Journal of Critical Care xxx (2014) xxx–xxx

Contents lists available at ScienceDirect

Journal of Critical Care journal homepage: www.jccjournal.org

Prognostic factors in critically ill cancer patients admitted to the intensive care unit Gulbin Aygencel, MD a,⁎, Melda Turkoglu, MD a, Gulsan Turkoz Sucak, MD b, Mustafa Benekli, MD c a b c

Gazi University Faculty of Medicine, Department of Internal Medicine, Division of Critical Care Medicine, Ankara, Turkey Gazi University Faculty of Medicine, Department of Internal Medicine, Division of Hematology, Ankara, Turkey Gazi University Faculty of Medicine, Department of Internal Medicine, Division of Medical Oncology, Ankara, Turkey

a r t i c l e Keywords: ICU mortality prognostic factor cancer patient

i n f o

a b s t r a c t Objective: The objective of this study is to identify factors predicting intensive care unit (ICU) mortality in cancer patients admitted to a medical ICU. Patients and methods: We conducted a retrospective study in 162 consecutive cancer patients admitted to the medical ICU of a 1000-bed university hospital between January 2009 and June 2012. Medical history, physical and laboratory findings on admission, and therapeutic interventions during ICU stay were recorded. The study end point was ICU mortality. Logistic regression analysis was performed to identify independent risk factors for ICU mortality. Results: The study cohort consisted of 104 (64.2%) patients with solid tumors and 58 patients (35.8%) with hematological malignancies. The major causes of ICU admission were sepsis/septic shock (66.7%) and respiratory failure (63.6%), respectively. Overall ICU mortality rate was 55 % (n = 89). The ICU mortality rates were similar in patients with hematological malignancies and solid tumors (57% vs 53.8%; P = .744). Four variables were independent predictors for ICU mortality in cancer patients: the remission status of the underlying cancer on ICU admission (odds ratio [OR], 0.113; 95% confidence interval [CI], 0.027-0.48; P = .003), Acute Physiology and Chronic Health Evaluation II score (OR, 1.12; 95% CI, 1.032-1.215; P = .007), sepsis/septic shock during ICU stay (OR, 8.94; 95% CI, 2.28-35; P = .002), and vasopressor requirement (OR 16.84; 95% CI, 3.98-71.24; P = .0001). Although Acute Physiology and Chronic Health Evaluation II score (OR, 1.30; 95% CI, 1.054-1.61; P = .014), admission through emergency service (OR, 0.005; 95% CI, 0.00-0.69; P = .035), and vasopressor requirement during ICU stay (OR, 140.64; 95% CI, 3.59-5505.5; P = .008) were independent predictors for ICU mortality in patients with hematological malignancies, Sequential Organ Failure Assessment score (OR, 1.83; 95% CI, 1.29-2.6; P = .001), lactate dehydrogenase level on admission (OR, 1.002; 95% CI, 1-1.005; P = .028), sepsis/septic shock during ICU stay (OR, 138.4; 95% CI, 12.54-1528.4; P = .0001), and complete or partial remission of the underlying cancer (OR, 0.026; 95% CI, 0.002-0.3; P = .004) were the independent risk factors in patients with solid tumors. Conclusion: Intensive care unit mortality rate was 55% in our cancer patients, which suggests that patients with cancer can benefit from ICU admission. We also found that ICU mortality rates of patients with hematological malignancies and solid tumors were similar. © 2014 Elsevier Inc. All rights reserved.

1. Introduction The growing number of patients living with cancer leads to a similar increase in the number of patients requiring intensive care. Despite the general opinion that admission of cancer patients to intensive care units (ICUs) is usually futile and costly based on some older studies, ICU survival has been reported to be improved

significantly in recent studies [1–4]. Increased survival expectancy in critically ill cancer patients led conduction of studies that investigate the prognostic factors that predict ICU outcome and guide ICU admission and management strategies [1–8]. We, therefore, analyzed our data retrospectively to determine the characteristics and outcomes of cancer patients admitted to our medical ICU and to identify the risk factors associated with ICU mortality. 2. Patients and methods

⁎ Corresponding author at: Gazi University Faculty of Medicine, Department of Internal Medicine, Division of Critical Care Medicine, Besevler, 06510, Ankara, Turkey. Tel.: + 90 312 2024216. E-mail addresses: [email protected] (G. Aygencel), [email protected] (M. Turkoglu), [email protected] (G. Turkoz Sucak), [email protected] (M. Benekli).

