Empiric Echinocandin Therapy in Sepsis: Echino-“Can It Work” or Echino-“Can You Not”? Luke Smedley, PharmD PGY-2 Critical Care Pharmacy Resident Department of Pharmacotherapy and Pharmacy Services, University Hospital Division of Pharmacotherapy, University of Texas at Austin College of Pharmacy Pharmacotherapy Education and Research Center, UT Health San Antonio San Antonio, Texas August 8 and 17, 2018 Learning Objectives 1. Differentiate various diagnostic tools for invasive candidiasis and describe situations where false positives or false negatives may arise 2. Appraise the data surrounding azole versus echinocandin therapy for invasive candidiasis 3. Defend the choice to initiate or defer initiation of echinocandin therapy in a patient with sepsis without proven invasive candidiasis
20
Embed
Empiric Echinocandin Therapy in Sepsis: Echino-“Can It ... › pharmacotherapy-rounds › files › 2018 › 08 › … · and Pakistan 1. C. parapsilosis. more common in pediatric
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Empiric Echinocandin Therapy in Sepsis: Echino-“Can It Work” or Echino-“Can You Not”?
Luke Smedley, PharmD PGY-2 Critical Care Pharmacy Resident
Department of Pharmacotherapy and Pharmacy Services, University Hospital Division of Pharmacotherapy, University of Texas at Austin College of Pharmacy
Pharmacotherapy Education and Research Center, UT Health San Antonio San Antonio, Texas
August 8 and 17, 2018
Learning Objectives
1. Differentiate various diagnostic tools for invasive candidiasis and describe situations wherefalse positives or false negatives may arise
2. Appraise the data surrounding azole versus echinocandin therapy for invasive candidiasis3. Defend the choice to initiate or defer initiation of echinocandin therapy in a patient with
sepsis without proven invasive candidiasis
2
Assessment Questions:
1. Which diagnostic tool provides the fastest results confirming the presence of a Candida spp. in the bloodstream?
a. BioFire FilmArray® Blood Culture ID Panel b. (1-3)-β-D-glucan c. T2Candida® d. Traditional blood culture
2. True or false: in clinical trials, fluconazole therapy has been consistently shown to
have better outcomes than echinocandin therapy.
3. True or false: due to their once daily dosing and limited side-effect profile, echinocandins can safely be given to all patients with septic shock without consequences.
***To obtain CE credit for attending this program please sign in. Attendees will be emailed a link to an electronic CE Evaluation Form. CE credit will be awarded upon completion of the electronic form. If you do not receive an email within 72 hours, please contact the CE Administrator at [email protected] *** Faculty (Speaker) Disclosure: Luke Smedley has indicated he has no relevant financial relationships to disclose relative to the content of this presentation.
Abbreviations: AUC = area under the time-concentration curve, CNS = central nervous system, GI = gastrointestinal, IV = intravenous, LFT = liver function test, MIC = minimum inhibitory concentration
3. Multiple studies favor use of an echinocandin for management of IC
Table 4. Studies of fluconazole versus echinocandins for invasive candidiasis
Study Intervention Results
Reboli AC, Rotstein C, Pappas PG, et al. Anidulafungin versus fluconazole for invasive candidiasis. N Engl J Med. 2007;356:2472-82.
25
Prospective, randomized controlled trial of anidulafungin (200 mg IV on day 1, then 100 mg IV daily) vs fluconazole (800 mg IV on day, then 400 mg IV daily)
Global response at end of IV therapy: 75.6% anidulafungin vs fluconazole 60.2% (p = 0.01) o Remained significant when
adjusted for baseline characteristics of immunosuppressive therapy, diabetes, prior azole therapy, presence of C. glabrata, and catheter removal
Remained significant at end of all treatment and 2-week follow-up
Lost statistical superiority at 6-week follow-up
Andes DR, Safdar N, Baddley JW, et al. Impact of treatment strategy on outcomes in patients with candidemia and other forms of invasive candidiasis: a patient-level quantitative review of randomized trials. Clin Infect Dis. 2012;54:1110-22.
26
Patient-level meta-analysis of treatments for IC
Echinocandin use associated with improved 30-day all-cause mortality (OR 0.65, 95% CI 0.45-0.94, p = 0.02)
Also associated with higher success rate at end of therapy (OR 2.33, 95% 1.27-4.35, p = 0.01)
Eschenauer GA, Carver PL, Lin SW, et al. Fluconazole versus an echinocandin for Candida glabrata fungaemia: a retrospective cohort study. J Antimicrob Chemother. 2013;68:922-6.
