Epidemiology of source-related candidaemia and breakthrough candidaemia in intensive care units Authors: Ayşe Sena Şişman, Uğur Önal, Büşra Çalışır, Emel Yılmaz, Esra Kazak, Yasemin Heper, Güven Özkaya, Remzi İşçimen, Cüneyt Özakın, Beyza Ener, Halis Akalın Published: 2026 ABSTRACT AbstractBackgroundA review of the literature reveals that there is no descriptive epidemiological study based on the source of candidaemia in intensive care units(ICUs). In addition to that the number and scope of studies evaluating breakthrough candidaemia in ICUs are limited. The primary objective of this study was to analyse the epidemiology of candidaemia in our ICUs based on the source of each case, and the secondary aim was to evaluate breakthrough candidaemia.Methods141 adult patients(≥ 18 years) who had Candida spp. growth in their blood cultures 48 h after admission to an ICU were included in the study. Statistical comparisons were made between groups for abdominal, central venous catheter and primary candidaemia. The level of significance was set at α = 0.05.ResultsIn the classification based on the sources of candidaemia, 52 (37%) were abdominal, 32 (23%) were central venous catheter-related, 55 (39%) were primary candidaemia and 2 (1%) were urinary tract-related. Total parenteral nutrition use rates were significantly lower in primary candidaemia than in abdominal candidaemia (p < 0.001). Steroid use rates were significantly higher in primary candidaemia than in abdominal candidaemia (p = 0.021). Mortality rates on days 14 and 28 did not differ significantly according to the source of infection. Beakthrough candidaemia rates were significantly higher in abdominal candidaemia. The acute physiology and chronic health evaluation II (APACHE-II) score at ICU admission was significantly higher in the breakthrough candidaemia(p = 0.037).ConclusionsA significant proportion of candidemia cases in our intensive care units were non–catheter-related candidemia. Breakthrough candidaemia was found to be more common in patients with high APACHE-II scores at ICU admission and in candidaemia originating from the abdomen.Keywords: Candidaemia, Fungal infections, Intensive care unit, Breakthrough candidaemia Introduction Recently, invasive fungal infections are on the rise due to increasing comorbidities and immunosuppressive treatments. It is estimated that approximately three million deaths worldwide are caused by such infections each year. Approximately 70% of all invasive fungal infections are reported to be caused by Candida spp [1]. In multicentre prospective study, the incidence of candidaemia in surgical patients in ICUs was reported to be 9 per 1,000 patient admissions [2]. In the candidaemia/intra-abdominal candidiasis in European ICU (EUCANDICU) study conducted by Bassetti et al., which included 23 ICUs from different European countries, the incidence of candidaemia was found to be 5.52 per 1,000 patient admissions [3]. In ICU-acquired candidemia, identifying the source of infection is crucial for guiding source control, predicting the likely Candida species, selecting appropriate antifungal therapy, and developing policies to reduce the incidence of candidemia [4]. A review of the literature reveals that, in general, there is no descriptive epidemiological study based on the source of candidaemia in ICUs. Candidemias that develop during prophylactic, pre-emptive, empirical, or targeted antifungal therapy are defined as breakthrough candidemia. They occur more frequently in immunosuppressed, neutropenic patients, in those with mucositis, and in patients receiving fluconazole treatment. Non-albicans strains are more commonly seen in breakthrough candidemias. Breakthrough candidemias are associated with antifungal resistance and an unfavorable prognosis [5]. The number of studies that have evaluated breakthrough candidemias developing in the intensive care unit in detail is quite limited [2, 6, 7]. The primary objective of this study was to analyse the epidemiology of ICU-acquired candidaemia in our ICUs based on the source of each case, and the secondary aim was to evaluate breakthrough ICU-acquired candidaemia. Methods This study included adult (≥ 18 years) patients who developed candidaemia in adult ICUs between 2004 and 2023. The study was approved by the Clinical Research Ethics Committee of Bursa Uludag University Faculty of Medicine on May 3, 2023, with decision number 2023-10/4. Since the study was retrospective, informed consent was not obtained from the patients. Bursa Uludag University Faculty of Medicine Hospital is a 900-bed tertiary university hospital accredited by the Joint Commission International in 2007 and 2012. Patients followed up in adult ICUs (Anaesthesiology and Reanimation, General Surgery, Neurosurgery, Thoracic and Cardiovascular Surgery, Cardiology) with Candida spp. growth in their blood cultures were retrospectively screened from our hospital’s mycology laboratory archive. Patient’s demographic and clinical data were obtained from medical records in the hospital’s electronic file system. Adult patients with Candida spp. growth in blood cultures taken 48 h after