2.1. Study design This study is a retrospective, observational study conducted in the 9bed medical ICU of the Gazi University Hospital, a 1000-bed university

0883-9441/$ – see front matter © 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.jcrc.2014.01.014

Please cite this article as: Aygencel G, et al, Prognostic factors in critically ill cancer patients admitted to the intensive care unit, J Crit Care (2014), http://dx.doi.org/10.1016/j.jcrc.2014.01.014

2

G. Aygencel et al. / Journal of Critical Care xxx (2014) xxx–xxx

hospital in Ankara, Turkey. Every adult patient (≥18 years old) with histologically proven cancer who required ICU admission was evaluated between January 1, 2009, and June 30, 2012. When ICU admission is considered in a cancer patient in our center, life expectancy should be longer than 3 months, and/or further treatment options to treat the underlying cancer should be available. Only the first admission was recorded in patients with multiple ICU admissions. Patients who stayed in the ICU for shorter than 24 hours were also excluded. This study was approved by the institutional review board. A total of 162 consecutive cancer patients admitted to ICU during the study period were included in the study. The following information was abstracted from the medical charts of the patients: age and sex; comorbidities; type of cancer; characteristics of the cancer including presence of metastases; current status of the underlying cancer (complete or partial remission, relapsed, or progressive1 disease); treatment modality that includes surgery, chemotherapy, and radiation therapy; the patient's preadmission performance status as determined by Eastern Cooperative Oncology Group (ECOG2) scale; hematopoietic stem cell transplantation (HSCT) status and type of HSCT (autologous and allogeneic); cause of ICU admission; source of admission (internal medicine, emergency service, etc); time from hospital to ICU admission; blood chemistries and complete blood count on day 1; presence and site of infection on admission and during ICU stay; severity of illness score using Acute Physiology and Chronic Health Evaluation (APACHE) II score; organ dysfunctions using Sequential Organ Failure Assessment (SOFA) score; therapeutic interventions during the ICU stay (use of vasopressors, mechanical ventilation, dialysis, chemotherapy), length of ICU stay; and ICU mortality rate. 2.2. Statistical analysis Data were analyzed using SPSS 13.5 for Windows (SPSS, Inc, Chicago, IL). Descriptive statistics were computed for all study variables. A Kolmogorov-Smirnov test was used, and histograms and normal-quantile plots were examined to verify the normality of distribution of continuous variables. Discrete variables are expressed as counts (percentage), and continuous variables, as means ± SD or median (interquartile range, 25%75%). For demographics and clinical characteristics of the study groups, differences between groups were assessed using a χ2, Fisher exact test, Student t test, or Mann-Whitney U test, as appropriate. Multivariate logistic regression analysis with ICU mortality as the dependent variable was conducted in cancer patients. Only variables associated with a higher risk of ICU mortality (P b .05) on a univariate basis were introduced in the multivariate model. P b .05 was considered statistically significant. 3. Results 3.1. Results in the whole study cohort There were 1130 admissions during the study period, among whom 162 (14.3%) were cancer patients who met the eligibility criteria of the study. Median age was 61 (48-71.3) years, and most patients were male (58.6%). The most common causes of ICU 1 Cancer recurrence or relapse is defined as the return of cancer after treatment and after a period during which the cancer cannot be detected. When cancer spreads or gets worse, it is called progression. When a treatment completely eliminates the tumor and the tumor cannot be seen on the tests or cannot be measured after a period, it is called a complete response or complete remission. A partial response or partial remission means the cancer partly responded to treatment. Treatment partly controls the tumor and reduces the tumor size. 2 Eastern Cooperative Oncology Group scale is used to assess how the disease affects the daily living abilities of the patient. They included the following: 0, fully active, able to carry on all predisease performance without restriction; 1, restricted in physically strenuous activity but ambulatory and able to carry out work of a light or sedantery nature, for example, light house work, office work; 2, ambulatory and capable of all self-care but unable to carry out any work activities; 3, capable of only limited selfcare, confined to bed or chair more than 50% of waking hours; and 4, completely disabled, cannot carry on any self-care, totally confined to bed or chair.