27
Retrospective cohort study of fluconazole versus an echinocandin for candidemia due to C. glabrata
In multivariate analysis, echinocandin use was associated with increased odds of complete response at day 14 (OR 2.305, 95% CI 1.124-4.727, p = 0.023)
Echinocandin use was not associated with 28-day survival (OR 1.843, 95% CI 0.835-4.069, p = 0.13)
4. Multiple mechanisms of resistance have been reported28
a. Adaptive stress response
i. When synthesis of (1-3)--D-glucan is inhibited, cell wall increases synthesis of chitin
ii. Compensatory increase in chitin synthesis occurs in response to echinocandin exposure in Candida
spp.
8
iii. Has been seen in most Candida spp., but is more common in C. albicans
iv. Elevated cell wall chitin content is associated with protection against echinocandins
v. Some studies show paradoxical effect of high levels of caspofungin on C. albicans, where C. albicans is
able to grow at higher serum concentrations
1. May be due to higher chitin contents in fungal cell wall, reducing ability of caspofungin to
exhibit its mechanism of action
b. Acquired FKS mutations
i. Glucan synthase (molecular target of echinocandins) is made of at least two subunits: FKS1p
(encoded by genes FKS1, FKS2, and FKS3) and Rho1p
ii. Amino acid substitutions in FKS subunits confer reduced echinocandin susceptibility
iii. These mutations can be gained, and have been seen in C. albicans, C. tropicalis, C. krusei, and C.
glabrata
c. Intrinsic FKS mutations
i. C. parapsilosis and C. guilliermondii have naturally occurring mutations in FKS genes that confer
reduced susceptibility to echinocandins
ii. These species have consistently higher MIC values than other Candida spp.
iii. Echinocandins have still shown clinical efficacy in treatment of these species
5. Resistance is emerging as use of echinocandins increases
a. Pre-exposure to fluconazole or caspofungin was associated with decreased prevalence of C. albicans and an
increased prevalence of less susceptible species, such as C. glabrata and C. krusei, in a multicenter
prospective observational study of candidemia29
i. Recent exposure to caspofungin was independently associated with increased risk of infection with
isolate with decreased echinocandin susceptibility (OR 4.79, 95% CI 2.47-9.28, p < 0.001)
b. A single-center study evaluating echinocandin resistance in C. glabrata showed resistance increased from
4.9% of isolates in 2001 to 12.3% in 201030
i. In addition, prior echinocandin therapy independently predicted presence of an FKS mutant strain of
C. glabrata (OR 19.647, 95% CI 7.19-58.1)
Empiric Antifungal Therapy in Critically Ill Patients
1. Various treatment strategies have been suggested to reduce mortality with IC31
a. Prophylaxis = treat patients with high-risk of IC (> 5-10%) to prevent development of infection
b. Pre-emptive therapy = treat patients with one or more biological markers of infection risk (e.g. elevated β-D-
glucan, widespread Candida spp. colonization)
c. Empiric treatment = treat patients with continued signs of infections and clinical suspicion of IC without
proven fungal infection
2. Guidelines suggest therapy should be considered in critically ill patients with risk factors for IC and no other cause of
fever10
Table 5. Risk factors for invasive candidiasis10
Candida spp. colonization Increased severity of illness
Exposure to broad-spectrum antibiotics Recent major surgery
Abdominal surgery/procedures Hemodialysis
Total parenteral nutrition (TPN) High-dose corticosteroids
Use of central venous catheters Necrotizing pancreatitis
9
3. Epidemiologic studies cited by guidelines for risk factors were done in surgical ICU patients32,33
a. Largest epidemiologic study of patients with candidemia identified risk factors listed above, but only prior
surgery (RR 7.3, 95% CI 1.0-53.8, p = 0.05), acute renal failure (RR 4.2, 95% CI 2.1-8.3, p < 0.001), and receipt
of parenteral nutrition (RR 3.6, 95% CI 1.8-7.5, p < 0.001) remained significant in multivariate analysis32
b. In subgroup of patients who actually underwent surgery, above risk factors remained significant as well as
triple-lumen catheter placement (RR 5.4, 95% CI 1.2-23.6, p = 0.03)
4. Risk factors included in guidelines are broad and apply to almost every ICU patient, however incidence of candidiasis
in the ICU is only about 10%10
Clinical Controversy #1
How do we predict who has invasive candidiasis in critically ill patients?
