ročník 11,2003 č.4

Pre-emptive therapy of infections caused by yeasts in surgical ICUs

Havlíček K., Sákra L., Motyčka V., Mencl, K.

Department of Surgery, Regional Hospital
Health Care Studies Institute
Department of Microbiology, Regional Hospital
Pardubice, Czech Republic

Grant Project (Identification Code: ND 6884-3) for the years 2001-2003 is being solved with the aid of IGA Czech Republic Ministry of Health.


The authors evaluate the incidence of yeast infections at the surgical ICU during the first two years of the three-year grant project. From a total of 3389 patients hospitalized in the ICU of the Department of Surgery, Pardubice, 852 have been included into the study (their parameters complied with the study requirements regarding main diagnosis, performed operative procedure, and concomitant diseases). Using a scoring scheme, high risk of yeast infection development has been assessed in 210 of them. These 210 patients were randomized into two groups. In the group A the patients received pre-emptive treatment with itraconazole, group B served as control. Our microbiological and clinical monitoring concerned the incidence of yeast colonization in different organic systems, the incidence of yeast infections and of disseminated yeast infections (bloodstream infections). Detailed assessment of the yeast species gamut took place. In group A receiving pre-emptive itraconazole treatment 52.3 % of patients had negative mycological tests, while in group B only 38.5 % were tested negative. As to colonization, we could find it in 21.2 % of patients in the group A whereas in group B this was the case only in 12.8 %. We diagnosed infection with yeast organisms in 25.7 % of subjects from group A, and in 42.3 % from group B. Septicaemia caused by yeasts was proven in 8 (12.8%) patients of group B, while in group A this serious condition appeared only in one case (0.7%). From the presented results obtained from our set of patients we conclude that pre-emptive therapy with antimycotics unequivocally reduces the number of developed forms of infections caused by yeasts in patients at risk. This is most apparent with regard to yeast septicaemia.
Keywords:Candida - yeast infection - yeast colonization - intensive care


Having regard to the permanently increasing incidence of nosocomial yeast infections in surgical ICUs, the research project deals with the problem of systemic yeast infections, called also deep yeast infections, especially candidosis.
Taking in account the difficult diagnostics, and the need to differentiate between diverse clinical forms, clear definition of the terms colonization, candidaemia, candidosis, disseminated candidosis, Candida spp. septicaemia is necessary. Colonization means isolation of yeasts from two, or more sites without clinical symptoms. Finding Candida spp. in one site or system (e.g. skin or the gastrointestinal tract) at a low quantitative level is not considered to represent a clinical unit; this is supposed to be a normal condition - commensalism, or epiphytism of the yeasts. Candidaemia (bloodstream infection) is defined as the detection of Candida spp. in the circulating blood, one single positive blood culture being sufficient to proclaim this diagnosis. Candidosis designates the condition when Candida spp. are found in sites normally sterile or where Candida spp. do not occur normally. Disseminated candidosis means invasion of an organ with a Candida sp., not by direct contact but as a result of candidaemia which must always precede. The term Candida spp. septicaemia is considered incorrect; it only describes the clinical outcome of one of the above conditions. The need to differentiate between these conditions is a consequence of the fact that presence of Candida spp. at some sites of the human organism is considered to be normal, and only 44 % of patients with Candida spp. colonization can be supposed to develop some of the more serious forms later.
In clinical practice, prevention and therapy of yeast infections is divided into 4 types: Prophylaxis, pre-emptive therapy, empirical treatment and targeted or definitive treatment.
Our study is interested in pre-emptive treatment.
Prophylaxis is defined as preventive therapy performed in all patients at risk of developing candidaemia or disseminated candidosis. However, no unanimity exists concerning risk factors specification, and the assessment of their importance. There are neither studies, nor papers evaluating the contribution of this therapy for patients at the ICU.
Pre-emptive therapy is defined as antimycotics administration (also with preventive purpose) but only in selected patients (who are at high risk of candidaemia or disseminated candidosis development), and in therapeutic doses, not only for decontamination. The various studies differ in risk factors assessment. Mostly, the following risk factors are indicated: repeated broad-spectrum antibiotics administration, renal impairment, central vein catheters, candiduria, high score APACHE II, and concomitant serious disease. From the above definitions results clearly that these two types of therapeutic approach are overlapping one another, and it depends mainly on the risk factors designation as well as on the distinction of their importance concerning candidaemia occurrence. There isn't any more relevant study, neither any paper, which would try to evaluate the contribution of these two therapy types. Studies concerned with this issues dispose only of small numbers of cases, and therefore no statistically significant data can be obtained from them.
Empirical therapy is used in cases of supposed candidaemia or disseminated candidosis without microbiological or histological proof. Most often we indicate this therapy in patients with risk factors of Candida spp. infections development in whom non-remittent fever, and high CRP levels persist, and where we have no other explanation for the related symptoms.
Targeted or definitive therapy is indicated in patients with histological or mycological Candida spp. infection. Such therapy should be absolutely initiated in patients with
- histological confirmed presence of yeasts in tissues;
- microbiological proof from sites, secretions, or tissues which under normal circumstances are sterile;
- microscopic proof of yeasts in sites, secretions, or tissues which under normal circumstances are sterile;
- one positive blood culture.

