Severe fungal infections in intensive therapy
*Danuta Gierek1, Józefa Dąbek2, Małgorzata Kuczera1, Małgorzata Marmaj1, Grzegorz Kluczewski1, Łukasz Krzych3
1Department of Anaesthesiology and Intensive Therapy, Teaching Hospital no 7, Medical University of Silesia in Katowice,
2Department of Cardiology, Medical University of Silesia in Katowice
31st Department of Cardiac Surgery, Teaching Hospital no 7, Medical University of Silesia in Katowice
Background. Identification of pathogens in severe fungal infections, by positive cultures, is usually difficult, delays appropriate therapy, and impairs treatment. Despite progress in biologic sciences, the diagnosis of candidiasis still poses a challenge. Early symptoms are not specific, and cultures are usually negative. Molecular methods are rarely used in clinical practice. Common empiric therapy of suspected fungal infection is based on examination, history, and analysis of risk factors. The aim of the study was to analyse fungal infections in ITU and to find factors which may help in their recognition.
Methods. In this retrospective study, the medical histories of ITU patients were analysed. Patients were divided into two groups: I – suspected and II – confirmed, fungal infections. The factors considered were: age, gender, suspected source of fungal infection, co-existing bacterial infection, Candida Score, laboratory tests taken on the day of fungal infection diagnosis (leukocyte count, platelets, and CRP), duration of hospitalisation, declarations of infection from departments, and results of treatment.
Results. Statistically significant differences between the groups were found in Candida Scores, duration of hospitalisation and departments of infection. The Candida Scores were higher among those patients in whom infections were already suspected. In this group, the duration of hospitalization was shorter, and infections frequently developed during hospitalisation in the ITU.
Conclusion. The analysis of infections and Candida Scores helped to initiate prompt antifungal therapy and to reduce the duration of hospitalisation. Infection markers that were routinely used in the ITU were not specific, did not allow identification of patients with fungal infection.
Despite vast advances in biologic sciences, the identification of a pathogen responsible for an infection is often very difficult. This concerns mainly fungal infections, in which the isolation of a fungus delays the institution of appropriate antibiotic therapy and impairs outcomes .
Generalized fungal infection – candidiasis in ITU patients has often the course of severe sepsis and sometimes leads to death . Its diagnosis, as in cases of infections of different etiology, is predominantly based on clinical examination. Since there are no characteristic features and isolation of a pathogen is difficult, in each case of infection whose symptoms (including fever) do not subside despite proper antibiotic therapy, the possible development of this type of infection should be considered . The risk factors include artificial lung ventilation, parenteral feeding, broad-spectrum antibiotic therapy, catheterization of vessels, abdominal surgery, neutropaenia, steroid therapy, diabetes mellitus, or high APACHE II scores (1, 3).
Fungal infections are caused by Candida albicans in 86.5% of cases, and by Aspergillus in 2.5 %; 11% of such infections are caused by other species, which also induce superficial mycoses .
Various diagnostic methods were designed for identification of pathogens, early diagnosis and treatment of fungal infections. The main of them include mycologic and serologic tests, D-arabinitol (fungal metabolite in serum) determinations, histological and molecular biology examinations .
Due to high incidence rates of infections caused by Candida albicans, the Candida Colonization Index (CCI) (>0.5) and/or Candida Score (>2.5)  are relevant elements enabling proper diagnosis.
Moreover, radiologic and bronchoscopic examinations are helpful.
In many cases, generalized candidiasis is diagnosed too late as early clinical symptoms are not specific, cultures from the suspected sites are negative, and serologic or molecular examinations are rarely used in clinical practice.
The objective of the present study was to analyse diagnostic management and treatment of patients with fungal infections and to determine the factors enabling proper diagnoses.
The retrospective study was carried out using medical records of patients treated in the Intensive Therapy Unit, Teaching Hospital no 7, Upper-Silesian Medical Centre in Katowice during the period of one year.
The criteria of fungal infection diagnosis were based on definitions accepted on the consensus conference of the American College of Chest Physicians and Society of Critical Care Medicine .
For comparative analysis, patients were divided into two groups: I − with suspected fungal infection, II − with confirmed fungal infection
The suspected fungal infection was recognized in patients with persisting symptoms of septic condition (fever, severe general condition) despite the broad-spectrum antibiotic therapy implemented, and with at least 3 risk factors, whose biological specimens did not demonstrate any fungi. The confirmed fungal infection was diagnosed in patients once microbiological examinations of the material collected from the supposed infection site showed the same fungal species twice.
