Sage Journals: Discover world-class research

Abstract

Background

Candidemia is the most common invasive fungal disease in intensive care units (ICUs).

Objective

We aimed to investigate cases of candidemia infection developing in the ICU and factors associated with mortality due to this infection.

Materials and Methods

This is a retrospective study including patients admitted to a tertiary university hospital ICU between January 2012 and December 2020. Patients over 18 years of age who had candida growth in at least one blood culture taken from central or peripheral samples (>48 h after admission to the ICU) without concurrent growth were evaluated.

Results

The study group consisted of 136 patients with candida. Eighty-seven (63.97%) patients were male, with a median age of 69.5 (59–76.5) years. The 7-day mortality rate was 35.29%, while the 30-day mortality rate was 69.11%. As a result of multiple logistic regression analysis, after adjusting for age and malignancy, high APACHE II score and low platelet-lymphocyte ratio (PLR) - were found to be significant factors in predicting both 7-day and 30-day mortality.

Conclusion

In this study, PLR and APACHE II scores were shown to be independent predictors of mortality in patients with candidemia in the ICU.

Keywords

Candidemia intensive care unit mortality

Introduction

Candidemia is the most common invasive fungal disease in intensive care units (ICUs) and is considered to be the bloodstream pathogen with the highest mortality.^{1, 2} Risk factors for candidemia have been identified in several studies and can be listed as prolonged stay in ICU, use of immunosuppressive drugs, invasive interventions, and use of broad-spectrum antibiotics. In addition, as these risks demonstrate an increasing trend in medicine, the frequency of candidemia has also increased gradually during the last two decades and has become an important health problem in modern medicine.^3,4

Many of the predisposing factors for candidemia are very common in critically-ill patients admitted to the ICU.⁵ In addition, the low sensitivity of diagnostic tools and delayed results prevent rapid detection and treatment of this infection.⁶ Therefore, early and targeted treatment in candidemia is very important in terms of mortality and prognosis.⁷ Although blood cultures are accepted as the gold standard for the diagnosis of candidemia, the most important disadvantage of this method is that the results are often delayed.⁸ For this reason, investigation of the role of various laboratory parameters in the diagnosis of candidemia has recently gained interest.⁹ In addition, the relationship of haematological parameters such as neutrophil-lymphocyte ratio (NLR) and platelet-lymphocyte ratio (PLR) with prognosis and mortality in some diseases have yielded interesting results.^10,11 However, the relationship between these indicators and mortality from candidemia has not been adequately discussed, especially in ICU patients with candidemia.

With a better understanding of the relationship between candidemia and haematological parameters, important clues can be obtained both in the prevention of its development and in the early treatment of candidemia, which has become an important problem, especially in ICUs. Thus, it is evident that the identification of possible relationships in this context will also be effective in reducing mortality and morbidity in patients suffering from candidemia. In this study, we investigated cases of candidemia in the ICU and factors associated with mortality.

Methods

This study is a retrospective cohort examining patients admitted to the ICU at Istanbul Medipol University Hospital between January 2012 and December 2020. Necessary permissions were obtained from the Clinical Research Ethics Committee of local Institutional Review Board for the study.

Inclusion criteria

Patients over 18 years of age who had candida growth in at least one blood culture obtained from central or peripheral sites (>48 h after admission to the ICU) without concurrent growth were included. A total of 136 samples fulfilling these criteria were identified during the study period. If the same patient had more than one episode of candidemia during the study period, only the first episode of candidemia infection was included.

Exclusion criteria

Patients who were under the age of 18, had suspected candidemia before being admitted to the ICU, and subjects who died within 48 h of ICU admittance or were transferred to a different hospital were not included in the study. Accordingly, a final total of 136 patients were included in the analyzes.

Measurements

Yeast production detected in blood cultures was typed with VITEK 2 Compact System (BioMerieux, France) automated identification system. After positive blood culture was detected, blood samples were obtained for laboratory analyzes. A complete blood count (CBC) was performed in the hematology laboratory of the same hospital with the Horiba ABX Pentra DF 120 automatic analyzer.

Neutrophil-to-lymphocyte ratio (NLR) was calculated by dividing the absolute neutrophil count by the absolute lymphocyte count. Platelet-to-lymphocyte ratio (PLR) was calculated by dividing the platelet count by the lymphocyte count. Lymphocyte-to-monocyte ratio (LMR) was calculated by dividing lymphocyte count by monocyte count. The prognostic nutritional index (PNI) is associated with serum albumin level and the total lymphocyte count and was calculated by the formula: PNI = 10 × serum albumin concentration (g/dL) + 0.005 × total lymphocyte count (mm3). Systemic inflammation index (SII) was calculated using the following formula: platelet × (neutrophil/lymphocyte).