Integral part of the treatment should be change of all venous catheters including change of the entrance port, change of urinary catheters and of other interventional ports.

Purpose of the study

The goal of the study was to verify and to confirm the efficacy of pre-emptive therapy of yeast infections at surgical ICUs. To prove that using this method of antimycotics administration reduces the incidence of candidaemia, candidosis and disseminated candidosis, decreases morbidity and mortality of the critically ill. Another objective of this project was also to determine if the more economic oral form - tablets or solutions applicable per nasogastric tube - is sufficient to prevent the development of clinically relevant yeast infections in patients at risk.

Patients and methods

The trial was carried out at the ICU of the Department of Surgery, Regional Hospital, Pardubice disposing of 21 beds. It is divided into two parts. The ICU 1 has 8 beds in single boxes, and it is aseptic. Here we use to treat patients after vascular and some thoracosurgical operations, polytraumatic patients, or patients with serious monotrauma, patients after endoprosthesis implant, and after neurosurgical operations. ICU 2 is mesoseptic, there are 13 beds used for the management of patients who underwent larger abdominal and thoracic operations.
In every ICU patient periodical sampling for mycological examination is performed. On the very first day of patients' ICU stay smears from throat, nose and rectum are taken, as well as samples of urine, and of all discharges from wounds and drains. In addition a tracheostomy smear is taken if tracheostomy has been performed; in ventilated patients a sample of bronchial secretion is sent for examination. The like, all removed ports are subject to mycological examination. The said sampling is repeated periodically every 7 days. Microbiological and mycological examination is ensured by the Department of Microbiology, Regional Hospital, Pardubice.
Yeast infection incidence risk factors assessment has been performed on the ground of a retrospective study, and a scoring scheme was established. Patients scoring 10 and more are considered to be at high risk of yeast infection development.
The patients complying with the requirements for inclusion into the study were randomized into two groups. In the first group (hereafter group A) patients received pre-emptive treatment with oral itraconazole (Sporanox sol., Sporanox cps., 100 mg 2 x / day) as soon as their score reached 10. Itraconazole was chosen as a broad spectrum preparation covering also some Candida species less susceptible to fluconazole. Moreover, itraconazole has not been yet used routinely on the surgical ICU, and hereby it complies with the microbiological criteria for a medicament convenient for prophylactic or pre-emptive administration. The oral form was chosen because of the low economic demandingness of this pre-emptive administration method. We did it although we knew that from the administration of this drug per nasogastric tube may ensue problems regarding absorption of the active substance, e.g. from an atonic stomach.
The second group (hereafter group B) was randomized to be a control group consisting of patients scoring 10 and more but not receiving itraconazole.


The data submitted in this paper represent results covering the period from 1 June 2001 to 31 October 2002.
During this period, 3389 patients were hospitalized in the ICU of the Department of Surgery, Regional Hospital, Pardubice. Among them, the parameters of 852 patients complied with the project assignment (main diagnosis, concomitant diagnoses, or performed operative procedure). Among these 852 patients we found 210 patients (130 men, 80 women) with a final score of 10 points, or more.
Into group A, where itraconazole was administered accordingly to the above described schedule, 132 patients (78 men, 54 women) were included; their average ICU hospitalization period amounted to 10.6 days. In this group, 69 patients (52.3 %) had negative mycological examinations. One positive blood culture occurred in this group; with this one exception no further signs of Candida septicaemia were observed. Yeast colonization occurred in 28 patients (21.2 %). The data of culture results from biologic materials are shown in table 1.

Site Number (n = 28), 21.2 %
Sputum 24
Urine 4
Wound discharge 0
Faeces 18
Tab. No.1 - Yeast colonization in group A - occurrence site

The incidence of identified yeast species in these persons is demonstrated in table 2.

Species Number
Candida albicans 25
Candida tropicalis 4
Candida glabrata 11
Candida kefyr 1
Tab. No.2 - Yeast colonization in group A - yeast species

In the group A, infection with yeasts occurred in 34 patients in all (25.7 %). The spectrum of biologic materials is demonstrated in table 3, and the specification of found yeasts is in table 4.

