Emergence of non-albicans Candida species in neonatal candidemia.

BACKGROUND: Candida species are one of the most common causes of blood stream infections among neonates and account for 9-13% of such infections. Although Candida albicans remains the most common fungal isolate from neonatal candidemia, longitudinal studies have detected a shift towards non-albicans Candida (NAC) species. AIM: To examine the prevalence and epidemiology of candidemia among infants admitted to our hospital.
MATERIALS AND METHODS: Blood samples were collected from 548 neonates and only those which yielded pure growth of Candida spp. were included in the study. The isolates were identified as per standard mycological techniques and antifungal susceptibility (AFS) was done by disc diffusion method.
RESULTS: Of the total 132 neonates included in the study, NAC species were responsible for 80.30% cases with C. parapsilosis (25.0%) and C. tropicalis (21.97%) as the most predominant species; whereas 19.70% of cases were caused by C. albicans. AFS results revealed that 65.91, 73.49, and 96.21% isolates were sensitive to fluconazole (FLK), itraconazole (ITR), and amphotericin B (AMB), respectively.
CONCLUSION: Candidemia in neonates is an ominous prognostic sign and is an important entity in our hospital. Strict infection control strategies, appropriate preventive and therapeutic measures such as prophylactic antifungal use and a restrictive policy of antibiotic use should be implemented.

Introduction

Significance of Candida spp. in neonatal intensive care units (NICU) is increasingly being recognized. It is the third most common cause of late onset sepsis in NICU patients and accounts for 9-13% of blood stream infections (BSI) in neonates.[1] Although Candida albicans has historically been the most frequently isolated species, recently non-albicans Candida (NAC) have emerged as important opportunistic pathogen, notably Candida tropicalis, C. parapsilosis, C. krusei, and C. glabrata.[23] There is growing evidence suggesting a role of increasing use of azole agents in this epidemiological shift. Several of these NAC species exhibit intrinsic resistance to traditional triazoles like fluconazole (FLK) and may also demonstrate cross resistance to newer triazoles.[4] This makes it imperative to perform both speciation and antifungal susceptibility (AFS) of all the yeast isolates from blood or otherwise. Number of factors including the use of indwelling devices, broad spectrum antibiotics, low birth weight (LBW), prematurity, total parenteral nutrition (TPN), gastrointestinal surgery, artificial ventilation, and/or history of fungal colonization contribute to the risk.[5] Preterm, very low birth weight (VLBW): ≤1,500 g; extremely low birth weight (ELBW): ≤1,000 g; and critically ill infants are at highest risk of invasive Candida infections[6] Candida spp. can also spread through vertical transmission from maternal flora or via horizontal transmission from hands of healthcare workers (HCW).[78]

Clinical presentation of candidemia resembles sepsis syndrome and to establish a clinical diagnosis is difficult.[79] Respiratory insufficiency, feeding intolerance, abdominal distention, temperature instability, lethargy, and decreased perfusion are the various clinical manifestations associated. Candida BSIs are associated with very high crude mortality of over 60%, while attributable mortality may be as high as 49%.[310] The incidence and associated mortality due to candidemia can be influenced by several factors including the population at risk, healthcare facility standards, Candida spp. involved, and antifungal resistance.[2] Due to considerable regional variability, the local epidemiological knowledge is critical in terms of prevention and management of invasive Candida infections.

We were noticing an increase in the isolation rate of NAC species over last few months from cases of neonatal septicemia, which prompted us to undertake the present study, to examine the prevalence and epidemiology of neonatal candidemia at our hospital. We also correlated our results with the associated risk factors and the clinical presentation of the neonates.

Materials and Methods

This prospective study was conducted between January 2012 and December 2012 in the Departments of Microbiology and Pediatrics of our rural tertiary care and referral center in Uttarakhand state, India. Average referrals from outside sources are approximately 200-250/year. Premature infants comprise 15-20% of total number of deliveries in the hospital.

Prior to the sample collection, approval from institutional ethical committee and written informed consent from patient's guardian were obtained. From a total of 548 neonates (age >28 days, admitted to NICU with clinical suspicion of septicemia) blood samples were collected into Bactec Peds plus/F culture vials of an automated blood culture system (Bactec 9120, Becton Dickinson, USA) and only those which yielded pure growth of Candida spp. were included in the study. Candidemia was defined as the presence of at least one positive blood culture containing pure growth of Candida spp. with supportive clinical features. Candidemia was attributed to the mortality of the neonate expired within 3 days of a positive blood culture.