admission to the ICU were included in the study; the exclusion criteria were mixed growth(growth of different Candida species in consecutive blood cultures obtained within 7 days), a diagnosis of COVID-19, inaccessible electronic data and those under the age of 18. In patients with multiple candidaemia episodes, the first episode was included in the study. Definitions Candidaemia: Defined as Candida spp. growth in at least one blood culture. The day the blood culture was taken was considered the first day of candidaemia. Candidaemia developing in the ICU: Candidaemia developing 48 h after the patient‘s admission to the ICU [3]. Neutropenia: Defined as an absolute neutrophil count of ≤ 500/mm³ in a complete blood count. Prolonged steroid use: Defined as the use of prednisolone or equivalent steroids at a dose of > 20 mg/day for more than three weeks. European Confederation of Medical Mycology(ECMM) quality of clinical candidaemia management (known as EQUAL) Candida score: A scoring system developed by the ECMM to measure compliance with guidelines in the management of candidaemia [8, 9]. It consists of a total of 10 criteria under the steps of diagnosis, treatment and follow-up. Candidaemia developing under antifungal treatment (breakthrough candidaemia): This is defined as Candida spp. growth in patients who have been receiving antifungal treatment for at least 72 h at the onset of candidaemia [7]. Concomitant bacteraemia: This was assessed by screening blood cultures taken within 48 h prior to the onset of candidaemia. Empirical treatment: This is defined as treatment initiated upon suspicion of candidaemia, while targeted treatment is defined as treatment initiated following Candida spp. growth in blood cultures. Central venous catheter-associated candidaemia: Defined as the growth of the same pathogen in blood cultures obtained from both the peripheral venous route and the central venous catheter [7]. Intra-abdominal Candida infection: Defined as the presence of any of the following conditions [10, 11]. presence of primary/secondary/tertiary peritonitis, presence of primary/secondary/tertiary peritonitis, growth of Candida spp. in cultures taken from the abdomen during abdominal surgery, growth of Candida spp. in cultures taken from the abdomen during abdominal surgery, presence of perforation/leakage in the bile ducts or gastrointestinal tract, or. presence of perforation/leakage in the bile ducts or gastrointestinal tract, or. presence of the same pathogen in cultures taken from drains placed within the first 24 h after surgery. presence of the same pathogen in cultures taken from drains placed within the first 24 h after surgery. The source of candidemia was accepted to be abdominal(abdominal candidemia) in the presence of intra-abdominal Candida infection without other sources. Urinary-related candidaemia: Defined as the growth of the pathogen in blood cultures after the procedure, which was present in urine cultures prior to urological instrumentation [12]. Primary candidaemia: Patients who did not fit into any of the three groups or whose source of candidaemia could not be clearly identified (presence of tracheoesophageal fistula, skin-soft tissue, deep tracheal aspirate, urinary colonisation, etc.). To better assess the clinical status of patients with candidaemia, acute physiology and chronic health evaluation II (APACHE-II) scores were calculated on the first day of intensive care follow-up. The clinical picture on the day of candidaemia onset was classified according to Sepsis-1 criteria [13]. Epidemiological evaluation was performed in four five-year periods. Candidaemia rates were calculated based on 1,000 patient admissions and 10,000 patient bed days. Overall mortality was assessed at 14 and 28 days. Statistical comparisons were made between groups for abdominal, central venous catheter and primary candidaemia. Urinary tract-related candidaemia was excluded from statistical analysis due to the small number of cases. Epidemiological assessment, incidence rates, and sources of candidaemia were evaluated in 141 patients. Breakthrough candidaemia was evaluated after excluding 10 patients who did not receive antifungal therapy. These 10 patients were excluded from the breakthrough candidemia analysis because they died within 48 h after blood cultures were obtained, did not receive antifungal therapy, and did not meet the criteria for breakthrough candidemia. Microbiological methods Blood cultures obtained from patients were incubated for five days in the BD BACTEC FX top (Becton-Dickinson™, USA) automated blood culture system. Gram staining was performed when a growth signal was detected in the blood culture during the incubation period. Concurrently, passages were made to 5% sheep blood agar medium and eosin methylene blue (EMB; BD) medium and incubated at 37 °C for 24–48 h. Additionally, cultures with Gram staining morphology resembling yeast were also inoculated onto Sabouraud dextrose agar (BD) containing gentamicin and chloramphenicol. Yeast colonies growing on the inoculated media were first identified using conventional methods (germ tube test, cornmeal