Table 6. Risk scores for prediction of invasive candidiasis and/or candidemia
Study Patient Population Risk Scoring Results
León C, Ruiz-Santana S, Saavedra P, et al. A bedside scoring system ("Candida score") for early antifungal treatment in nonneutropenic critically ill patients with Candida colonization. Crit Care Med. 2006;34:730-7.
34
Patients admitted for at least seven days to ICU
Excluded neutropenic patients
TPN = 1 point
Surgery on ICU admission = 1 point
Multifocal Candida spp. colonization = 1 point
Severe sepsis = 2 points
A score of > 2.5 had sensitivity of 81% and specificity of 74% for diagnosis of IC
OR of proven infection with score > 2.5 was 7.75 (95% CI 4.74-12.66)
AUROC = 0.847 (95% CI 0.8-0.894)
Shorr AF, Tabak YP, Johannes RS, et al. Candidemia on presentation to the hospital: development and validation of a risk score. Crit Care. 2009;13:R156.
35
Admitted to acute care hospital
Age < 65 years
Temp ≤ 98° F OR severe altered mental status
Cachexia
Previous hospitalization within 30 days
Admitted from other healthcare facility
Mechanical ventilation at admission
Candidemia rates by score (P < 0.0001) o 0 = 0.4% o 1 = 0.8% o 2 = 1.6% o 3 = 3.2% o 4 = 4.2% o 5 = 9.6% o 6 = 27.3%
AUROC = 0.70
NPV 99% with score < 3
Hermsen ED, Zapapas MK, Maiefski M, et al. Validation and comparison of clinical prediction rules for invasive candidiasis in intensive care unit patients: a matched case-control study. Crit Care. 2011;15:R198.
36
ICU stay ≥ 4 days
Excluded patients with IC or receipt of antifungal agents prior to day 4 of ICU stay
Broad-spectrum antibiotics on any of days 1-3 of ICU stay = 1.537 points
CVC on any of days 1-3 of ICU stay = 0.873 points
TPN on any of days 1-3 of ICU stay = 0.922 points
Corticosteroids from 7 days before ICU admission to day 3 of ICU stay = 0.402 points
Abdominal surgery = 0.879 points
Increasing pre-ICU LOS = 0.039 per day
Sensitivity 84.1% and specificity 60.2% for score > 2.45
PPV 4.7% and NPV 99.4% for score > 2.45
AUROC = 0.77
Abbreviations: AUROC = area under the receiver operating curve, CVC = central venous catheter, LOS = length of stay, NPV = negative predictive value, OR = odds ratio, PPV = positive predictive value,
10
1. Further studies confirmed no score is perfect, either having low sensitivity (thus missing many patients with IC) or low
specificity (including many patients without IC)
a. Validation study of Candida score in patients with sepsis or septic shock showed Candida score > 3 had PPV of
23.8% and NPV of 100% for IC37
b. Study comparing use of (1-3)--D-glucan levels versus Candida score or Candida colonization index showed
varied sensitivity, specificity, PPV, and NPV38
Table 7. Sensitivity, specificity, PPV, and NPV of various prediction tools38
2. Development of a clinical prediction rule for candidemia specifically in patients presenting with sepsis or septic shock
Table 8. Guillamet CV, Vazquez R, Micek ST, et al. Development and validation of a clinical prediction rule for candidemia in hospitalized patients with severe sepsis and septic shock. J Crit Care. 2015;30:715-20.39
Design and Methods
Retrospective, cohort study to develop and internally validate a prediction rule to identify patients with sepsis or septic shock at risk for candidemia
Population Inclusion Exclusion
Presence of positive blood culture combined with primary or secondary ICD-9-CM codes indicative for sepsis and acute organ dysfunction and/or need for vasopressors
o Sepsis had to be temporally related (± 24 hours) to positive blood cultures
Isolation of usual blood culture contaminants (e.g. CoNS, Corynebacterium spp.)