Any growth indicated was subcultured on 5% sheep blood agar, MacConkey's agar, and Sabouraud's dextrose agar (SDA) with chloramphenicol (0.05%) and incubated at 37°C. The Candida spp. isolated was identified as per standard mycological techniques.[11] Preliminary identification was done by colony morphology on SDA, chromogenic media (Hichrome, Himedia Pvt. Ltd.), growth at 45°C, germ tube test, chlamydospore formation, and was confirmed by carbohydrate fermentation and assimilation tests.[11]

AFS was performed for FLK (25 mg), itraconazole (ITR, 10 mg), and amphotericin B (AMB, 100 units) using disc diffusion method[1213] on Muller-Hinton agar supplemented with 2% glucose and methylene blue (5 mg/ml). Zone diameters were interpreted as per the approved National Committee for Clinical Laboratory Standards (NCCLS) guidelines[14] Quality control for AFS was performed using C. albicans-ATCC 90028 and C. parapsilosis-ATCC 22019.

Statistical analysis

Statistical analysis was done using Statistical Package for Social Sciences (SPSS) version 11 and the prevalence of organisms was determined and expressed in percentage.

Results

A total of 381/548 (69.53%) cases were blood culture positive. Pure growth of Candida spp. was isolated from 132/381 (34.65%) cases. Pure growth of bacteria and mixed growth of bacteria and yeast were isolated from 207/381 (54.33%) and 42/381 (11.02%) of the cases, respectively.

Of the total 132 neonates included in the study 73 (55.30%) were females and 59 (44.70%) were males. The average gestational age was 32.2 weeks (30-39 weeks) and average birth weight was 1.2 kg (0.93-2.21 kg).

NAC spp. were responsible for 80.30%, whereas 19.70% of the cases were caused due to C. albicans. C. parapsilosis (25.0%), and C. tropicalis (21.97%) were the predominant NAC species isolated, followed by C. glabrata (14.39%) and C. krusei (10.61%) [Table 1]. AFS results showed 65.91% isolates sensitive to FLK, 73.49 % to ITR, and 96.21% to AMB. Five isolates (one C. tropicalis, two C. krusei, and two other Candida spp.) were found resistant to AMB. NAC species, especially C. glabrata and C. krusei were more resistant to azoles, particularly FLK, than C. albicans [Table 2].

Table 1

Characterization of various Candida species isolated from blood (n = 132)

Table 2

Antifungal susceptibility profile of Candida isolates (n = 132)

Among the risk factors observed for candidemia [Table 3], prematurity (73.49%) and LBW (67.42%) were the commonest followed by indwelling catheters (64.39%) and broad spectrum antibiotic use (61.36%). Failure to thrive (71.21%), feed intolerance (65.91%), and respiratory distress (63.64%) were the most common clinical presentations seen [Table 4]. Crude mortality due to candidemia was 34.85% as 46/132 of neonates died within 72 h of the candidemia diagnosis. Additional 13 (9.85%) neonates died post 72 h of the diagnosis of candidemia, so were not included in crude mortality. All the other patients were treated successfully.

Table 3

Potential risk factors for candidemia among neonates (n = 132)

Table 4

Various clinical presentations observed in cases of neonatal candidemia (n = 132)

Discussion

In the present study NAC species accounted for 80.30% of the cases of neonatal candidemia, whereas C.albicans was responsible for 19.70% of cases. This corroborates well with the results of other authors.[15161718] Striking feature of the present study was isolation of C. parapsilosis (25.0%) and C. tropicalis (21.97%) as the most common NAC species.

Although C. parapsilosis is less virulent, but under certain conditions (IV catheters, high IV glucose concentrations) virulence may increase many folds and it is relatively difficult to eradicate this organism.[19] This subsequently has pharmacotherapeutic and pharmacoeconomic implication as to treat such infections is quite difficult. C. parapsilosis is an emerging fungal pathogen and the major threat for neonates in NICU as it frequently colonizes the hands of HCW, has high affinity for intravascular devices, and parenteral nutrition.[2021] In the present study, 64.39 and 55.30% of patients were on indwelling catheters and TPN respectively, out of these, 14 patients had C. parapsilosis candidemia which is 42.42% of all the C. parapsilosis isolates. Higher affinity of C. parapsilosis to adhere on foreign material and ability to form biofilms are important factors for the development of fungemia.[22] Outbreaks of C. parapsilosis BSI in NICUs have been previously reported.[2324]

C. tropicalis causes infections with high mortality in adults and children with hematological malignancies or in immunocompromised individuals.[25] Ability of this organism to produce clusters is one of its major virulence factors. Once introduced into the immunocompromised host, C. tropicalis may be more virulent than C. albicans and can rapidly progress from colonization to invasion. It is the second leading cause of candidemia in adults, but is quite infrequent among neonates.[25] However, what we speculate is that premature and LBW infants have an immature immune system and may behave like an immunocompromised adult patient in this regard. In the present study, C. parapsilosis and C. tropicalis have emerged as predominant species accounting for 46.97% of infected neonates.