tween 80, etc.). In addition to these methods, isolates were also identified using API ID 32 C (BioMérieux, France) until 2020 and matrix-assisted laser desorption ionisation time of flight mass spectrometry (MALDI TOF MS; Bruker Daltonics, Bremen, Germany) from 2020 to the present. The in vitro susceptibilities of the isolates to fluconazole, voriconazole, itraconazole, posaconazole, amphotericin B, micafungin and anidulafungin were determined using the liquid microdilution method recommended in the CLSI M27-A3 guideline. The minimum inhibitory concentration data obtained were interpreted using species-specific epidemiological breakpoint values or clinical cutoff values according to the CLSI M60 document. Statistics Due to the retrospective nature of the study design, sample size calculations were not performed. The study retrospectively analyzed adult patients who developed candidemia in our intensive care units between 2004 and 2023. The Shapiro-Wilk test was used to examine whether the data were normally distributed. Descriptive statistics were presented as mean, standard deviation and median (minimum-maximum) values for quantitative data, while qualitative data were presented as frequencies and percentages. In independent two-group comparisons, the t-test was used for normally distributed data, and the Mann-Whitney U test was used for variables that were not normally distributed. In more than two-group comparisons, one-way analysis of variance was used for normally distributed variables, and the Kruskal-Wallis test was used for variables that were not normally distributed. Categorical data were analysed using the Pearson Chi-square, Fisher-Freeman-Halton and Fisher‘s Exact Chi-square tests. Significant results were evaluated using the Bonferroni test, a multiple comparison test. Relationships between variables were examined using the Spearman correlation coefficient. The level of significance was set at α = 0.05. Statistical analysis of the data was performed using IBM SPSS 29.0.2.0 (IBM Corp. Released 2023. IBM SPSS Statistics for Windows, Version 29.0.2.0 Armonk, NY: IBM Corp.) statistical package program. Results One hundred forty-one patients were included in the study. 71 (50.4%) of the patients were male and 70 (49.6%) were female. The mean age was 56.2 ± 18.4 years (range, 19–95 years). The mean total hospital stay and the mean ICU stay were 30.9 ± 37.4 days (range, 2–321) and 20.6 ± 34.4 days (range, 2–321), respectively. The mean APACHE-II score on ICU admission was 23.67 ± 7.92 (range, 4–46). A total of 63,944 patient admissions were observed in ICUs during the study period. The incidence of candidaemia was 2.2 per 1,000 patient ICU admissions and 4.1 per 10,000 patient hospitalisation days in the ICU. The number of candidaemia episodes and rates by department are shown in Table 1. Table 1Candidaemia incidence according to intensive care unitsCandidaemia(n)Patients(n)HospitalizationDaysCandidaemia Incidenceper 1000 Patient AdmissionsCandidaemia Incidenceper 10,000 Hospitalization DaysAnaesthesiology and Reanimation ICU795589110,62414,17,1General Surgery ICU44864741,6655,110,6Neurosurgery ICU310,26152,9280,30,6Thoracic andCardiovascular Surgery ICU1318,79250,6880,72,6Cardiology ICU220,65586,4590,10,2Total14163,944342,3642,24,1ICU: intensive care unit Candidaemia incidence according to intensive care units Thoracic and Cardiovascular Surgery ICU ICU: intensive care unit Analysing comorbidities revealed cardiovascular disease in 40 (28%) patients, hypertension in 33 (23%), solid organ malignancy in 28 (20%) and diabetes mellitus in 27 (19%). No patients were diagnosed with endocarditis. C. albicans ranked first at 39% (55/141), C. parapsilosis was second at 33.4% (47/141) and C. glabrata was third at 16.3% (23/141). The distributions of C. albicans and non-C. albicans spp. are shown by period in Fig. 1. Candida species based on periods In all, 52 cases (37%) were diagnosed as abdominal candidaemia, 32 (23%) as central venous catheter-related, 2 (1%) as urinary candidaemia and 55 (39%) as primary candidaemia. When looking at the species distribution according to sources, C. albicans was the leading cause in abdominal and primary candidaemia cases (42%), while C. parapsilosis was the leading cause in central venous catheter-related candidaemia (44%) (Table 2). Table 2Distribution of candida species based on candidaemia sourcesCandida albicansCandida tropicalisCandida parapsilosisCandida glabrataCandida kruseiCandidakefyrCandida pelliculosaCandida lusitaniaeCandida haemuloniiAbdominal candidaemia22 (42%)2 (4%)15 (29%)11 (21%)1 (2%)1 (2%)Primary candidaemia23 (42%)4 (7%)18 (33%)6 (10%)2 (4%)1 (2%)1 (2%)CVC-associated candidaemia9 (28%)3 (9%)14 (44%)5 (16%)1 (3%)Urinary tract-related candidaemia1 (50%)1 (50%) Distribution of candida species based on candidaemia sources A significant difference was found in the European Confederation of Medical Mycology quality of clinical candidaemia management (EQUAL) Candida score between abdominal, central venous catheter-associated and primary candidaemia cases (p = 0.010). The EQUAL Candida score was significantly higher in