o Included if had multiple cultures positive for same organism or if clinical scenario qualified organism as true pathogen
Factors Evaluated Presence of CVC for ≥48 hours prior to positive blood cultures
Mechanical ventilation
Septic shock (versus sepsis)
Prior antibiotic use within preceding 30 days
Diabetes mellitus
Presence of immunosuppression (e.g. hematologic malignancies, solid organ or bone marrow transplants, AIDS, long term or high dose corticosteroid administration, or administration of chemotherapy and/or radiation therapy)
TPN
Recent surgery
Duration of hospitalization prior to bloodstream infection
Recent hospitalization within preceding 90 days
Endpoints Presence of candidemia
Baseline Characteristics
n = 2597
Variable Candidemia
(n = 266) Non-candidemia
(n = 2331) p-value
Age in yrs, mean ± SD 61.4 ± 16 60.3 ± 15.8 0.275
Sex, male, n (%) 125 (47) 1299 (55.8) 0.008
Admission source, n (%) <0.001
Home 124 (46.6) 1540 (66.1)
Nursing home 31 (11.7) 197 (8.5)
Transferred from outside hospital 107 (40.2) 559 (24)
Charlson comorbidity score, median (IQR) 4 (2-7) 4 (2-7) 0.532
11
Diabetes mellitus, n (%) 71 (26.7) 630 (27) 0.907
Cirrhosis, n (%) 40 (15) 345 (14.8) 0.918
Hemodialysis, n (%) 46 (17.3) 291 (12.5) 0.023
Immunosuppression, n (%) 84 (31.6) 819 (35.3) 0.259
TPN, n (%) 51 (19.2) 125 (5.4) <0.001
Surgery, n (%)
Abdominal 54 (20.3) 209 (13.3) 0.001
Any surgery 99 (37.2) 602 (25.8) <0.001
CVC, n (%) 213 (80.1) 1320 (56.6) <0.001
Prior hospitalization, n (%) 142 (58.9) 1190 (51.1) 0.168
Prior antibiotics, n (%) 197 (74.1) 1215 (52.1) <0.001
Days hospitalization prior to candidemia, median (IQR)
Probability prediction equation (using 1 if factor is present and 0 if factor absent): o (0.93 x TPN) + (0.88 x prior antibiotics within 30 days) + (0.61 x transfer from outside hospital) + (0.81 x admission from
nursing home) + (0.52 x mechanical ventilation) + (0.8 x presence of CVC for at least 48 hours) + (-3.22 x lung as presumed infection source) – 3.61
Area under the receiver operating characteristic curve (AUROC) for prediction equation was 0.798 (95% CI 0.77-0.82), indicating
12
that the prediction equation will assign a randomly chosen positive patient a higher prediction score than a randomly chosen negative patient 80% of the time
Integer score, where score >3 indicates high risk of candidemia: o Prior antibiotics within 30 days = 2 points o CVC at least 48 hours = 2 points o Admitted from nursing home = 2 points o TPN = 2 points o Transferred from outside hospital = 1 point o Mechanical ventilation = 1 point o Lung as presumed source of infection = -6 points
AUROC for integer score with value of 3 to predict candidemia was 0.72 (0.7-0.74)
Specificity of score of 3 or higher was 87.6% (83-91.3%) and sensitivity was 55.9% (53.9-57.9%)
PPV of score of 3 or higher was 18.5% (16.4-20.7%) and NPV was 97.5% (96.5-98.3%) o Using cutoff score of 2 provided a negative predictive value of 98.8%
Authors’ Conclusions
The developed and validated prediction rule outperformed previous prediction rules
Locally derived prediction models may be superior by accounting for local case mix and risk factor distribution
Reviewer’s Critique
Strengths Limitations
Large sample size
Specifically evaluated patients with sepsis or septic shock
Prediction equation had highest AUROC of all available prediction scores (other scores ranged between 0.595 and 0.762)
Retrospective study
Score has not been externally validated
Although prediction equation had best AUROC, simplified version had lower AUROC than other rules
Unable to include dose of corticosteroids in analysis
Only evaluated risk of candidemia and not invasive candidiasis
Conclusion: Use of a complicated risk prediction equation was superior to all other models for prediction of candidemia in sepsis and septic shock, however the simplified version, which would be easier to implement in practice, was no better than other models.
3. Take-Home Points
a. First study to develop risk score specifically in patients with sepsis or septic shock
b. Useful for assessing risk to decide whether antifungal treatment is warranted
c. Risk tool still has limitations, including difficulty implementing into practice, low PPV, and limitation to
candidemia only
Clinical Controversy #2
Which patients should be treated empirically with an echinocandin versus fluconazole?
Table 9. Ostrosky-Zeichner L, Harrington R, Azie N, et al. A risk score for fluconazole failure among patients with candidemia. Antimicrob Agents Chemother. 2017;64:e02091-16.40
Design and Methods
Retrospective, cohort study to develop a risk score for fluconazole failure in patients with candidemia
Population Inclusion Exclusion
Aged 18 years or older
At least one positive blood culture for Candida spp.