The analysis in the study groups involved: age, gender, primary location of infection, etiological factor of coexisting bacterial infection, profile of patients (surgically or conservatively treated), laboratory findings on the day of infection diagnosis – WBC, PLT, CRP, duration of hospitalization, place of diagnosis (ITU or other ward), treatment outcomes. Additionally, the Candida Score was calculated according to the following rule: after surgery -1, parenteral feeding -1, severe sepsis – 2, multifocal colonization -1.
The collected data were analysed descriptively and comparatively. Quantitative variables were presented as a mean, maximal and minimal values whereas qualitative parameters as absolute values and percentages. Intergroup differences were assessed using the Student’s t test or Mann-Whitney U test for quantitative variables, χ2 and Fischer tests for qualitative variables. Normal distribution of quantitative variables was verified using the Shapiro-Wilk test. P <0.05 was considered as significant.
In general, infections were diagnosed in 153 individuals, 46.6% of all treated patients. Fungal infections were diagnosed in 24 patients (15.7%), 13 in group I and 11 in group II. There were no significant differences between groups according to age, gender or profile of patients (Fig. 1).
The respiratory system was the primary location of infections in 13 patients, the abdominal cavity in 5, the cardiovascular system in 3, soft tissues and bones in 2 and the mediastinum in 1 patient. The neoplastic disease was the dominating pathology in 2 patients treated surgically and in 1 patient treated conservatively.
In all material samples collected from the supposed site of infection, Candida albicans was detected, except for one case in which Candida tropicalis was found.
Fungal infections were treated with iv fluconazole, starting with a loading dose of 400 mg and continuing with the daily dose of 200 mg. In one case of confirmed infection, anidulafungine was administered, initially 200 mg followed by 100 mg × 1. The duration of antifungal therapy depended on the kind of infection: in suspected infections, the therapy was used for 21 days; in confirmed cases for further 7 days after the first negative mycological result.
The commonest etiologic factors of coexisting bacterial infection were: Acinetobacter baumanii – detected in 8 patients, Pseudomonas aeruginosa – 5, Klebsiella pneumoniae – 4, Stenotrophomonas maltophilia – 3, Enterobacter species – 3, beta-lactamase producing Klebsiella pneumoniae of wider substrate spectrum – 2, Escherichia coli – 2, methicillin-resistant Staphylococcus aureus − 2, methicillin-susceptible Staphylococcus aureus − 1, Staphylococcus epidermidis − 1 and methicillin-resistant Staphylococcus epidermidis – 1 patient. There were no intergroup differences in the etiological factors of coexisting bacterial infection.
The mean Candida Scores on the day of fungal infection diagnosis were: in group I – 1.07 and in group II – 2.45, the differences were significant (p = 0.02) (Fig. 2).
The mean serum levels of CRP on the day of fungal infection diagnosis were: 134.69 mg L-1 in group I, and 128.6 mg L-1 in group II, WBC – 11.07 G L-1 and 9.06 G L-1 , PLT 151.76 G L-1 and 132.07 G L-1, respectively.
The mean duration of hospitalization in patients with confirmed fungal infections was 41.18 days and was significantly longer (p=0.04) compared to patients with suspected fungal infections – 20.76 days (Fig. 3).
In group I, 5 patients were admitted with infections from other wards, in 8 the infection developed during ITU hospitalization. In group II, 9 patients were admitted with infections, and in 2 the infection was diagnosed in ITU. In the group with confirmed fungal infections, the infections were more commonly diagnosed before admission to ITU (p = 0.04).
In the group of 24 patients with fungal infections treated in ITU, the mortality was 66.7%. There were no intergroup differences in treatment outcomes (survival/death).
The number of fungal infections in patients is associated with the type of ITU and is most common in surgical and oncohaematologic wards . Our results confirm that generalized infections called invasive candidiasis affect about 15% of ITU patients [7, 8, 9]; in almost 46%, diagnosed fungal infections are confirmed by microbiological tests.
Fungal infections are treated with azoles, including fluconazole − the most popular in clinical practice, pollens – amphotericin B and its newer and less toxic forms as well as expensive and less commonly used candins. The type of pathogen cultured is essential for the proper choice of antifungal agent, as there are species, e.g. Candida glabrata and Candida krusei naturally resistant to standard fluconazole therapy . Another factor affecting the failures in fluconazole therapy is earlier use of azole drugs and the capacity of Candida species to form a biofilm inhibiting the antibiotic penetration to the fungal cell, which impairs its antifungal efficacy .