All clinical scores were recorded on the day of admission to the ICU. The Modified Glasgow Prognostic Score (mGPS) provides scores based on C-reactive protein (CRP) (CRP > 10 mg/L = 1 point) and albumin (albumin < 3.5 g/dL = 1 point) levels. Cases are classified as low risk (0 points), medium risk (1 point) and high risk (2 points) according to mGPS scores.^12,13 The Acute Physiology and Chronic Health Evaluation (APACHE) II is a parameter calculated based on the clinical and laboratory characteristics of the cases, and it is well established that higher scores indicate unfavorable outcomes.¹⁴

After recording all the above data, we identified 7-day and 30-day mortality status and evaluated associated factors.

Statistical analysis

All analyzes were subject to a significance threshold of p ≤ 0.05 and were performed on SPSS v25 (SPSS Inc., Chicago, IL, USA). For the normality check, histograms and Q-Q plots were used. Data are given as mean ± standard deviation or median (interquartile range; IQR) for continuous variables according to the normality of distribution, and as frequency (percentage) for categorical variables. According to distribution normality/non-normality, the independent samples t-test or the Mann-Whitney U test, respectively, were used for comparison of continuous variables. Categorical variable analyzes employed chi-square tests or Fisher's exact tests. The prediction performance of the variables was assessed by using Receiver Operating Characteristic (ROC) curve analysis. Multiple logistic regression analysis (forward conditional method) was performed to determine the best predictive factors of mortality.

Results

The study group consisted of 136 patients with candida. Eighty-seven (63.97%) patients were male and their median age (IQR) was 69.5 (59–76.5) years. While 7-day mortality rate was 35.29%, 30-day mortality rate was 69.11%. Factors associated with 7-day mortality were: malignancy (p = 0.001), high APACHE II score (p = 0.032), low platelet count (p = 0.002), low eosinophil count (p = 0.016), low PLR (p = 0.001) and low SII (p = 0.038) (Table 1)

Table 1.

Summary of patients characteristics with regard to 7-days mortality.

		7-days mortality
	Total (n = 136)	No (n = 88)	Yes (n = 48)	p
Age	69.5 (59–76.5)	68 (58–75.5)	72 (60.5–80)	0.186
Sex
Male	87 (63.97%)	60 (68.18%)	27 (56.25%)	0.231
Female	49 (36.03%)	28 (31.82%)	21 (43.75%)	0.231
Diagnosis
Pneumonia	17 (12.50%)	8 (9.09%)	9 (18.75%)	0.175
Dyspnea	29 (21.32%)	20 (22.73%)	9 (18.75%)	0.747
Clouding/Loss of consciousness	8 (5.88%)	6 (6.82%)	2 (4.17%)	0.712
General condition impairment	19 (13.97%)	11 (12.50%)	8 (16.67%)	0.681
Fever	10 (7.35%)	6 (6.82%)	4 (8.33%)	0.742
Post-operative	5 (3.68%)	3 (3.41%)	2 (4.17%)	1.000
Arrest/Post-CPR	13 (9.56%)	8 (9.09%)	5 (10.42%)	0.770
Sepsis	8 (5.88%)	6 (6.82%)	2 (4.17%)	0.712
Gastrointestinal problems	13 (9.56%)	9 (10.23%)	4 (8.33%)	1.000
Hemorrhage	7 (5.15%)	4 (4.55%)	3 (6.25%)	0.697
Other	21 (15.44%)	15 (17.05%)	6 (12.50%)	0.651
Reason of ICU admission
Diabetes mellitus	33 (24.26%)	22 (25.00%)	11 (22.92%)	0.951
Hypertension	57 (41.91%)	37 (42.05%)	20 (41.67%)	1.000
Coronary artery disease	23 (16.91%)	17 (19.32%)	6 (12.50%)	0.439
Heart diseases	25 (18.38%)	19 (21.59%)	6 (12.50%)	0.282
COPD	20 (14.71%)	15 (17.05%)	5 (10.42%)	0.430
Renal diseases	26 (19.12%)	15 (17.05%)	11 (22.92%)	0.546
Cerebrovascular disease	19 (13.97%)	11 (12.50%)	8 (16.67%)	0.681
Gastrointestinal problems	17 (12.50%)	12 (13.64%)	5 (10.42%)	0.786
Malignancy	48 (35.29%)	22 (25.00%)	26 (54.17%)	0.001
Other	25 (18.38%)	14 (15.91%)	11 (22.92%)	0.437
Mechanical ventilation
None	33 (24.26%)	26 (29.55%)	7 (14.58%)	0.150
NIV	13 (9.56%)	8 (9.09%)	5 (10.42%)
Intubation	90 (66.18%)	54 (61.36%)	36 (75.00%)
APACHE II score	23.74 ± 4.56	23.13 ± 4.55	24.88 ± 4.39	0.032
Candida
Albicans	70 (51.47%)	44 (50.00%)	26 (54.17%)	0.776
Non-albicans	66 (48.53%)	44 (50.00%)	22 (45.83%)	0.776
Hemoglobin	9.31 ± 1.76	9.35 ± 1.64	9.25 ± 1.99	0.762
Platelet (x10³)	155.5 (72–254)	196 (100–269)	79 (53–212)	0.002
WBC (x10³)	10.28 (7.20–15.97)	9.85 (6.93–13.91)	11.28 (7.20–20.61)	0.224
Neutrophil (x10³)	7.97 (5.32–12.75)	7.80 (5.58–11.71)	8.78 (5.05–18.95)	0.430
Lymphocyte (x10³)	0.88 (0.53–1.44)	0.79 (0.53–1.50)	1.02 (0.49–1.43)	0.821
Eosinophil (x10³)	0.05 (0.01–0.13)	0.06 (0.01–0.17)	0.02 (0.00–0.08)	0.016
Monocyte (x10³)	0.55 (0.31–0.89)	0.61 (0.36–0.90)	0.50 (0.25–0.89)	0.345
CRP	151.05 (85.90–261.03)	137.20 (77.82–238.56)	177.34 (102.39–281.41)	0.067
Albumin	2.65 ± 0.48	2.70 ± 0.50	2.56 ± 0.44	0.202
NLR	9.07 (5.20–17.31)	9.15 (5.23–15.75)	8.98 (4.59–22.39)	0.887
PLR	166.67 (89.39–303.03)	190.28 (128.04–334.35)	101.61 (45.16–235.90)	0.001
LMR	1.75 (1.00–2.83)	1.66 (1.03–2.42)	2.17 (0.88–5.04)	0.164
SII (x10³)	1319.89 (640.39–2612.44)	1640.84 (771.96–2648.97)	1115.71 (321.48–1834.29)	0.038
mGPS
0	3 (2.29%)	2 (2.38%)	1 (2.13%)	0.912
1	7 (5.34%)	5 (5.95%)	2 (4.26%)
2	121 (92.37%)	77 (91.67%)	44 (93.62%)
PNI	26.50 (7.00–34.55)	26.48 (5.85–34.35)	26.75 (7.15–35.80)	0.764
≥ 45	8 (5.93%)	3 (3.41%)	5 (10.64%)	0.126
< 45	127 (94.07%)	85 (96.59%)	42 (89.36%)	0.126