central venous catheter-associated candidaemia than in abdominal candidaemia cases but similar to primary candidaemia cases. Total parenteral nutrition use rates were significantly lower in primary candidaemia cases than in abdominal candidaemia cases, but similar to central venous catheter-associated candidaemia cases (p < 0.001). Steroid use rates were significantly higher in primary candidaemia cases than in abdominal candidaemia cases, but similar to central venous catheter-associated candidaemia cases (p = 0.021). Cardiovascular disease rates were significantly higher in central venous catheter-associated candidaemia compared to abdominal candidaemia, but similar to primary candidaemia (p = 0.027). Twenty-six cases of candidaemia with concomitant bacteraemia growth were detected. No difference was found between accompanying bacteremia in cases of abdominal, primary, and central venous catheter-associated candidaemia (p = 0.170). No significant differences were observed when evaluating Gram-positive and Gram-negative growths (p = 0.457 and p = 0.442, respectively). The 14-day and 28-day mortality rates for all candidaemias were 41.8% (n = 59) and 54.6% (n = 77), respectively. When mortality was examined according to candidaemia sources, the highest mortality was observed in primary candidaemia, with 14-day and 28-day mortality rates of 49% and 58%, respectively. In abdominal candidaemia, the 14th and 28th day mortalities were 44.2% and 55.7%, respectively, while in central venous catheter-associated candidaemia, these rates were 28% and 50%, respectively. Demographic, epidemiological and clinical data based on candidaemia sources are shown in Table 3. Table 3Demographic, epidemiological, and clinical characteristics of patients based on candidaemia sourcesCVC (n = 32)Abdomen (n = 52)Primary (n = 55)Ürınary (n = 2)P1 Age (years) 51,7 ± 20,8 (19–94)55,9 ± 18,1 (19–95)58,9 ± 17,4 (23–90)61,5 ± 10,6 (54–69)0,256 Gender n (%) Female Male14 (43)18 (57)27 (52)25 (48)27 (49)28 (51)2(100)00,767 APACHE-II Score On The Day Of ICU Admission 22,9 ± 6,9 (12–40)23,9 ± 7,5 (9–46)23,6 ± 8,7 (4–42)31 ± 9,8 (24–38)0,746 EQUAL Candida Score 10,9 ± 4,2 (6–21)9,9 ± 4,1 (6–18)10 ± 4 (6–21)11 ± 2,8 (9–13)0,010* Total Hospital Stay 2 (days) 26,6 ± 25,2 (2–97)27,3 ± 19 (2–87)37,4 ± 53,2 (2-321)11,5 ± 4,9 (8–15)0,577 Total ICU stay 2 (days) 18,4 ± 22,4 (2–87)17,5 ± 17 (2–72)25,3 ± 49,4(2-321)6,5 ± 0,7 (6–7)0,818 At least one comorbidity n (%) 32 (98)48 (92)50 (90)2 (100)0,585 At least one immunosuppressive condition n (%) 3 14 (44)25 (48)26 (47)2 (100)0,924 Candida colonization in at least one anatomical site n (%) 9 (28)21 (40)26 (47)1 (50)0,214 Concomitant bacteremia n %) 4 3 (9)9 (17)15 (26)00,170 Breakthrough candidaemia n (%) 5 3 (9)18(35)9 (16)2 (100)0,012* Abdominal surgery (Last 3 months) (%) 5 (16)38 (73)6 (10)1 (50)< 0,001** The day the blood culture was taken (%) Sepsis Severe sepsis or septic shock7 (22)25 (78)12 (23)40 (77)16 (29)39 (71)1 (50)1 (50)0,747 ICU department n (%) Anesthesiology and Reanimation General Surgery Neurosurgery Thoracic-Cardiovascular Surgery Cardiology23 (72)3 (9)06 (19)015 (29)34 (65)03 (6)039 (71)7 (13)3 (6)4 (7)2 (3)2 (100)0000< 0,001** Species n (%) C. albicans C. tropicalis C. parapsilosis C. glabrata C. krusei C. kefyr Others9 (28)3 (9)14 (44)5 (16)01 (3)022 (42)2 (4)15 (29)11 (21)1 (2)01 (2) ****23 (41,8)4 (7,3)18 (32,8)6 (10,9)02 (3,6)2 (3,6) *****1 (50)001 (50)0000,592 TPN n (%) 16 (50)39 (75)20 (36)1(50)< 0,001*** CRRT n (%) 5 (16)9 (17)8 (14,5)00,815 ECMO n (%) 002 (3,6)00,338 Neutropenia n (%) 1 (3)04 (7,2)00,109 Steroid Use n (%) Prolonged Steroid Use 6 Pulse Steroid Use7 (22)2 (6)07 (13)3 (6)020 (36,3)7 (13)0000 0021*** 0,401 Chemotherapy/radiotherapy (CT/RT) (Last 6 months) n (%) 6 (19)5 (10)8 (14,5)00,693 Immunomodulator use n (%) 7 2 (6)2 (4)3 (5,4)00,891 Comorbidities n (%) Chronic Renal Failure Diabetes Mellitus Coronary Artery Disease Hypertension Solid Organ Malignancy Hematologic Malignancy Pancreatitis Multitrauma Autoimmune Disease Asthma-COPD Solid Organ Transplant Metabolic-Genetic Disease Liver Failure Cranial Disease Others 83 (9)8 (25)13 (41)8 (25)6 (19)3 (9)03 (9)2 (6)1 (3)1 (3)2 (6)03 (9)3 (9)4 (8)8 (15)8 (15,4)11 (21)18 (34,6)02 (4)3 (6)3 (6)2 (4)2 (4)2 (4)4 (8)1 (2)6 (12)7 (13)11 (20)18 (33)14 (25)3 (5)5 (9)2 (2)8 (15)6 (10)5 (9)2 (2)3 (5)2 (2)8 (15)3 (5)001 (50)01 (50)0001 (50)0000000,6800,5520,027*0,9120,001***0,0540,6150,3170,6660,4620,8340,8910,2400,0550,373Not receiving antifungal treatment, n (%)3 (9)3 (6)4 (7)00,842 Early antifungal therapy (< 48 h) 12 (38)35 (67)29 (53)2 (100)0,104 Antifungal n (%) Fluconazole Echinocandin Others 9 22 (69)6 (19)1 (3)34 (65)14 (27)1 (2)35 (64)12 (22)4 (7)2 (100)000,711 14-day mortality n (%) 9 (28,1)23 (44)27 (49)00,153 28-day mortality n (%) 16 (50)29 (55,7)32 (58)00,759Descriptive statistics are expressed as mean, standard deviation, frequency (n), and percentage (%)CVC: Central venous catheter, ICU: Intensive care unit, TPN: Total parenteral nutrition, CRRT: Continuous renal replacement therapy, ECMO: Extracorporeal membrane oxygenation, CT-RT: Chemotherapy-Radiotherapy, COPD: Chronic