admission were sepsis/septic shock (66.7%) and respiratory failure (63.6%). The most common comorbidities of the patients were cardiovascular diseases, diabetes, and hypertension. Most patients were admitted to our medical ICU from emergency service. Two or more organ dysfunctions were found in 84 patients (52%) on admission. Renal and pulmonary dysfunctions were the most common organ dysfunctions. Most patients had thrombocytopenia (53.7%), whereas 46 patients had neutropenia. Sixty-eight patients required mechanical ventilation on admission. A possibility of an infectious etiology or isolation of a microorganism was seen in 147 patients on ICU admission. Fifty-eight patients (35.8%) in our cohort had hematological malignancies, and 104 patients (64.2%) had solid tumors. Lymphoma was the most common solid tumor, whereas acute leukemia was the most common hematological malignancy. Colon, lung, and breast cancers were the other common solid tumors in our cohort (all types of cancer in our study group are presented in Table 1). Fifty-five patients had relapsed or progressive cancer, whereas 54 patients were newly diagnosed cancer patients. Performance status of the cancer patients was well (ECOG, 0-2) in 117 patients according to ECOG performance scale on ICU admission. One hundred fifteen patients had been treated with chemotherapy, and 84 patients had cancer therapy within a month before ICU admission. Eighteen of the patients were stem cell transplant recipients. Renal replacement therapy was required in 53 patients (32.7%) during ICU stay. Mechanical ventilation (invasive or noninvasive) was applied to 109 patients. Intensive care unit–acquired nosocomial infection was detected in 62 patients (38.3%). Of the cancer patients, 73 (45%) survived, and 89 cancer patients (55%) died at the end of their ICU stay. Table 1 shows some baseline characteristics of the study population. 3.2. Results in hematological malignancies and solid tumors groups Patients are categorized into 2 groups in terms of their type of cancer (patients with hematological malignancies and patients with solid tumors). Patients with solid tumors were older than the patients with hematological malignancies. Sequential Organ Failure Assessment scores were significantly higher in patients with hematological malignancies than in those with solid tumors. Patients with hematological malignancies had longer duration of hospital stay before ICU admission. Comorbidities varied among the groups, with a higher prevalence in patients with solid tumors. Anticancer treatment within a month before ICU admission was more common in patients with hematological malignancies. The frequency of infection on ICU admission was also more common in patients with hematological malignancies. Episodes of bacteremia were more frequent in patients with hematological malignancies compared with patients with solid tumors. The number of organ failure and renal dysfunction was more common in patients with hematological malignancies during ICU stay. Although neutropenia on and during admission to ICU and thrombocytopenia were more common in patients with hematological malignancies, ICU mortality rate was similar in patients with hematological malignancies and solid tumors. Table 1 shows some characteristics of the patients with hematological malignancies and solid tumors in our ICU. 3.3. Results in survivors and nonsurvivors We subcategorized our cohort according to their outcome and reanalyzed: patients who survived (survivor—discharge or transfer) (73 patients, 45%) and who died (nonsurvivors) (89 patients, 55%). Acute Physiology and Chronic Health Evaluation II and SOFA scores were significantly lower, Glasgow Coma Scale (GCS) was significantly higher, and length of ICU stay and length of hospital stay before ICU admission were significantly shorter in survivors when compared with nonsurvivors. Nonsurvivors had more progressive and relapsed

Please cite this article as: Aygencel G, et al, Prognostic factors in critically ill cancer patients admitted to the intensive care unit, J Crit Care (2014), http://dx.doi.org/10.1016/j.jcrc.2014.01.014

G. Aygencel et al. / Journal of Critical Care xxx (2014) xxx–xxx

3

Table 1 Some baseline characteristics of cancer patients in our ICU Variable Age (y) Male sex Length of hospital stay before ICU admission (d) Length of ICU stay (d) Types of cancer Solid tumors Non-Hodgkin lymphoma Colon cancer Breast cancer Prostate cancer Lung cancer Urinary bladder cancer Gynecologic cancers Other solid tumorsa Hematological malignancies Acute myeloid leukemia Multiple myeloma Acute lymphoblastic leukemia Chronic lymphocytic leukemia Chronic myelogenous leukemia Characteristics of the cancer on ICU admission Newly diagnosed Complete or partial remission Relapsed or progressive disease End stage Patient's preadmission performance status Good (ECOG 0-2) Poor (ECOG 3 and 4) Cancer treatment status Chemotherapy Radiation therapy Surgery Cancer therapy within the last month before ICU admission BMT recipient Allogeneic BMT Comorbidities Chronic obstructive pulmonary disease Diabetes mellitus Hypertension Cardiovascular diseases Cause of ICU admission Sepsis/septic shock Respiratory failure Acute renal failure Source of admission Emergency service Morbidities on ICU admission Renal dysfunction Neutropenia Thrombocytopenia Requirement of mechanical ventilation (invasive or noninvasive) Hepatic dysfunction Vasopressors requirement ≥2 organ dysfunctions Suspected or culture-proven infection Bacterial infection Fungal infection Pulmonary infection Bloodstream infection On ICU admission APACHE II score GCS SOFA score Hemoglobin level (g/dL) Leukocytes (/mm3) Thrombocytes (/mm3) Sodium (Na, mEq/L) ALT (IU/L) LDH (IU/L) C-reactive protein (mg/L) Procalcitonin (ng/mL) Calcium (mg/dL) Albumin (g/dL)