Initiation of IV fluconazole treatment during hospital stay and no more than 5 days before positive Candida spp. blood culture
Second hospitalization in study period (4/01/2004 to 03/31/2013)
Time from admission until start of fluconazole treatment
Fluconazole use prior to positive blood culture
Underlying diagnosis
Procedures received (e.g. CVC placement, drain placement, ileostomy)
Treatments (e.g. parenteral or enteral nutrition, corticosteroids, other antifungals)
Endpoints Failure of fluconazole, defined as patients who met any of the following criteria o Switched to or added another antifungal after fluconazole initiation o Had a subsequent positive blood culture for Candida infection during index hospitalization and at
least 10 days after fluconazole initiation o Died during period from index date to end of index hospitalization
Baseline Characteristics
n = 987
Variable Fluconazole failure
(n = 488) No fluconazole failure
(n = 499) p-value
Age in yrs, mean ± SD 60.95 ± 16.55 61.24 ± 17.68 0.7
Sex, male, n (%) 252 (51.6) 255 (51.1) 0.866
Days from admission to fluconazole initiation, mean ± SD
22.88 ± 66.52 12.76 ± 15.2 <0.001
Fluconazole use prior to positive blood culture, n (%)
*other antifungals = amphotericin B, flucytosine, itraconazole, voriconazole, posaconazole, or an echinocandin
Candida sp. isolated, n (%) Fluconazole failure
(n = 488) No fluconazole failure
(n = 499) p-value
Candida albicans 256 (52.5) 291 (58.3) 0.064
Candida glabrata 134 (27.5) 84 (16.8) <0.001
Candida krusei 6 (1.2) 2 (0.4) 0.173
Candida parapsilosis 48 (9.8) 62 (12.4) 0.196
Other Candida spp. 47 (9.6) 65 (13.0) 0.093
14
Results
Risk factor Coefficient Standard Error Odds Ratio
Days to initiation of fluconazole after admission
0.02 0.01 1.02
Candida sp.
Candida glabrata 0.68 0.20 1.98
Candida krusei 2.1 1.12 8.20
Hematologic malignancy 1.18 0.60 3.25
Venous thromboembolism -0.67 0.41 0.51
Nonoperative intubation or irrigation 0.43 0.29 1.54
Mechanical ventilation 0.12 0.24 1.13
Enteral nutrition 0.44 0.32 1.56
Other antifungal use 0.34 0.21 1.4
AUROC was 0.65, meaning model will give a randomly selected positive case a higher score than a randomly selected negative case 65% of the time
Risk score cutoff # correctly predicted*
Sensitivity Specificity PPV NPV
0.1 163 100 0 49 0
0.2 162 99 2 50 75
0.3 158 97 17 53 85
0.4 127 78 42 57 66
0.5 91 56 68 63 61
0.6 53 33 84 66 56
0.7 25 15 95 74 63
0.8 10 6 98 77 52
0.9 5 3 100 100 51
Authors recommend a cutoff of 0.5 to balance sensitivity and specificity
Authors’ Conclusions
This study identified clinical factors that may predict fluconazole failure in hospitalized patients with candidemia
Reviewer’s Critique
Strengths Limitations
First study to provide scoring system to predict fluconazole failure
Included many different variables to assess risk of failure
Retrospective, claims-based study
AUROC is less than generally accepted cutoff for a “good” model (0.7 or higher)
Recommended cutoff of 0.5 only provided 56% sensitivity and 68% specificity
Conclusion: This prediction model for risk of fluconazole failure may be used in combination with other tools to assess whether patients at high risk of candidiasis may benefit from initial echinocandin therapy over fluconazole therapy.
1. Take-Home Points
a. Although this tool may help predict patients who need echinocandin therapy rather than fluconazole therapy,
the tool did not perform well enough to recommend in clinical practice
b. Has not been validated as a tool to make initial treatment decisions
c. Risk score does not have the best accuracy, with an AUROC of 0.65
15
Clinical Controversy #3
What is the clinical benefit of early antifungal therapy in critically ill patients with sepsis?
Table 10. Timsit JF, Azoulay E, Schwebel C, et al. Empirical micafungin treatment and survival without invasive fungal infection in adults with ICU-acquired sepsis, Candida colonization, and multiple organ failure: the EMPIRICUS randomized controlled trial. JAMA. 2016;316:1555-64.41
Design and Methods
Multicenter, double-blind, placebo-controlled trial to evaluate whether micafungin increases 28-day invasive fungal infection-free survival in patients with ICU-acquired sepsis, Candida spp. colonization at multiple sites, and multiple organ failure
Population Inclusion Exclusion
Mechanically ventilated for at least 5 days
At least 1 colonization site (other than rectal swab or stool) positive for Candida sp.