Data are given as mean ± standard deviation or median (1st quartile - 3rd quartile) for continuous variables according to normality of distribution and as frequency (percentage) for categorical variables.

Abbreviations: APACHE: Acute physiology and chronic health evaluation, COPD: Chronic obstructive pulmonary disease, CPR: Cardiopulmonary resuscitation, CRP: C-reactive protein, ICU: Intensive care unit, LMR: Lymphocyte/monocyte ratio, mGPS: Modified Glasgow prognostic score, NIV: Non-invasive ventilation, NLR: Neutrophil/lymphocyte ratio, PLR: Platelet/lymphocyte ratio, PNI: Prognostic nutritional index, SII: Systemic inflammatory index, WBC: White blood cell.

When the performances of various variables in predicting 7-day mortality were evaluated, it was found that high APACHE II score (p = 0.008), low PLR (p = 0.001) and low SII (p = 0.038) were significant. Among these variables, PLR was found to predict 7-day mortality with the highest accuracy (73.68%) (Table 2,Figure 1).

Figure 1.

ROC curve of the variables to predict 7-days mortality.

Table 2.

Performance of the variables to predict 7-days mortality.

	Cut-off	Sensitivity	Specificity	Accuracy	PPV	NPV	AUC (95.0% CI)	p
APACHE II score	≥ 22	87.50%	38.64%	55.88%	43.75%	85.00%	0.638 (0.542–0.733)	0.008
NLR	< 2.05	14.89%	96.47%	67.42%	70.00%	67.21%	0.492 (0.384–0.601)	0.887
PLR	< 85	46.81%	88.37%	73.68%	68.75%	75.25%	0.675 (0.573–0.778)	0.001
LMR	≥ 2.15	51.06%	72.09%	64.66%	50.00%	72.94%	0.573 (0.461–0.685)	0.165
SII (x10³)	< 1600	74.47%	50.59%	59.09%	45.45%	78.18%	0.609 (0.504–0.715)	0.038
mGPS	2	93.62%	8.33%	38.93%	36.36%	70.00%	0.510 (0.407–0.613)	0.855
PNI	< 45	89.36%	3.41%	33.33%	33.07%	37.50%	0.484 (0.380–0.589)	0.764

Abbreviations: APACHE: Acute physiology and chronic health evaluation, AUC: Area under ROC curve, CI: Confidence intervals, LMR: Lymphocyte/monocyte ratio, mGPS: Modified Glasgow prognostic score, NLR: Neutrophil/lymphocyte ratio, NPV: Negative predictive value, PLR: Platelet/lymphocyte ratio, PNI: Prognostic nutritional index, PPV: Positive predictive value, SII: Systemic inflammatory index.