obstructive pulmonary disease1 Urinary candidaemia was not included in the statistical analysis due to the number of cases. P values for the analysis of abdominal, primary, and central venous catheter-associated candidaemia are presented2 The time elapsed until the day the blood culture was taken3 It was defined as the presence of at least one complicating immunosuppressive condition such as malignancy, autoimmune disease, solid organ transplant, CT-RT, steroid use, immunomodulator use, and neutropenia4 Blood culture growths within 48 h before the onset of candidaemia were screened and concomitant bacterial growths were recorded5 Patients who received antifungal therapy for ≥ 72 h at the onset of candidaemia were in this group6 Defined as use of > 20 mg/day prednisolone or equivalent steroid for more than three weeks7 All immunomodulators other than steroids are included in this group (Tacrolimus, cyclosporine (CsA), methotrexate (MTX), mycophenolate mofetil (MMF), etc.)8 Benign prostatic hyperplasia, urolithiasis, substance abuse9 Amphotericin-B was used in four patients, voriconazole in one patient, and itraconazole in one patient* A difference was found between abdominal and central venous catheter-related candidaemia** A difference was found between abdominal and central venous catheter-related candidaemia and between abdominal and primary candidaemia*** A difference was found between abdominal candidaemia and primary candidaemia**** C. haemulonii***** C. pelliculosa, C. .lusitaniae Demographic, epidemiological, and clinical characteristics of patients based on candidaemia sources Female Male 14 (43) 18 (57) 27 (52) 25 (48) 27 (49) 28 (51) 2(100) 0 Sepsis Severe sepsis or septic shock 7 (22) 25 (78) 12 (23) 40 (77) 16 (29) 39 (71) 1 (50) 1 (50) Anesthesiology and Reanimation General Surgery Neurosurgery Thoracic-Cardiovascular Surgery Cardiology 23 (72) 3 (9) 0 6 (19) 0 15 (29) 34 (65) 0 3 (6) 0 39 (71) 7 (13) 3 (6) 4 (7) 2 (3) 2 (100) 0 0 0 0 C. albicans C. tropicalis C. parapsilosis C. glabrata C. krusei C. kefyr Others 9 (28) 3 (9) 14 (44) 5 (16) 0 1 (3) 0 22 (42) 2 (4) 15 (29) 11 (21) 1 (2) 0 1 (2) **** 23 (41,8) 4 (7,3) 18 (32,8) 6 (10,9) 0 2 (3,6) 2 (3,6) ***** 1 (50) 0 0 1 (50) 0 0 0 Prolonged Steroid Use 6 Pulse Steroid Use 7 (22) 2 (6) 0 7 (13) 3 (6) 0 20 (36,3) 7 (13) 0 0 0 0 0021*** 0,401 Chemotherapy/radiotherapy (CT/RT) (Last 6 months) n (%) Chronic Renal Failure Diabetes Mellitus Coronary Artery Disease Hypertension Solid Organ Malignancy Hematologic Malignancy Pancreatitis Multitrauma Autoimmune Disease Asthma-COPD Solid Organ Transplant Metabolic-Genetic Disease Liver Failure Cranial Disease Others 8 3 (9) 8 (25) 13 (41) 8 (25) 6 (19) 3 (9) 0 3 (9) 2 (6) 1 (3) 1 (3) 2 (6) 0 3 (9) 3 (9) 4 (8) 8 (15) 8 (15,4) 11 (21) 18 (34,6) 0 2 (4) 3 (6) 3 (6) 2 (4) 2 (4) 2 (4) 4 (8) 1 (2) 6 (12) 7 (13) 11 (20) 18 (33) 14 (25) 3 (5) 5 (9) 2 (2) 8 (15) 6 (10) 5 (9) 2 (2) 3 (5) 2 (2) 8 (15) 3 (5) 0 0 1 (50) 0 1 (50) 0 0 0 1 (50) 0 0 0 0 0 0 0,680 0,552 0,027* 0,912 0,001*** 0,054 0,615 0,317 0,666 0,462 0,834 0,891 0,240 0,055 0,373 Fluconazole Echinocandin Others 9 22 (69) 6 (19) 1 (3) 34 (65) 14 (27) 1 (2) 35 (64) 12 (22) 4 (7) 2 (100) 0 0 Descriptive statistics are expressed as mean, standard deviation, frequency (n), and percentage (%) CVC: Central venous catheter, ICU: Intensive care unit, TPN: Total parenteral nutrition, CRRT: Continuous renal replacement therapy, ECMO: Extracorporeal membrane oxygenation, CT-RT: Chemotherapy-Radiotherapy, COPD: Chronic obstructive pulmonary disease 1 Urinary candidaemia was not included in the statistical analysis due to the number of cases. P values for the analysis of abdominal, primary, and central venous catheter-associated candidaemia are presented 2 The time elapsed until the day the blood culture was taken 3 It was defined as the presence of at least one complicating immunosuppressive condition such as malignancy, autoimmune disease, solid organ transplant, CT-RT, steroid use, immunomodulator use, and neutropenia 4 Blood culture growths within 48 h before the onset of candidaemia were screened and concomitant bacterial growths were recorded 5 Patients who received antifungal therapy for ≥ 72 h at the onset of candidaemia were in this group 6 Defined as use of > 20 mg/day prednisolone or equivalent steroid for more than three weeks 7 All immunomodulators other than steroids are included in this group (Tacrolimus, cyclosporine (CsA), methotrexate (MTX), mycophenolate mofetil (MMF), etc.) 8 Benign prostatic hyperplasia, urolithiasis, substance abuse 9 Amphotericin-B was used in four patients, voriconazole in one patient, and itraconazole in one patient * A difference was found between abdominal and central venous catheter-related candidaemia ** A difference was found between abdominal and central venous catheter-related candidaemia and between abdominal and primary candidaemia *** A difference was found between abdominal candidaemia and primary candidaemia **** C. haemulonii ***** C. pelliculosa, C. .lusitaniae Of the 131 patients receiving antifungal therapy, empiric therapy was initiated in 73(56%) cases, and targeted therapy