At least 1 additional organ dysfunction (SOFA score ≥ 3)
Previous treatment for > 4 days using broad-spectrum antibacterial agents within last 7 days
Arterial or CVC
New finding of ICU-acquired sepsis of unknown origin (based on SIRS criteria)
Neutropenia (< 500 WBCs/mm2)
Previous organ or stem cell transplant
Ongoing systemic immunosuppression agent therapy other than 2 mg/kg/day of prednisolone or equivalent
Recent chemotherapy (within past six months)
Proven invasive infection at time of randomization
Antifungal treatment with an echinocandin agent for more than one day or with any other antifungal agent for more than 72 hours during the week prior to inclusion
Intervention Micafungin 100 mg IV daily x 14 days (n = 128) vs placebo (n = 123)
Endpoints 28-day survival free of proven invasive fungal infection o Also assessed in subgroups: medical vs surgical, low vs high SOFA score, low vs high (1-3)-β-D-
glucan level, low vs high colonization index, Candida score < 3 vs ≥ 3
New proven invasive fungal infections during follow-up
Survival at days 28 and 90
Antifungal-free survival at day 28
Incidence of ventilator-associated bacterial pneumonia
Evolution throughout 28-day study period of SOFA score and (1-3)-β-D-glucan
Baseline Characteristics
N = 251
Micafungin (n = 128) Placebo (n = 123)
Age in years, median (IQR) 65 (56-74) 64 (52-74)
Men, n (%) 81 (66) 82 (64)
Weight in kg, median (IQR) 84 (72-97) 80 (68-95)
Chronic disease categories, n (%)
Cardiac 30 (24) 34 (27)
Respiratory 20 (16) 33 (26)
Hepatic 11 (9) 14 (11)
Renal 15 (12) 7 (6)
Immunosuppression 4 (3) 8 (6)
Receiving corticosteroids, n (%) 11 (9) 11 (9)
Admission category, n (%)
Medical 92 (75) 94 (73)
Emergency surgery 29 (24) 31 (24)
Scheduled surgery 2 (2) 3 (2)
Main surgical procedures, n (%)
Cardiac 25 (20) 94 (73)
Abdominal 5 (4) 31 (24)
Other surgery or trauma 2 (2) 4 (3)
16
Main reason for ICU admission, n (%)
Acute respiratory failure 48 (39) 54 (41)
Septic shock 37 (31) 48 (37)
Cardiogenic shock 21 (17) 17 (13)
Coma 15 (12) 10 (8)
Acute pancreatitis 7 (6) 7 (6)
Duration of ICU stay prior to inclusion in days, median (IQR)
11 (7-17) 10 (7-15)
Variables assessed at inclusion
SOFA score, median (IQR) 8 (5-12) 8 (6-11)
Candida score, median (IQR) 3 (2.5-4) 3 (2-4)
No. of positive colonization sites, median (IQR)
3 (2-4) 3 (2-4)
Epinephrine or norepinephrine use, n (%) 70 (57) 71 (56)
# of invasive fungal infections at follow-up (day 28)
≥ 1 4 (3) 15 (12) 9.1 (2.5 – 16.3)
2 0 2 (2) 1.6 (-1.5 – 5.7)
Invasive fungal infections by species
Candida albicans 3 10 19.4 (-29.7 – 49.4)
Candida glabrata 0 2 11.1 (-38.5 – 32.8
Candida parapsilosis 0 3 16.7 (-33.5 – 39.2)
Candida inconspicua 1 0 25.0 (-2.0 – 69.9)
Trichosporon 0 2 11.1 (-38.5 – 32.8)
Aspergillus fumigatus 0 1 5.6 (-43.7 – 25.8)
17
Authors’ Conclusions
Among nonneutropenic critically ill patients with ICU-acquired sepsis, Candida species colonization at multiple sites, and multiple organ failure, empirical treatment with micafungin did not increase fungal infection-free survival at day 28
Reviewer’s Critique
Strengths Limitations
Double-blind, placebo-controlled randomized trial
Excluded transplant and neutropenic patients
Only included patients at highest risk of invasive candidiasis (exposure to broad-spectrum antibiotics, presence of central venous catheter)
Underpowered (expected 18% difference, found 8% difference)
Included invasive candidiasis diagnosed by samples found before initiation of therapy as an occurrence of the primary endpoint
Low percentage of abdominal surgery and acute pancreatitis as these patients are at higher risk of IC
Reviewer’s Conclusions
Underpowered to find an improvement in mortality with use of empiric echinocandin therapy in patients at highest risk of invasive candidiasis
a. Patients at risk for invasive candidiasis are extremely sick
i. Median SOFA score was 8, correlating with 15-20% mortality42,43
ii. Risk factors for IC are seen in patients with many chronic issues with a higher baseline risk of death
iii. Presence of IC could arguably be considered marker of disease severity in ICU patients
iv. Questions futility of echinocandin intervention in patients at high-risk