Site Number (n = 34), 25.7 %
Sputum 27
Urine 26
Wound discharge 14
Faeces 14
Blood culture 1
Tab. No.3 - Yeast infection in group A - the occurrence site

Species Number
Candida albicans 29
Candida tropicalis 7
Candida krusei 3
Candida glabrata 6
Candida kefyr 1
Tab. No.4 - Yeast infection in group A - yeast species

Into group B where itraconazole was not administered, 78 patients (52 men, 26 women) were entered; their average ICU hospitalisation period was 15.2 days - the final average hospitalisation period is skewed by 1 patient who was hospitalized at the ICU for 126 days. Positive blood cultures or clinical diagnosis of clearly demonstrable septicaemia in conjunction with microbiologically proven invasive yeast infection occurred in 8 cases (10.3 %). In compliance with sensitivity tests, these infections were treated either with fluconazole, or amphotericin B . Yeast colonization occurred in 10 patients (12.8%). In table 5 the sites of positive findings are given, while table 6 demonstrates the involvement of single species.

Site Number (n=10), 12.8%
Sputum 8
Urine 1
Wound discharge 0
Faeces 8
Tab. No.5 - Yeast colonization in group B - occurrence site

Species Number
Candida albicans 7
Candida krusei 1
Candida glabrata 3
Candida prapsilosis 1
Trichosporum asahii 1
Candida kefyr 1
Tab. No.6 - Yeast colonization in group B - yeasts species

In the group B, yeast infection was identified in 33 patients (42.3 %). Positive sites, and there identified yeast species are indicated in tables 7 and 8.

Site Number (n=33), 42.3%
Sputum 27
Urine 25
Wound discharge 13
Faeces 19
Blood culture 4
Sepsis without positive blood culture 3
Tab. No.7 - Yeast infection in group B - occurrence site

Species Number
Candida albicans 29
Candida tropicalis 4
Candida krusei 1
Candida glabrata 11
Candida kefyr 1
Candida famata 1
Tab. No.8 - Yeast infection in group B - yeasts species

In the group A (group with pre-emptive itraconazole treatment); mycological examination resulted negative in 52.3 % of patients compared with only 38.5 % in the group B. As far as colonization is concerned, we showed this phenomenon in 21.2 % of patients in group A, and 12.8 % in group B. Yeast infection was diagnosed in 25.7 % of subjects in group A, while in group B this number was 42.3%. Yeast septicaemia was registered in 8 patients group B (12.8 %), whereas in the group A this serious condition was proven only once (0.7 %).


Considering, how difficult and expensive the treatment of fully developed yeast infections may be, as well as the pitfalls of large scale prevention, the use of pre-emptive therapy gains importance ever more.
The submitted results obtained from monitoring the studied patient file permit to draw the conclusion that pre-emptive treatment with antimycotics in patients at risk unequivocally reduces the number of advanced yeast infections and most of all the development of yeast septicaemia.


  1. Cohen, J. et al, British Society for Antimicrobial Chemotherapy Working Party. Management of Deep Candida Infection in Surgical and Intensive Care Unit Patients, Intensive Care Med, 1994, 20, s. 522-528
  2. EPIC Study Group. The EPIC study - preliminary results. HOST 1992, 9, s.12-13
  3. Flanagan, P.G., Barnes, R.A. Fungal infection in the intensive care unit. Journal of Hospital Infection, 1998, 38, s.163-177
  4. Gelfand, M.S. Candidiasis: Implications of Intensivists and Pulmonologists. Seminars in Respiratory and Critical Care Medicine, 1997, Vol. 18. NO. 3
  5. Motyčka, V., Sákra, L., Mencl, K., Havlíček, K. Výskyt kvasinkových infekcí na chirurgické JIP [Yeast infection incidence in a surgical ICU], Rozhl. Chir., 2001, roč. 80, č. 6, s. 320-323
  6. Petri, M.G., Konig, J., Moecke, H.P. et al. Epidemiology of Invasive Mycosis in ICU Patients: A Prospective Multicentre Study in 435 Non-neutropenic Patients, Intensive Care Med, 1997, 23, s.317-325
  7. Pittet, D., Monod, M., Suter, P., Frenk, E., Auckenthaler, R. Candida Colonisation and Subsequent Infection in Critically Ill Surgical Patients, Ann Surg, 1994, 220, s. 751-758
  8. Robin, R.H., Pre-emptive Therapy in Immunocompromised Hosts, N Engl J Med, 1991, 324, s. 1057-1058
  9. Sákra, L., Motyčka, V., Mencl, K., Havlíček, K. Kvasinkové infekce na chirurgické JIP [Yeast infections at the surgical ICU], Rozhl. Chir., 2000, roč. 79, č. 4, s. 179-185
  10. Solomkin, J., Flohr, A., Simmons, R.L. Indications for therapy for fungaemian postoperative patients, Arch Surg 1982, 117, s. 1272-1275
  11. Vincent J.L. et al. Epidemiology, diagnosis and treatment of systemic Candida infection in surgical patients under intensive care, Intensive Care Med, 1998, 24, s. 206-216

Addresse for correspondence:

Prof. MUDr. K. Havlíček, PhD.
Department of Surgery, Regional Hospital Pardubice,
Czech Republic