ITU patients are in severe conditions, haemodynamically unstable, with impaired immunity and numerous risk factors, therefore generalized fungal infections in them should be treated with modern antifungal candins of low toxicity .
In our study, the choice of echinocandin – anidulafungine was decided based on the severe condition of the 27-year-old patient with suspicion of candidaemia, haemodynamically unstable, and with features of hepatic failure. The patient was treated for the systemic disease and transferred to ITU due to increasing parameters of multi-organ failure and shock. In the district hospital, she had been earlier treated with azoles due to suspected fungal infection. During ITU hospitalization, the fungal infection was confirmed, Candida albicans was detected twice in the bronchial washings collected bronchoscopically. Thanks to extensive treatment – artificial lung ventilation, intensive pharmacotherapy with catecholamines, broad-spectrum antibiotics and targeted antifungal therapy, her general condition improved and she recovered.
Fungal infections confirmed by microbiological examinations are common in patients with lung cancer and chronic obstructive pulmonary disease . The risk factors in such cases include steroid therapy, immunosuppression, artificial lung ventilation, broad-spectrum antibiotic therapy. The literature data show that the primary site of infection within the abdominal cavity, particularly in patients after surgery, is the factor predisposing to the development of invasive fungal infection. In the European study on fungal infections, abdominal surgery was found the most important risk factor accounting for 48% of candidaemia cases [9, 13, 14]. Our study revealed that the primary site of infection and profile of patients were essential. This was likely to be associated with the specificity of surgical procedures performed in our surgical ward – vascular procedures, which do not induce the discontinuity of the alimentary tract, enabling early institution of enteral feeding and limitation of parenteral feeding, one
of the candidiasis risk factors.
Our findings demonstrated that Candida Score evaluated on the day of fungal infection diagnosis was significantly higher in the group with confirmed fungal infections. It was shown that patients with the Candida Score >2.5 were most commonly predisposed to generalized candidiasis . In such patients, the antifungal therapy should be considered before microbiological confirmation is available. Since the mortality in fungal infections is high, even 70%, early institution of therapy is recommended based on the simple scoring, which should improve the outcomes in this group of patients .
The clinical symptoms of fungal infections are poorly characteristic and cultures from the suspected sites of infections are often negative; therefore, the laboratory indices are being searched for to define the state of immune response. Laboratory indices of generalized infections, except for procalcytonine, are not specific as their increase is observed in such conditions as trauma, surgery, burn, and pancreatitis. Procalcytonine was shown to be the most useful parameter of differential diagnosis in fungal infections, particularly in the initial stage of disease . Moreover, it was observed that an increase in this parameter is strictly correlated with the development of invasive candidiasis . However, in cases of fungal superinfections and coexisting bacterial infections, procalcytonine is not the factor differentiating these types of infections.
Close observation of ITU patients, i.e. daily evaluation of clinical state, laboratory indices of inflammation, meticulous analysis of bacteriological findings enabled early diagnosis of septic conditions and institution of appropriate therapy to inhibit the growth of fungi. In the material analysed, such an infection was diagnosed during ITU hospitalization in the majority of patients with suspected fungal infections. It may be supposed that for these reasons those pathogens were not microbiologically confirmed. Amongst patients with confirmed fungal infections, in the majority of cases the diagnosis was established in other wards, before ITU hospitalization. Less scrupulous observation of patients and worse bacteriological surveillance in general wards might have delayed the diagnosis and institution of suitable therapy leading to increased multiplication of the pathogen. Therefore, in this group of patients, fungal infections were confirmed microbiologically. The obtained results are in accordance with the observations of other authors who have demonstrated that earlier therapy of patients in general wards is the factor predisposing to the development of candidaemia .
The Internet system for reporting severe sepsis, introduced in 2003 by the Polish Working Group for Sepsis, revealed that the mortality rates in this group of patients reached 55% and the mean hospitalization duration – 19 days . In our study, the mean time of hospitalization of patients with confirmed fungal infections was 41.1 days, which increased the ward costs. There is hope that results of the studies being carried out will contribute to identification of patients with fungal infections, in whom the institution of early targeted antibiotic therapy should improve the therapy outcomes.
1. Meticulous analysis of the course of infection and its assessment according to the Candida Score enable early institution of antifungal therapy and shortening of hospitalization.
2. Routine laboratory indices of inflammations used in intensive therapy are not sufficiently specific and do not allow to identify patients with fungal infections.