b. Micafungin was started late in hospital stay
i. Study only evaluated patients who had received broad-spectrum antibiotics for four or more days
ii. With knowledge that early treatment is associated with better outcomes, delayed micafungin
initiation may have resulted in lower rate of efficacy
c. Higher doses may have been needed to hit pharmacokinetic/pharmacodynamic targets
i. In population pharmacokinetic study of critically ill patients based on data from EMPIRICUS trial,
results showed higher doses may be needed to meet AUC:MIC targets for efficacy44
1. Covariates that significantly influenced clearance (CL), central distribution volume (Vc), and
peripheral distribution volume (Vp) were body weight, serum albumin, and SOFA score
2. In simulations of patients with C. albicans or C. glabrata with MICs > 0.015 mg/L, micafungin
doses of at least 150 mg were needed to achieve high probability of target attainment (PTA)
a. 100 mg dose adequate for isolates with MICs < 0.015 mg/L
3. In simulations of patients with C. parapsilosis, doses between 150 mg and 300 mg were
necessary to reach a PTA of 90% for isolates with MICs = 0.25-0.5 mg/L
a. High PTA never able to be obtained with isolates with MIC > 1 mg/L
b. 100 mg dose only adequate in patients with isolates with MICs ≤ 0.125 mg/L
c. Lower PTAs seen in models of patients with serum albumin < 2.5 g/dL
ii. Micafungin pharmacokinetics in obese, critically ill, and morbidly obese critically ill patients showed
difficulty meeting AUC/MIC targets45
1. 100 mg dose was never able to achieve > 90% fractional target attainment (FTA) for any
Candida spp.
2. 150 mg dose met 90% FTA for C. albicans in patients ≤115 kg, but never achieved > 90% for
other species
3. 200 mg dose meet 90% FTA for C. albicans in all patients and C. glabrata in patients ≤ 115 kg,
but never met goal for C. tropicalis or C. parapsilosis
18
Recommendation
1. Guillamet score should be used to assess the risk of candidemia in patients with septic shock presenting from the
community
a. Full calculation should be used rather than simplified scoring system better accuracy
2. Patients in the ICU for ≥7 days can be assessed with Candida score better accuracy than Guillamet score
3. For patients with either score ≥ 3, prior antifungal history should be assessed
a. Patients with history of fluconazole use should receive empiric echinocandin therapy
i. Should be given at least 150 mg of micafungin or 200 mg if ≥ 115 kg
b. Patients with no history of antifungal therapy AND low SOFA score (< 8) may be initially managed with
fluconazole
4. Blood cultures should be drawn before receipt of antifungal therapy, and T2Candida® and/or Biofire PCR testing
should be utilized if possible to enable targeted therapy
5. Empiric therapy can be discontinued in the following situations: a. Negative blood cultures after 5 days b. No clinical improvement after 4-5 days without proven IC c. Back-to-back negative (1-3)-β-D-glucan on separate days
6. If blood cultures come back positive, therapy can be de-escalated based on organism and susceptibilities
Conclusion
1. Although early treatment of IC is extremely important, diagnosis remains immensely difficult
2. Patients presenting with sepsis or septic shock with extensive exposure to the healthcare system should be assessed
for risk of IC
3. Care should be taken to avoid overuse of echinocandins in light of emerging resistance
4. Although IC is associated with high mortality, patients who are diagnosed are generally chronically ill, and the disease
6. Vincent JL, Rello J, Marshall J, et al. International study of the prevalence and outcomes of infection in intensive care units. JAMA.
2009;302:2323-9.
7. Quenot JP, Binquet C, Kara F, et al. The epidemiology of septic shock in French intensive care units: the prospective multicenter cohort EPISS
study. Crit Care. 2013;17:R65.
8. Kollef M, Micek S, Hampton N, et al. Septic shock attributed to Candida infection: importance of empiric therapy and source control. Clin Infect
Dis. 2012;54:1739-46.
9. Bassetti M, Righi E, Ansaldi F, et al. A multicenter study of septic shock due to candidemia: outcomes and predictors of mortality. Intensive Care
Med. 2014;40:839-45.