AFS results showed that 34.09% of Candida isolates were resistant to antifungal drugs. High degree of resistance to azoles was seen among C. krusei (78.57%) and C. glabrata (63.16%). Previous studies have reported that resistance to azole compounds among C. glabrata species can range from 3.6 to 64%.[2627] A significant proportion of C. tropicalis isolates were resistant to azoles especially FLK. Although resistance to AMB was quite low (3.79%), but is a matter of concern as emergence of such isolates may pose serious therapeutic challenges and also increases risk of nosocomial infection. A noteworthy feature observed was that 3/46 neonates who died were infected with these AMB resistant strains.

Combinations of various risk factors are known to be strongly associated with development of candidemia, and our results also suggest the same. The major risk factors identified in our study were prematurity, LBW, indwelling catheters, and broad spectrum antibiotic therapy. More than 65% of our cases were either premature or LBW, highlighting the significant burden of this disease among such infants. Use of multiple invasive devices (catheters, endotracheal tubes) or surgery causes break in the skin/mucosal integrity, which predisposes these sites for colonization/infection by Candida spp. Broad spectrum antibiotics ranging two to four in number were being administered to most of the neonates in the present study. Antibiotics promote fungal overgrowth at the expense of normal bacterial flora and encourage translocation of yeast across the intact mucosa. The risk of candidemia is known to increase exponentially with each class of antimicrobial used. TPN induces gut mucosal atrophy and has immunosuppressive effects which again predisposes individual for infection. Moreover, certain Candida spp. like C. parapsilosis has higher affinity towards parenteral nutrition, and can be responsible for outbreaks in NICUs. The hands of HCW and environmental surfaces are newly appreciated potential reservoirs for nosocomial strains of Candida spp.

Neonatal candidemia is generally associated with high mortality. In the present study, crude mortality was 34.85%, which was quite high. Even though candidemia is associated with prolonged hospitalization, we found that most of the fatal cases occurred in infants >3 weeks of age and had LBW compared with the ones who survived. Given that infants of this age have decreased immunity, their host response to Candida may have contributed to mortality. It has been observed that clearance of fungemia is directly proportional to catheter removal as soon as candidemia is detected and failure to do so is a risk factor for death.[1] Also removal of catheter alone is not sufficient and administration of antifungal agents is equally important as early treatment with antifungals is associated with enhanced survival.[2829] However in the present study, both these factors were frequent as majority of neonates who died had delay in the initiation of therapy and catheter retention for <48 h after diagnosis of candidemia which may have contributed to the high mortality.

Based on high perinatal risk factors for early onset sepsis, the current hospital antibiotic policy recommends empiric use of ampicillin-gentamicin for neonates born within the facility and cefotaxime-amikacin for neonates reffered from elsewhere. Long-term use of these broad spectrum antibiotics must have created a negative pressure and favorable environment for Candida spp. to flourish. This substantiates the need of prophylactic antifungals to be used in a set up where continuous upsurge in the incidence of candidemia is seen.

Conclusion

Candidemia in neonates is an ominous prognostic sign and is an important entity in our hospital. Preventive measures such as use of filters for parenteral nutrition, prophylactic antifungal use, and a restrictive policy of antibiotic use to decrease Candida colonization/infection rates should be implemented to reduce the morbidity and mortality associated with these infections. Reporting of fungal BSI and the spectrum of species involved are essential measures in any ICU in order to implement appropriate preventive and therapeutic strategies. Though powered to detect significant risk factors for fungal sepsis, this is a single center study and our findings may not be generalizable to other institutions. Additional studies are necessary to validate our findings and define more accurately the reservoirs, mode of transmission, emergence of new species, and their sensitivity patterns. Epidemiological data of our study can serve as a template for the development of local guidelines for prevention and appropriate treatment of neonatal candidemia.