We performed multiple logistic regression to determine significant predictive factors of 7-day mortality. We found the presence of malignancy, high APACHE II score and low PLR were significant factors. Patients with a high APACHE II score (≥22) had a 7.127-fold higher risk of death than those with a lower score (OR: 7.127, 95% CI: 2.188–23.217; p = 0.001). Patients with low PLR (<85) had a 9.328-fold higher risk of death than those with higher PLR (OR: 9.328, 95% CI: 3.303–26.344, p < 0.001). Other variables included in the model, age (p = 0.681), sex (p = 0.727), mechanical ventilation (p = 0.978), type of candida (p = 0.602), NLR (p = 0.222), LMR (p = 0.565), SII (p = 0.583), mGPS (p = 0.741) and PNI score (p = 0.718) were found to be non-significant (Table 3).

Table 3.

Significant predictive factors of the 7-days mortality, multiple logistic regression analysis.

	β coefficient	Standard Error	p	Exp(β)	95.0% CI for Exp(β)
Malignancy	1.804	0.474	<0.001	6.072	2.397	15.384
APACHE II score (≥22)	1.964	0.603	0.001	7.127	2.188	23.217
PLR (< 85)	2.233	0.530	<0.001	9.328	3.303	26.344
Constant	−3.427	0.657	<0.001	0.032

Dependent Variable: 7-days mortality; Nagelkerke R² = 0.413; Correct prediction = 78.20%

Abbreviations: APACHE: Acute physiology and chronic health evaluation, CI: Confidence intervals, PLR: Platelet/lymphocyte ratio.

When the factors associated with 30-day mortality were analyzed, significant relationships were found between 30-day mortality and malignancy (p = 0.014), mechanical ventilation (p = 0.038), high APACHE II score (p < 0.001), high CRP (p = 0.049), and low albumin (p = 0.025) There was no significant difference between Candida species (albicans vs non-albicans) in terms of 7-day and 30-day mortality rates (p = 0.776, p = 0.247, respectively) (Table 4).

Table 4.

Summary of patients characteristics with regard to 30-days mortality.

	30-days mortality
	No (n = 42)	Yes (n = 94)	p
Age	71 (61–77)	67 (57–76)	0.281
Sex
Male	28 (66.67%)	59 (62.77%)	0.807
Female	14 (33.33%)	35 (37.23%)	0.807
Diagnosis
Pneumonia	7 (16.67%)	10 (10.64%)	0.483
Dyspnea	8 (19.05%)	21 (22.34%)	0.836
Clouding/Loss of consciousness	2 (4.76%)	6 (6.38%)	1.000
General condition impairment	6 (14.29%)	13 (13.83%)	1.000
Fever	4 (9.52%)	6 (6.38%)	0.498
Post-operative	3 (7.14%)	2 (2.13%)	0.171
Arrest/Post-CPR	2 (4.76%)	11 (11.70%)	0.344
Sepsis	2 (4.76%)	6 (6.38%)	1.000
Gastrointestinal problems	4 (9.52%)	9 (9.57%)	1.000
Hemorrhage	1 (2.38%)	6 (6.38%)	0.436
Other	7 (16.67%)	14 (14.89%)	0.994
Reason of ICU admission
Diabetes mellitus	13 (30.95%)	20 (21.28%)	0.317
Hypertension	21 (50.00%)	36 (38.30%)	0.276
Coronary artery disease	7 (16.67%)	16 (17.02%)	1.000
Heart diseases	9 (21.43%)	16 (17.02%)	0.709
COPD	9 (21.43%)	11 (11.70%)	0.223
Renal diseases	7 (16.67%)	19 (20.21%)	0.803
Cerebrovascular disease	4 (9.52%)	15 (15.96%)	0.464
Gastrointestinal problems	6 (14.29%)	11 (11.70%)	0.888
Malignancy	8 (19.05%)	40 (42.55%)	0.014
Other	7 (16.67%)	18 (19.15%)	0.916
Mechanical ventilation
None	16 (38.10%)	17 (18.09%)	0.038
NIMV	4 (9.52%)	9 (9.57%)
Intubation	22 (52.38%)	68 (72.34%)
APACHE II score	20.76 ± 3.05	25.07 ± 4.50	<0.001
Candida
Albicans	18 (42.86%)	52 (55.32%)	0.247
Non-albicans	24 (57.14%)	42 (44.68%)	0.247
Hemoglobin	9.59 ± 1.62	9.19 ± 1.82	0.228
Platelet (x10³)	189 (112–250)	127 (64–254)	0.130
WBC (x10³)	9.63 (7.39–13.08)	10.76 (7.20–16.14)	0.525
Neutrophil (x10³)	8.17 (5.68–10.93)	7.92 (5.25–13.22)	0.699
Lymphocyte (x10³)	0.77 (0.54–1.50)	0.95 (0.50–1.44)	0.857
Eosinophil (x10³)	0.05 (0.01–0.10)	0.04 (0.01–0.15)	0.914
Monocyte (x10³)	0.55 (0.38–0.79)	0.56 (0.29–0.91)	0.862
CRP	110.12 (73.31–178.52)	162.35 (102.39–268.64)	0.049
Albumin	2.83 ± 0.48	2.58 ± 0.47	0.025
NLR	9.33 (5.23–15.98)	8.98 (5.17–18.39)	0.729
PLR	189.66 (138.16–334.35)	146.88 (73.24–272.76)	0.052
LMR	1.77 (1.02–2.69)	1.72 (0.99–3.16)	0.819
SII (x10³)	1939.74 (804.64–2907.53)	1207.79 (554.11–2558.25)	0.098
mGPS
0	2 (4.88%)	1 (1.11%)	0.120
1	4 (9.76%)	3 (3.33%)
2	35 (85.37%)	86 (95.56%)
PNI	27.78 (5.85–34.80)	26.25 (7.15–34.50)	0.919
≥ 45	2 (4.76%)	6 (6.45%)	1.000
< 45	40 (95.24%)	87 (93.55%)	1.000