was initiated in 58 (44%) cases. All antifungal agents were administered for therapeutic purposes. Fluconazole was used in 93 (71%) patients, echinocandins in 32 (24%) patients (20 caspofungin, 12 anidulafungin), amphotericin B in 4 (3%) patients (three liposomal, one conventional), itraconazole in 1 (1%) patient, and voriconazole in 1 (1%) patient. In the 131 patients who received antifungal therapy, the mean duration of treatment was 16.7 ± 11 days (1–74 days). Antifungal treatment was modified in 10 patients. In one patient, therapy was switched from an echinocandin to fluconazole based on susceptibility results. In seven patients, treatment was changed from fluconazole to an echinocandin, and in two patients from fluconazole to liposomal amphotericin B. All patients had received their initial regimen for at least 72 h prior to the treatment change. Of the 131 patients receiving antifungal therapy, 32 (23%) were diagnosed with breakthrough candidaemia. There was no significant difference in the type of antifungal agent between the breakthrough candidemia group and the non-breakthrough candidemia group (Table 4). Table 4Comparison of groups with and without breakthrough candidaemiaBreakthroughWithout Breakthrough p Total n = 1313299 Age (year) 60,9 ±17,6 (25–95)55,5 ± 18,3 (19–94)0,146 Gender n (%) Female Male20 (63)12 (37)44 (44)55 (56)0,076 APACHE-II Score On The Day Of ICU Admission 25,9 ± 8,8 (5–46)22,5 ±7,4 (4–42)0,037 EQUAL Candida Score 9,4 ± 4 (6–18)10,6 ± 4,2 (6–21)0,164Total Hospital Stay, (day) 131,2 ± 21,9 (2–87)28,6 ± 31 (2-132)0,068Total ICU stay, (day) 120,5 ±21 (2–74)18,5 ± 24,7 (2-132)0,248 Concomitant bacteremia n (%) 2 5 (20)20 (16)0,567 ICU department n (%) Anesthesiology and Reanimation General Surgery Neurosurgery Thoracic-Cardiovascular Surgery Cardiology10 (31)18 (56)04 (13)061 (62)25 (25)3 (3)9 (9)1 (1)0,008 Species n (%) C. albicans Non-C. albicans 10 (31)22 (69)38 (38)61 (72)0,467 Source of candidaemia n (%) Abdomen Primary Central venous catheter Urinary18 (56)9 (28)3 (10)2 (6)31 (31)42 (43)26 (26)00,003 The day the blood culture was taken n (%) Sepsis Severe sepsis or septic shock8 (25)24 (75)28 (28)71 (72)0,718 TPN n (%) 21(66)48 (49)0,091 CRRT n (%) 5 (19)15 (19)1 ECMO n (%) 02 (2)1 Neutropenia n (%) 2 (6)3 (3)0,595 Chemotherapy/radiotherapy (Last 6 months) n (%) 6 (19)9 (9)0,197 Abdominal surgery (Last 3 months) n (%) 19(59)30 (30)0,003 Immunomodulator use n (%) 3 07 (7)0,194 Steroid Use n (%) Prolonged Steroid Use 46 (19)3 (9)27 (27)8 (8)0,3340,729 At least one immunosuppressive condition n (%)5 17 (53)45 (46)0,45 Candida colonization in at least one anatomical site n (%) 12 (38)43 (43)0,554 At least one comorbidity n (%) 30 (94)87 (88)0,516 Chronic Renal Failure2 (7)10 (10)0,730 Diabetes Mellitus5 (15)20 (20)0,567 Coronary Artery Disease8 (25)29 (29)0,639 Hypertension6 (19)25 (25)0,452 Solid Organ Malignancy10 (31)15 (15)0,044 Hematologic Malignancy2 (6)6 (6)1 Pancreatitis 3 (9)00,014 Multitrauma2 (6)12 (12)0,516 Autoimmune Disease3 (9)7 (7)0,706 Asthma-COPD1 (3)7 (7)0,679 Solid Organ Transplant04 (4)0,572 Metabolic-Genetic Disease04 (4)0,572 Liver Failure04 (4)0,572 Cranial Disease1 (3)9 (9)0,45 Others64 (13)9 (9)0,519 Antifungal n (%) 7 Fluconazole Echinocandin Others 8 23 (72)6 (19)3 (9)70 (71)26 (26)3 (3)0,489 14-day mortality n (%) 15 (47)34 (34)0,203 28-day mortality n (%) 19 (59)48 (49)0,284Descriptive statistics are expressed as mean, standard deviation, frequency (n), and percentage (%)CVC: Central venous catheter, ICU: Intensive care unit, TPN: Total parenteral nutrition, CRRT: Continuous renal replacement therapy, ECMO: Extracorporeal membrane oxygenation, CT-RT: Chemotherapy-Radiotherapy, COPD: Chronic obstructive pulmonary disease1 The time elapsed until the day the blood culture was taken2 Blood culture growths within 48 h before the onset of candidaemia were screened and concomitant bacterial growths were recorded3 All immunomodulators other than steroids are included in this group (Tacrolimus, cyclosporine (CsA), methotrexate (MTX), mycophenolate mofetil (MMF), etc.)4 Defined as use of > 20 mg/day prednisolone or equivalent steroid for more than three weeks5 It was defined as the presence of at least one complicating immunosuppressive condition such as malignancy, autoimmune disease, solid organ transplant, CT-RT, steroid use, immunomodulator use, and neutropenia6 Benign prostatic hyperplasia, urolithiasis, substance abuse7 Antifungal therapy was evaluated between fluconazole and echinocandin8 Amphotericin-B was used in four patients, voriconazole in one patient, and itraconazole in one patient Comparison of groups with and without breakthrough candidaemia Female Male 20 (63) 12 (37) 44 (44) 55 (56) Anesthesiology and Reanimation General Surgery Neurosurgery Thoracic-Cardiovascular Surgery Cardiology 10 (31) 18 (56) 0 4 (13) 0 61 (62) 25 (25) 3 (3) 9 (9) 1 (1) C. albicans Non-C. albicans 10 (31) 22 (69) 38 (38) 61 (72) Abdomen Primary Central venous catheter Urinary 18 (56) 9 (28) 3 (10) 2 (6) 31 (31) 42 (43) 26 (26) 0 Sepsis Severe sepsis or septic shock 8 (25) 24 (75) 28 (28) 71 (72) 6 (19) 3 (9) 27 (27) 8 (8) 0,334 