3. A clinical diagnosis of fungal infection is of high importance due to lack of explicit differences between patients with confirmed and suspected fungal infections.
1. Shorr AF, Tabak YP, Johannes RS, Sun X, Spalding J, Kollef MH: Candidemia on presentation to the hospital: development and validation of a risk score. Critical Care 2009; 13: R 156.
2. Ciok-Pater E, Gospodarek E: Zróżnicowanie gatunkowe grzybów Candida spp. w materiałach pobranych od pacjentów leczonych w wybranych klinikach Szpitala Uniwersyteckiego im. A. Jurasza w Bydgoszczy w latach 2003-2005. Medical and Biological Sciences 2007; 21: 37-42.
3. Piotr Sowiński: Zakażenia grzybicze w OIT; w: Zakażenia grzybicze – wybrane zagadnienia. (Red.: Dzierżanowska D.), α-medica press, Bielsko Biała, 2006; 126-136.
4. León C, Ruiz-Santana S, Saavedra P, Almirante B, Nolla-Salas J, Alvarez-Lerma F, Garnacho-Montero J, León MA; EPCAN Study Group: A bedside scoring system („Candida score”) for early antifungal treatment in nonneutropenic critically ill patients with Candida colonization. Crit Care Med 2006; 34: 730-737.
5. Kűbler A: Sepsa – definicja i epidemiologia. Med Sci Rev 2005; 1: 5-10.
6. Bedlin A, Venturelli C, Mussini C, Guaraldi G, Codeluppi M, Borghi V, Rumpianesi F, Barchiesi F, Esposito R: Epidemiology of candidemia and antifungal susceptibility patterns in an Italian tertiary – care hospital. Clin Mikrobiol Infec 2006; 12: 75-80.
7. Bille J, Marchetti O, Calandra T: Changing face of health-care associated fungal infections. Curr Opin Infect Dis 2005; 18: 314-319.
8. Eggimann JP, Garbino G, Pittet D: Epidemiology of Candida species infections in critically ill non-immunosuppressed patients. Lancet Infect Dis 2003; 11: 685-702.
9. Tortorano AM, Kibbler C, Peman J, Bernhardt H, Klingspor L, Grillot R: Candidaemia in Europe: epidemiology and resistance. Int J Antimicrob Agents 2006; 27 5: 359-366.
10. Szymankiewicz M, Kowalewski J: Ocena zdolności do wytwarzania śluzu przez grzyby Candida albicans, w zależności od miejsca izolacji szczepów. Mikol Lek 2005; 12 4: 249-252.
11. Guery BP, Arendrup MC, Auzinger G, Azoulay E, Borges Sá M, Johnson EM, Müller E, Putensen Ch, Rotstein C, Sganga G, Venditti M, CrespoRZ, Kullberg BJ: Management of invasive candidiasis and candidemia in adult non-neutropenic intensive care unit patients: part II treatment. Intensive Care Med 2009; 35: 206-214.
12. Batura-Gabryel H: Inwazje wieloogniskowe grzybami z rodzaju Candida u chorych na raka płuca i przewlekłą obturacyjną chorobę płuc. Mikol Lek 1999; 3: 21-24.
13. Calandra T, Marchetti O: Clinical trials of antifungal prophylaxis among patients undergoing surgery. Clin Infect Dis 2004; 39: 185-192.
14. Safdar N, Maki DG: The commonality of risk factors for nosocomial colonization and infection with antimicrobial-resistant Staphylococcus aureus, enterococcus, Gram-negative bacilli, Clostridium difficile, and Candida. Ann Intern Med 2002; 136: 834-844.
15. Balcl C, Sungurtekin H: Usefulness of procalcitonin for diagnosis of sepsis in the intensive care unit. Critical Care 2003; 7: 85-90.
16. Charles PE, Castro C, Ruiz-Santana S, Leon C, Saavedra P, Martin E: Serum procalcitonin levels in critically ill patients colonized with Candida spp: new clues for the early recognition of invasive candidiasis? Intensive Care Med 2009; 35: 2146-2150.
17. Kubler A, Durek G, Zamirowska A, Pałysińska B, Gaszyński W, Pluta A: Severe sepsis in Poland - results of internet surveillance of 1043 cases. Med Sci Monit 2004; 10: 35-41.
Oddział Anestezjologii i Intensywnej Terapii
SPSK nr 7, Śląski Uniwerystet Medyczny
GCM im. prof. Leszka Gieca
ul. Ziołowa 45/47, 40-635 Katowice-Ochojec
tel. 32-359 81 00