The performance of the variables in predicting 30-day mortality was evaluated. Among the variables, only the APACHE II score was found to significantly predict 30-day mortality with an accuracy of 77.94%. No significant value was found in terms of other variables (Table 5,Figure 2).

Figure 2.

ROC curve of the variables to predict 30-days mortality.

Table 5.

Performance of the variables to predict 30-days mortality.

	Cut-off	Sensitivity	Specificity	Accuracy	PPV	NPV	AUC (95.0% CI)	p
APACHE II score	≥ 22	85.11%	61.90%	77.94%	83.33%	65.00%	0.790 (0.711–0.870)	<0.001
NLR	< 11.5	64.84%	48.78%	59.85%	73.75%	38.46%	0.519 (0.412–0.625)	0.729
PLR	< 105	39.13%	87.80%	54.14%	87.80%	39.13%	0.606 (0.505–0.706)	0.052
LMR	≥ 2.15	39.13%	70.73%	48.87%	75.00%	34.12%	0.512 (0.409–0.616)	0.819
SII (x10³)	< 1950	69.23%	48.78%	62.88%	75.00%	41.67%	0.590 (0.484–0.696)	0.098
mGPS	2	95.56%	14.63%	70.23%	71.07%	60.00%	0.551 (0.441–0.661)	0.348
PNI	< 45	93.55%	4.76%	65.93%	68.50%	25.00%	0.506 (0.400–0.611)	0.919

We performed multiple logistic regression to determine significant predictive factors of 30-day mortality. We found age, malignancy presence, high APACHE II score and low PLR as significant factors. Patients with a high APACHE II score (≥22) had 31.699-fold higher risk of death than those with a lower score (OR: 31.699, 95% CI: 8.394–119.709; p < 0.001). Patients with low PLR (<105) had 10.467-fold higher risk of death than those with higher PLR (OR: 10.467, 95% CI: 2.540–43.123; p = 0.001). Other variables included in the model, sex (p = 0.767), mechanical ventilation (p = 0.631), type of candida (p = 0.156), NLR (p = 0.569), LMR (p = 0.697), SII (p = 0.393), mGPS (p = 0.103) and PNI score (p = 0.237) were found to be non-significant (Table 6).

Table 6.

Significant predictive factors of the 30-days mortality, multiple logistic regression analysis.

	β coefficient	Standard Error	p	Exp(β)	95.0% CI for Exp(β)
Age	−0.044	0.018	0.013	0.957	0.924	0.991
Malignancy	2.076	0.684	0.002	7.973	2.087	30.462
APACHE II score (≥22)	3.456	0.678	<0.001	31.699	8.394	119.709
PLR (< 105)	2.348	0.722	0.001	10.467	2.540	43.123
Constant	0.412	1.208	0.733	1.510

Dependent Variable: 30-days mortality; Nagelkerke R² = 0.527; Correct prediction = 79.70%

Abbreviations: APACHE: Acute physiology and chronic health evaluation, CI: Confidence intervals, PLR: Platelet/lymphocyte ratio.

Discussion

According to the results of our study, 7-day mortality rate was 35.2% and the 30-day mortality rate was 69.1% in ICU patients that developed candidemia. We also found a high APACHE II score and low PLR to be significant independent risk factors for both 7-day and 30-day mortality in ICU patients with candida infection –with adjustment for age and malignancy by inclusion in the models.