10. Pappas PG, Kauffman CA, Andes DR, et al. Clinical practice guideline for the management of candidiasis: 2016 update by the Infectious Diseases
Society of America. Clin Infect Dis. 2016;62:e1-50.
11. Fan D, Coughlin LA, Neubauer MM, et al. Activation of HIF-1α and LL-37 by commensal bacteria inhibits Candida albicans colonization. Nat Med.
2015;21:808-14.
12. Netea MG, Joosten LA, van der Meer JW, et al. Immune defence against Candida fungal infections. Nat Rev Immunol. 2015;15:630-42.
13. Lionakis MS, Netea MG. Candida and host determinants of susceptibility to invasive candidiasis. PLoS Pathog. 2013;9:e1003079.
14. Zaoutis TE, Prasad PA, Localio AR, et al. Risk factors and predictors for candidemia in pediatric intensive care unit patients: implications for
prevention. Clin Infect Dis. 2010;51:e38-45.
15. Centers for Disease Control and Prevention. Candida auris. Available at: https://www.cdc.gov/fungal/candida-auris/index.html. Accessed July 15,
2018.
16. Miller JM, Binnicker MJ, Campbell S, et al. A guide to utilization of the microbiology laboratory for diagnosis of infectious diseases: 2018 updated
by the Infectious Diseases Society of America and the American Society for Microbiology. Clin Infect Dis. 2018 Jun 28. doi: 10.1093/cid/ciy381.
[Epub ahead of print]
17. Desoubeaux G, Franck-Martel C, Bailly E, et al. Prospective assessment of FilmArray® technology for the rapid identification of yeast isolated
from blood cultures. J Microbiol Methods. 2014;106:119-122.
18. Mylonakis E, Clancy CJ, Ostrosky-Zeichner L, et al. T2 magnetic resonance assay for the rapid diagnosis of candidemia in whole blood: a clinical
trial. Clin Infect Dis. 2015;60:892-9.
19. Wilson NM, Alagaden G, Tibbetts RJ, et al. T2 magnetic resonance assay improves timely management of candidemia. J Antimicrob Stew.
2017;1:12-8.
20. Karageorgopoulos DE, Vouloumanou EK, Ntziora F, et al. β-D-glucan assay for the diagnosis of invasive fungal infections: a meta-analysis. Clin
Infect Dis. 2011;52:750-70.
21. Lu Y, Chen YQ, Guo YL, et al. Diagnosis of invasive fungal disease using serum (1→3)-β-D-glucan: a bivariate meta-analysis. Intern Med.
2011;50:2783-91.
22. Onishi A, Sugiyama D, Kogata Y, et al. Diagnostic accuracy of serum 1,3-β-D-glucan for Pneumocystis jirovecii pneumonia, invasive candidiasis,
and invasive aspergillosis: systematic review and meta-analysis. J Clin Microbiol. 2012;50:7-15.
23. Cappelletty D, Eiselstein-McKitrick K. The echinocandins. Pharmacotherapy. 2007;27:369-88.
24. Nett JE, Andes DR. Antifungal agents: spectrum of activity, pharmacology, and clinical indications. Infect Dis Clin North Am. 2016;30:51-83.
25. Reboli AC, Rotstein C, Pappas PG, et al. Anidulafungin versus fluconazole for invasive candidiasis. N Engl J Med. 2007;356:2472-82.
26. Andes DR, Safdar N, Baddley JW, et al. Impact of treatment strategy on outcomes in patients with candidemia and other forms of invasive
candidiasis: a patient-level quantitative review of randomized trials. Clin Infect Dis. 2012;54:1110-22.
27. Eschenauer GA, Carver PL, Lin SW, et al. Fluconazole versus an echinocandin for Candida glabrata fungaemia: a retrospective cohort study. J
Antimicrob Chemother. 2013;68:922-6.
28. Beyda ND, Lewis RE, Garey KW. Echinocandin resistance in Candida species: mechanisms of reduced susceptibility and therapeutic approaches.
Ann Pharmacother. 2012;46:1086-96.
29. Lortholary O, Desnos-Ollivier M, Sitbon K, et al. Recent exposure to caspofungin or fluconazole influences the epidemiology of candidemia: a
prospective multicenter study involving 2,441 patients. Antimicrob Agents Chemother. 2011;55:532-8.
20
30. Alexander BD, Johnson MD, Pfeiffer CD, et al. Increasing echinocandin resistance in Candida glabrata: clinical failure correlates with presence of