0,729 Fluconazole Echinocandin Others 8 23 (72) 6 (19) 3 (9) 70 (71) 26 (26) 3 (3) Descriptive statistics are expressed as mean, standard deviation, frequency (n), and percentage (%) CVC: Central venous catheter, ICU: Intensive care unit, TPN: Total parenteral nutrition, CRRT: Continuous renal replacement therapy, ECMO: Extracorporeal membrane oxygenation, CT-RT: Chemotherapy-Radiotherapy, COPD: Chronic obstructive pulmonary disease 1 The time elapsed until the day the blood culture was taken 2 Blood culture growths within 48 h before the onset of candidaemia were screened and concomitant bacterial growths were recorded 3 All immunomodulators other than steroids are included in this group (Tacrolimus, cyclosporine (CsA), methotrexate (MTX), mycophenolate mofetil (MMF), etc.) 4 Defined as use of > 20 mg/day prednisolone or equivalent steroid for more than three weeks 5 It was defined as the presence of at least one complicating immunosuppressive condition such as malignancy, autoimmune disease, solid organ transplant, CT-RT, steroid use, immunomodulator use, and neutropenia 6 Benign prostatic hyperplasia, urolithiasis, substance abuse 7 Antifungal therapy was evaluated between fluconazole and echinocandin 8 Amphotericin-B was used in four patients, voriconazole in one patient, and itraconazole in one patient The mean treatment duration for the 32 patients diagnosed with breakthrough candidaemia was 10.1 ± 8.3 (3–44) days. The demographic, epidemiological and clinical characteristics of patients with and without breakthrough candidaemia are shown in Table 4. In terms of the source of candidaemia, breakthrough candidaemia rates were significantly higher in abdominal candidaemia compared to central venous catheter-associated candidaemia (p = 0.012) (Table 4). No significant difference was found in fluconazole and echinocandin use rates between patients with and without breakthrough candidaemia (p = 0.489). While the rates of severe sepsis or septic shock were not statistically different between the breakthrough candidaemia and non-breakthrough candidaemia groups, the APACHE-II score on ICU admission was significantly higher in the breakthrough candidaemia group (p = 0.718 and p = 0.037, respectively) (Table 4). In 63 Candida strains for which antifungal susceptibility was studied, three Candida spp. were found to be resistant to fluconazole (two C. parapsilosis and one C. glabrata), while no echinocandin-resistant strains were found. Nine C. glabrata, three C. parapsilosis, and one C. albicans were found to be dose-dependently susceptible to fluconazole. Discussion The EUCANDICU study, which included nine European countries and 23 ICUs, found the incidence of candidaemia in ICUs to be 5.52 per 1,000 ICU admissions [3]. Studies from Türkiye reported candidaemia at rates of 1.76–41.4 per 1,000 ICU admissions and 23.1–28 per 10,000 hospitalisation days [14–18]. In our study, the incidence of candidaemia in ICUs was found to be 2.2 per 1,000 ICU admissions and 4.1 per 10,000 hospitalisation days. The overall incidence was highest in the Anaesthesiology and Reanimation ICU at 14.1 per 1,000 ICU admissions, while the General Surgery ICU ranked second at 5.1 per 1,000 ICU admissions. In other surgical ICUs, this rate was 0.7 and 0.3 (Cardiovascular and Thoracic Surgery ICU and Neurosurgery ICU, respectively). Klingspor et al. [2] reported that the most common preceding surgery was abdominal (51.5%), followed by thoracic (20%) and neurosurgery (8.2%). In our study, the high incidence of candidaemia in the Anaesthesiology and Reanimation ICU may be explained by the high rate of central venous catheter use, lower patient number and longer ICU stay. The significantly higher incidence of central venous catheter-related candidaemia in this unit also supports this hypothesis. The second-highest incidence of candidaemia in the General Surgery ICU may be explained by the significantly higher rate of abdominal surgery in patients [19]. The incidence of candidaemia varies across ICUs where different patient groups are monitored. The multicentre EUCANDICU study with European data found that the most common causes of candidaemia in ICUs were C. albicans, C. glabrata and C. parapsilosis, respectively [3]. In two retrospective studies from Türkiye, C. albicans ranked first, while C. parapsilosis and C. tropicalis were the second and third most frequently detected, respectively [17, 18, 20]. In our study, C. albicans was the most frequently detected causative agent (n = 55, 39%), while C. parapsilosis (n = 44, 31%) was the second most common. A significant increase was observed in the number of C. parapsilosis cases (n = 19, 37%) in the last quarter, with non-C. albicans species surpassing C. albicans. Our findings are similar to many studies, including recent Centers for Disease Control and Prevention (CDC) data [20–22]. The increase in non-C. albicans species is associated with increased rates of immunosuppressed patients, widespread use of antifungal prophylaxis and increased resistance rates [23]. In our study, breakthrough candidaemia was detected in 32(32/131, 24%) patients. Breakthrough