Prior studies show that 33–55% of all candidemia infections occur in the ICU and 5–71% of these cases are reported to result in mortality.^15,16 In a multicentre study evaluating candida infections in 23 ICUs in Europe, crude 30-day mortality was reported to be 42%.¹⁷ Other studies estimating attributable mortality rates show values between 40–50%.^18–20 In our study, the 30-day mortality rate was found to be 69.1%, which was considerably higher compared to the literature. However, our results may have been affected because our study was conducted in a tertiary ICU, and there was no evaluation of whether the deaths were due to candidiasis or worsening of the underlying disease; thus, attributable mortality rates could not be calculated.

In different studies, many factors associated with mortality and prognosis in candidemia have been identified. These factors include the use of mechanical ventilation, hypoproteinaemia, high APACHE II score, delay in catheter removal, and inadequate antifungal therapy.^21–24 In addition, a recent study reported that the cumulative number of related risk factors is the most useful variable in predicting hospital death due to candidemia among intensive care patients.²⁵ In our study, factors associated with both 7-day and 30-day mortality were evaluated. Accordingly, 7-day mortality was found to be associated with (univariate analysis): malignancy, high APACHE II score, platelet level, eosinophil level, low PLR level and SII level. The factors that were significantly associated with 30-day mortality (again, univariate results) were malignancy, mechanical ventilation, high APACHE II score, high CRP, and low albumin. However, with further analysis after adjusting for age and malignancy via regression models, high APACHE II scores and low PLR levels were found to be independent risk factors for both 7-day mortality and 30-day mortality.

The APACHE II score, which is widely used in the ICU and is one of the important tools in showing prognosis, has been shown to accurately measure the severity of the disease and is also strongly associated with mortality in critically-ill patients.²⁶ In addition, it has been frequently emphasized that a high APACHE II score may be a predictor of mortality in patients with candida infection,^16,27,28 similar to our findings. In a recent study, it was reported that an APACHE II score above 20 was an independent risk factor for mortality in patients with candidiasis.²⁹ In our study, patients with a high APACHE II score (≥22) had a 7.1-fold higher risk of death in terms of 7-day mortality and a 31.6-fold higher risk of death in terms of 30-day mortality, in line with the literature. In addition, it was found that the APACHE II score significantly predicted 30-day mortality with an accuracy of 77.94%.

Studies have reported that PLR, which is a haematological index reflecting inflammation and thrombosis, may provide prognostic efficacy in various cancers and heart-lung diseases.^30–33 In addition, in a study, it was reported that PLR value was associated with increased mortality in sepsis patients.³⁴ In a recent study, Zhai et al. reported that PLR is an independent predictor of mortality in patients hospitalized in the ICU.³⁵ Contrary to these reports, in a study evaluating bloodstream infections, including candida, it was suggested that PLR level was not associated with clinical outcomes.³⁶ In our study, however, it was found that low PLR level was independently associated with increased 7-day mortality (9.3 fold) and 30-day mortality (10.4 fold). In addition, PLR levels was found to predict 7-day mortality with the highest accuracy (73.68%). We thought that this result was mainly due to the decrease in platelet level. Consistent with our findings, recent studies have reported that newly developing thrombocytopenia is associated with mortality in candidemia cases.^37,38 In the literature, it can be said that the relationship between PLR and mortality in candidemia cases has been examined in very few studies, and the fact that our results indicate greater accuracy compared to APACHE II score is crucial and warrants further research on this topic. New studies are needed to better understand the relationship between haematological parameters and candida infections, and how these factors are associated with mortality and prognosis.

This study had some limitations. First, the study design had inherent disadvantages due to its retrospective nature. Second, we did not distinguish whether death was due to candidemia or the underlying disease, as it is difficult to determine the exact cause of death in critically ill patients. In relation, previous studies have reported that culture sampling experience may affect the outcomes.³⁹ Although all cultures were obtained in a standardized fashion, possible differences in operator experience may have caused limited but unavoidable variations in culture positivity for both candida and other agents. The fact that the study was single-centred can be noted as another limitation. In addition, although there have been many studies examining candidemia and the factors affecting mortality, one of the strengths of our study is that it is one of the few studies examining the relationships between CBC-derived haematological parameters and mortality.

Conclusions

In this study, it was observed that low PLR level and high APACHE II score increased both 7-day and 30-day mortality in patients who developed candidemia in the ICU. These two parameters were independent predictors of mortality, in addition to malignancy (7-day and 30-day) and age (30-day). In clinical practice, it may be useful to consider APACHE II and PLR levels in cases with candidemia in the ICU, and it appears that PLR may in fact be more accurate for prognostic assessment in the shorter term. Since patients with candidemia have greater mortality in clinical practice, empirical antifungal therapy may be considered for patients with high APACHE II score, low PLR or thrombocytopenia at ICU admission in order to reduce mortality. The relationship between CBC-derived haematological parameters and candidemia can be examined in more detail with future prospective, multicentre studies that include more cases.

