *Candida* infections, and verified an association between healthcare-associated sepsis, antibiotic therapy in day 1, the duration of parenteral nutrition and the use of central vascular catheter. For each extra week on gestational age, the risks declined in

Kung et al. [35] affirm that infants in NICU have a higher incidence of *Candida* infections than any other pediatric or adult population. The predisposing factors were evaluated in Taiwan

20%, and for each day of parenteral nutrition, the risk increased 22%.

The collected potential risk factors consisted of: (1) prenatal and maternal history such as toxemia, multiple gestation, intra-uterine growth retardation and perinatal infections; (2) perinatal history such as premature rupture of membrane greater than 18 hours and delay in initial crying; (3) invasive procedures such as instrument insertion and its duration (e.g., placement of nasogastric tubes, endotracheal tubes, mechanical ventilation, peripherally inserted central catheters, chest tubes, blood transfusion or exchanged blood transfusion and lumbar puncture); (4) the concomitant use of medications such as parenteral nutrition and intravenous lipid, antibiotics and steroids and (5) comorbidities such as meconium aspiration syndrome (MAS), persistent pulmonary hypertension of the newborn (PPHN), intraventricular hemorrhage (IVH), necrotizing enterocolitis (NEC), respiratory distress syndrome (RDS), inborn error of metabolism and cardiac anomalies (except for patent ductus arteriosus and secundum type of atrial septal defect). However, transfusion, antimicrobial treatment, use of steroids and the presence of other comorbidities were not associated with *Candida* infections in neonates [35].

Invasive neonatal candidiasis presented an overall mortality rate of 35% during a study in Los Angeles (USA) in a neonatal intensive care unit. In general, every infant used a central venous catheter (CVC), required mechanical ventilation and previous administration of antibacterial agents. According to the authors, delayed institution of antifungal therapy was associated with increased mortality as well as length of hospitalization and the duration of prior antibacterial therapy [36].

Frequently, *C. albicans* is the most fungal clinical isolate; however, the incidence of bloodstream infections caused by *Candida* non-*albicans*, mainly *C. parapsilosis* complex and *C. glabrata*, has increased over the past 15 years. The current high rate of *Candida parapsilosis* infections may be attributed the capacity of this isolate to form biofilms and contaminate solutions, as those used in parenteral nutrition [8].

In recent Italian study, *C. albicans* was the most frequently identified strain, but nearly 20% of infections were due to *Candida* non-*albicans*, mainly *C. krusei* and *C. glabrata* [10]. Since both these strains can be resistant to fluconazole, that is the antifungal drug with the best urinary penetration [37], treatment of these patients could be challenging, despite a recent report that showed effective concentrations of micafungin in the urinary tract [38]. In these search, fungemia was the second most frequent diagnosis and was more frequent in children with malignancy/hematopoietic stem cell transplantation, those undergoing abdominal surgery and in low birth weight neonates, also in this case, confirming other recent pediatric data such as Ota et al. [39] and Steinbach et al. [40].

Invasive fungal infections in NICUs show high mortality. The better prognosis of the patient with invasive candidiasis/or candidemia admitted in NICU is associated with the early diagnosis and fast treatment. Evidence suggests an estimated mortality rate of 40% if therapy is not initiated early. Therefore, it is not a good practice to wait for cultures to become positive. This need for early therapy must be balanced against the need to use antifungal agents to avoid selection of resistant strains. Early empiric therapy guided by stratification systems for high-risk patients should help address these cases [41].

Other less common symptoms include lethargy, malnutrition, bradycardia and hepatosplenomegaly. However, no signs of erythema, swelling or purulence appear at the catheter entry site. Signs of skin rash are also not evident in children with systemic infections. Interstitial

Fungal Infections in Neonatal Intensive Care http://dx.doi.org/10.5772/intechopen.70302 115

The diagnosis of fungal infection by *Malassezia* is made by isolating the microorganism from blood collected through the catheter or by culturing the catheter tip after its removal. In suspected sepsis by *Malassezia*, the tip of the catheter should be cultured in broth enriched with

The standard therapeutic management for systemic infections by *Malassezia* is still not well defined, since the fungemia by this microorganism is relatively unusual. However, some authors recommend the use of amphotericin B to treat these infections [45, 46, 49]. Morrison and Weisdorf [50] found that all patients enrolled in their study were cured without the

Studies have indicated that the most important factor for therapeutic success against systemic infection is the removal of the infected catheter and the interruption of lipid infusion, with or

The appropriate use of antifungals agents is of particular importance in the prevention and treatment of invasive fungal infection in neonates; however, guidelines to facilitate the optimal therapy choice do not exist. The current therapeutic options that are available to treat fungemia among newborns and children are based on clinical trials in adults, since there are few comparative studies of antifungal agents in infants. The optimal treatment of fungal infection in this special population requires detailed studies on pharmacokinetics, safety and

Similar to neonatal invasive infections by species of *Candida*, the management of *Malassezia* sp. fungemia requires the removal of any catheter as soon as the first positive blood culture occurs and the temporary discontinuation of parenteral nutrition in combination with an intravenous antifungal therapy. The most commonly used agents for the treatment of invasive fungal infections in NICU are classified into four different classes: polyene, azoles, analogs of pyrimidines and echinocandins. Among many years, the drugs of choice in this group of patients were amphotericin B alone or in combination with fluocitosin, liposomal formulation of amphotericin B or fluconazole. However, the development of a new generation of azoles and echinocandins, such as micafungin, has increased the therapeutic options

Amphotericin B deoxycholate and lipid preparations are traditional choices for invasive fungal infections being active against a majority of clinical important *Candida* species and with reported use for *Malassezia* [45, 55]. Amphotericin B deoxycholate is well tolerated

bronchopneumonia can be found in 40% of children [44–46].

administration of systemic antifungal therapy.

**5. Treatment of neonatal fungal infections**

without antifungals [18, 45, 46].

efficacy of antifungal therapies [51–54].

for the treatment [45, 54].

lipids [45, 46].

The score for exact risk measurement of invasive candidiasis has yet to be developed. The "Candida Score" presented by Spanish group in 2006 provides an easy-to-use tool to assist the health professionals with critically ill adults [42]. However, we believe that will should be adapted to pediatric patients, in the near future. In this stratification, the selected variables by logistic regression model with increasing weight are total parenteral nutrition, surgery, multifocal *Candida* species colonization and severe sepsis [42].

Recent Infectious Disease Society American guidelines suggest that "empirical antifungal therapy should be considered in critically ill patients with risk factors for invasive candidiasis and no other known cause of fever." Risk factors for invasive candidiasis are well identified. When analyzing clinical data, surveillance culture and levels of anti-*Candida* antibodies plus β-D-glucan in the serum, the same Spanish researchers showed a positive correlation among increasing values of the "*Candida* score" and the rate of invasive *Candida* infections. Such a score was calculated by variables such as total parenteral nutrition, surgery, multifocal *Candida* colonization and severe sepsis. Thus, *Candida* score ≥3 suggest patients at high risk for invasive candidiasis and enable to differentiate patients who would benefit from early antifungal treatment from those for whom invasive candidiasis is highly improbable [43].