candidaemia rates were significantly higher in patients with abdominal candidaemia. This may be explained by the early initiation of antifungal therapy and delay in source control due to clinical instability in patients undergoing abdominal surgery. On the other hand, Durmus et al. reported that initiating prophylactic antifungal therapy in liver transplant patients increased the risk of candidemia [24]. While studies on breakthrough candidaemia in patients with haematological malignancies and the paediatric population are more numerous in the literature, investigations of breakthrough candidaemia developing in the ICU are limited [25–27]. The APACHE-II score on ICU admission was significantly higher in the group with breakthrough candidaemia (p = 0.037) in our study. This may be explained by the clinical severity of the overall condition negatively affecting the immune system, leading to this result. In our study, no significant difference was found in the rates of severe sepsis or septic shock between the groups with and without breakthrough candidaemia (p = 0.718). A retrospective study by Cuervo et al. [28] including data from 44% of ICU patients showed no difference in the presence of hypotension between patients with(9% of 409 candidemia episodes) and without breakthrough candidaemia (p = 0.504). A five-year retrospective study by Breda et al. [29] including data from 52% of ICU patients found no difference in shock clinical findings between patients with(18% of 148 candidemia episodes) and without breakthrough candidaemia (p = 0.07). In our study, there was no difference in 14th and 28th-day mortality between the groups with and without breakthrough candidaemia (p = 0.203 and p = 0.284, respectively). An ICU study by Ohki et al. [7] showed that breakthrough candidaemia had no effect on 30th-day mortality (p = 0.681). Similarly, a study by Cuervo et al. [28] reported no significant difference in 5-day (early) and 30-day (late) mortality between patients with and without breakthrough candidaemia (p = 0.802 and p = 0.954, respectively). In their study conducted in the pediatric intensive care unit, Duyu et al. reported that breakthrough candidemia increased mortality by 10.5-fold [30]. In our study, fluconazole treatment rates were high in the group with breakthrough candidaemia (n = 23, 72%), while no difference was found in fluconazole and echinocandin use rates between the groups with and without breakthrough candidaemia (p = 0.489). While some studies have shown that prior azole use increases breakthrough candidaemia rates [25, 31], high breakthrough candidaemia rates have also been reported under echinocandins [32]. Our study is the first intensive care study to examine the source of candidaemia, which is one of its strengths. Limitations of our study include its retrospective nature, extended timeframe, single-center and predominantly descriptive and epidemiological in nature. Conclusions Candidaemia rates and type distribution in ICUs vary depending on the type of ICU and patient population. A significant proportion of candidemia cases in our intensive care units were non–catheter-related candidemia. It was determined that breakthrough candidaemia was seen at a higher rate in patients with high APACHE II scores at admission to intensive care and in patients with candidaemia developing from the abdomen. Acknowledgements I would like to thank all the authors for their contributions to this work. Abbrevations Intensive care unit Acute physiology and chronic health evaluation II Extended Prevalence of Infection in Intensive Care European Confederation of Medical Mycology Candidemia/intra-abdominal candidiasis in European ICU Quality of clinical candidaemia management Eosine Methylen Blue Medium Matrix-assisted laser desorption ionisation time of flight mass spectrometry Clinical & Laboratory Standards Institute Author contributions HA and ASŞ proposed and developed research question. ASŞ and HA wrote the first draft of the manuscript. All authors contributed the study design, interpretation of data and the revision of the manuscript. All authors have seen and contributed to the final version of manuscript. Funding There is no funding. Data availability The data that support the findings of this study are not openly available due to reasons of sensitivity and are available from the corresponding author upon reasonable request. Halis Akalınhalis@uludag.edu.tr. Declarations Ethics approval and consent to participate This study was conducted in accordance with the principles of the Declaration of Helsinki. The study was approved by the Clinical Research Ethics Committee of Bursa Uludag University Faculty of Medicine on May 3, 2023, with decision number 2023-10/4. Since the study was retrospective, informed consent was not obtained from the patients. Consent for publication Not applicable. Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Data Availability Statement The data that support the findings of this study are not openly available due to reasons of sensitivity and are available from the corresponding author upon reasonable request. Halis Akalınhalis@uludag.edu.tr.