Footnotes

Authors' contributions

All authors analyzed, interpreted the patient data, and were major contributors in writing the manuscript. All authors read and approved the final manuscript.

Availability of data and material

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request

Consent for publication

Informed consent was not obtained from the patients because the study was a retrospective study.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Ethical approval

The study has been conducted by the principles of the Helsinki Declaration and approved by the local Institutional Review Board.

ORCID iD

Burcu Tunay

Author biographies

Burcu Tunay is an experienced anesthesia and reanimation specialist working in the general intensive care unit of Istanbul Medipol University Hospital. She has articles published in the field at home and abroad. She continues her scientific activities in her field.

Selda Aydin is a Clinical Microbiology and Infectious Diseases Specialist working at Istanbul Medipol University. She has many articles written in his field in Turkey and abroad. She still continues active clinical studies.

References

Pfaller

Diekema

Epidemiology of invasive mycoses in North America

. Crit Rev Microbiol 2010 ; 36 : 1 – 53 .

Guery

Arendrup

Auzinger

, et al. Management of invasive candidiasis and candidemia in adult non-neutropenic intensive care unit patients: part I. Epidemiology and diagnosis. Intensive Care Med 2009; 35: 55–62.

Leroy

Bailly

Gangneux

J-P

, et al. Systemic antifungal therapy for proven or suspected invasive candidiasis: the AmarCAND 2 study. Ann Intensive Care 2016; 6: 1–11.

Wisplinghoff

Ebbers

Geurtz

, et al. Nosocomial bloodstream infections due to candida spp. in the USA: species distribution, clinical features and antifungal susceptibilities. Int J Antimicrob Agents 2014; 43: 78–81.

Pfaller

Diekema

. Epidemiology of invasive candidiasis: a persistent public health problem. Clin Microbiol Rev 2007; 20: 133–163.

Pfaller

Castanheira

. Nosocomial candidiasis: antifungal stewardship and the importance of rapid diagnosis. Med Mycol 2016; 54: 1–22.

Bassetti

Righi

Ansaldi

, et al. A multicenter study of septic shock due to candidemia: outcomes and predictors of mortality. Intensive Care Med 2014; 40: 839–845.

Clancy

Nguyen

. Finding the “missing 50%” of invasive candidiasis: how nonculture diagnostics will improve understanding of disease spectrum and transform patient care. Clin Infect Dis 2013; 56: 1284–1292.

Ríos-Toro

J-J

Márquez-Coello

García-Álvarez

J-M

, et al. Soluble membrane receptors, interleukin 6, procalcitonin and C reactive protein as prognostic markers in patients with severe sepsis and septic shock. PLoS One 2017; 12: e0175254.

10.

Wang

Jiang

, et al. Prognostic value of neutrophil-to-lymphocyte ratio and platelet-to-lymphocyte ratio in acute pulmonary embolism: a systematic review and meta-analysis. Int Angiol 2017; 37: 4–11.

11.

Zhou

, et al. The prognostic value of the platelet-to-lymphocyte ratio in acute coronary syndrome: a systematic review and meta-analysis. Kardiologia Polska (Polish Heart Journal) 2017; 75: 666–673.

12.

Forrest

McMillan

McArdle

, et al. Comparison of an inflammation-based prognostic score (GPS) with performance status (ECOG) in patients receiving platinum-based chemotherapy for inoperable non-small-cell lung cancer. Br J Cancer 2004; 90: 1704–1706.

13.

Zhou

Hong

, , et al. A systemic inflammation-based prognostic scores (mGPS) predicts overall survival of patients with small-cell lung cancer. Tumour Biol 2015; 36: 337–343.

14.

Knaus

Draper

Wagner

, et al. APACHE II: a severity of disease classification system. Crit Care Med 1985; 13: 818–829.

15.

Patolia

Kennedy

Zahir

, et al. Risk factors for candida blood stream infection in medical ICU and role of colonization-A retrospective study. British Journal of Medical Practitioners 2013; 6(1): a618.

16.

Al-Dorzi

Sakkijha

Khan

, et al. Invasive candidiasis in critically ill patients: a prospective cohort study in two tertiary care centers. J Intensive Care Med 2020; 35: 542–553.

17.

Bassetti

Giacobbe

Vena

, et al. Incidence and outcome of invasive candidiasis in intensive care units (ICUs) in Europe: results of the EUCANDICU project. Crit Care 2019; 23: 1–7.

18.

Shastri

Shankarnarayan

Oberoi

, et al. Candida auris candidaemia in an intensive care unit–prospective observational study to evaluate epidemiology, risk factors, and outcome. J Crit Care 2020; 57: 42–48.

19.

Ostrosky-Zeichner

Pappas

. Invasive candidiasis in the intensive care unit. Crit Care Med 2006; 34: 857–863.

20.

Gudlaugsson

Gillespie

Lee

, et al. Attributable mortality of nosocomial candidemia, revisited. Clin Infect Dis 2003; 37: 1172–1177.

21.

Raja

. Epidemiology, risk factors, treatment and outcome of candida bloodstream infections because of candida albicans and candida non–albicans in two district general hospitals in the United Kingdom. Int J Clin Pract 2021; 75: e13655.

22.

Zeng

Z-r

Tian

Ding

Y-h

, et al. Surveillance study of the prevalence, species distribution, antifungal susceptibility, risk factors and mortality of invasive candidiasis in a tertiary teaching hospital in southwest China. BMC Infect Dis 2019; 19: 1–12.

23.

Zhang

X-B

S-J

J-X

, et al. Retrospective analysis of epidemiology and prognostic factors for candidemia at a hospital in China, 2000–2009. Jpn J Infect Dis 2012; 65: 510–515.

24.

Garnacho-Montero

Díaz-Martín

García-Cabrera

, et al. Impact on hospital mortality of catheter removal and adequate antifungal therapy in candida spp. Bloodstream infections. J Antimicrob Chemother 2013; 68: 206–213.

25.

Kawano

Togawa

Nakamura

, et al. Prognostic factors for candidaemia in intensive care unit patients: a retrospective analysis. Singapore Med J 2017; 58: 196.

26.

Moon

Park

Jang

, et al. Use of APACHE II and SAPS II to predict mortality for hemorrhagic and ischemic stroke patients. J Clin Neurosci 2015; 22: 111–115.

27.

Hirano

Sakamoto

Kudo

, et al. Retrospective analysis of mortality and candida isolates of 75 patients with candidemia: a single hospital experience. Infect Drug Resist 2015; 8: 199.

28.

Ahmed

Azim

Baronia

, et al. Risk prediction for invasive candidiasis. Indian Journal of Critical Care Medicine: Peer-Reviewed, Official Publication of Indian Society of Critical Care Medicine 2014; 18: 682.

29.

Chen

Yang

, et al. Clinical characteristics and risk factors for death in patients with candida bloodstream infection in intensive care unit. Zhong nan da xue xue bao Yi xue ban = Journal of Central South University Medical Sciences 2021; 46: 719–724.

30.

Guo

Liu

, et al. Prognostic value of neutrophil–to–lymphocyte ratio and platelet–to–lymphocyte ratio for breast cancer patients: an updated meta–analysis of 17079 individuals. Cancer Med 2019; 8: 4135–4148.

31.

Zhou

Liu

, et al. Platelet-to-lymphocyte ratio in advanced cancer: review and meta-analysis. Clin Chim Acta 2018; 483: 48–56.

32.

Dong

Huang

Liu

. Platelet–to–lymphocyte ratio and prognosis in STEMI: a meta–analysis. Eur J Clin Invest 2021; 51: e13386.

33.

Galliazzo

Nigro

Bertù

, et al. Prognostic role of neutrophils to lymphocytes ratio in patients with acute pulmonary embolism: a systematic review and meta-analysis of the literature. Intern Emerg Med 2018; 13: 603–608.

34.

Shen

Huang

Zhang

. Platelet-to-lymphocyte ratio as a prognostic predictor of mortality for sepsis: interaction effect with disease severity—a retrospective study. BMJ open 2019; 9: e022896.

35.

Zhai

Wang

Liu

, et al. Platelet-lymphocyte ratio as a new predictor of in-hospital mortality in cardiac intensive care unit patients. Sci Rep 2021; 11: 23578–.

36.

Tang

Zhang

, et al. Hematological parameters in patients with bloodstream infection: A retrospective observational study. J Infect Dev Ctries 2020; 14(11): 1264–1273.

37.

Kutlu

Sayın-Kutlu

Alp-Çavuş

, et al. Mortality-associated factors of candidemia: a multi-center prospective cohort in Turkey. Eur J Clin Microbiol Infect Dis 2022; 41: 597–607.

38.

Yakut

Kepenekli

Ergenc

, et al. Antifungal susceptibility, species distribution and risk factors associated with mortality of invasive candidiasis in children in Turkey: a six-year retrospective, single-centre study. Journal of Medical Mycology 2021; 31: 101082.

39.

Yalçinkaya

Öz

Erdoğan

, et al. Turkish pediatric residents’ knowledge, perceptions, and practices of blood culture sampling. Archives de Pédiatrie 2021; 28: 191–196.

Investigation of inflammation-related parameters in patients with candidemia hospitalized in the intensive care unit: A retrospective cohort study

Abstract

Background

Objective

Materials and Methods

Results

Conclusion

Keywords

Introduction

Methods

Inclusion criteria

Exclusion criteria

Measurements

Statistical analysis

Results

Discussion

Conclusions

Footnotes

Authors' contributions

Availability of data and material

Consent for publication

Declaration of conflicting interests

Funding

Ethical approval

ORCID iD

Author biographies

References