Cancer patients in ICU (n = 162) 61 (48-71.3) 95 (59%) 4 (1-16) 5 (3-12)

Patients with hematological malignancies (n = 58) 52.5 (33-61.25) 40 (69%) 7 (1.75-26.25) 4 (3-9.25)

Patients with solid tumors (n = 104) 66.5 (33-61.25) 55 (53%) 2 (1-11) 5(3-14)

P .0001 .067 .001 .56

104 (64%) 26 (16%) 19 (12%) 12 (7%) 8 (5%) 7 (4%) 7 (4%) 6 (4%) 19 (12%) 58 (36%) 27 (16.7%) 13 (8%) 8 (5%) 6 (4%) 4 (2%) 54 (33%) 37 (23%) 55 (34%) 16 (10%)

21 (36%) 10 (17%) 25 (43%) 2 (3%)

33 27 30 14

117 (72.2%) 45 (27.8%)

39 (67%) 19 (33%)

78 (75%) 26 (25%)

.36

115 (71%) 41 (25%) 44 (27%) 84 (52%) 18 (11%) 16 (10%)

45 (78%) 7 (12%)

70 (67%) 34 (33%)

.207 .004

39 (67%) 14 (24%) 12 (21%)

45 (43%) 4 (4%) 4 (4%)

.005 .0001 .001

17 (10%) 33 (20%) 36 (22%) 24 (15%)

2 (3%) 4 (7%) 4 (7%) 5 (9%)

15 29 32 19

.033 .001 .0001 .11

108 (67%) 103 (64%) 51 (31%)

42 (72%) 40 (69%) 16 (28%)

66 (63%) 63 (61%) 35 (34%)

.298 .311 .483

55 (34%)

9 (16%)

46 (44%)

.0001

75 (46%) 46 (28%) 87 (54%) 68 (42%) 56 (35%) 54 (33%) 84 (52%) 147 (91%) 69 (43%) 21 (13%) 102 (63%) 28 (17%)

24 (41%) 37 (64%) 48 (83%) 29 (50%) 17 (29%) 23 (40%) 41 (71%) 55 (95%) 32 (55%) 13 (22%) 39 (67%) 17 (29%)

51 (49%) 9 (9%) 39 (38%) 39 (38%) 39 (38%) 31 (30%) 43 (41%) 92 (88%) 37 (36%) 8 (8%) 63 (61%) 11 (11%)

.412 .0001 .0001 .137 .308 .226 .001 .26 .02 .013 .498 .004

23 (18-29) 14 (10-15) 8 (5-10) 9.1 (8-10.6) 8870 (3150-15 950) 91 000 (36 000-192 000) 136 (135-140) 21 (12-39.8) 348 (246.5-779.25) 149 (76.5-248.5) 2.3 (0.65-20) 7.7 (7.1-8.3) 2.6 (2.2-2.9)

24 (18-32.5) 14 (10-15) 9 (6.25-12) 8.2 (6.8-9.45) 4300 (890-18 800) 43 000 (21050-83800) 136 (133.5-140) 16 (12-26) 345 (239-830) 180 (102.7-318) 5 (0.9-31) 7.4 (6.8-7.85) 2.5 (2.2-2.9)

(32%) (26%) (29%) (13%)

(14%) (28%) (31%) (18%)

22 (17-28) 14 (10-15) 7 (4-10) 9.7 (8.5-11.2) 10 180 (5400-15 350) 146 500 (66 750-232 250) 136 (132-141) 26 (13-46) 349 (246-697) 133 (63.8-208) 1.9 (0.39-10.92) 7.95 (7.32-8.5) 2.6 (2.2-2.9)

.604 .244 .084 .053

.217 .6 .003 .0001 .014 .0001 .55 .01 .66 .035 .048 .0001 .54

(continued on next page)

Please cite this article as: Aygencel G, et al, Prognostic factors in critically ill cancer patients admitted to the intensive care unit, J Crit Care (2014), http://dx.doi.org/10.1016/j.jcrc.2014.01.014

4

G. Aygencel et al. / Journal of Critical Care xxx (2014) xxx–xxx

Table 1 (continued) Variable Creatinine (mg/dL) Glucose (mg/dL) Morbidities during ICU stay Sepsis/septic shock Neutropenia Hepatic dysfunction Renal dysfunction Renal replacement therapy Requirement of mechanical ventilation Chemotherapy in ICU Blood/blood products transfusion Central venous catheterization Arterial catheterization Suspected or culture-proven infection Bacterial infection Pulmonary infection Steroid use Vasopressor use Results Survived Died

Cancer patients in ICU (n = 162)

Patients with hematological malignancies (n = 58)

1.1 (0.7-2.1) 138 (120-185) 73 27 24 48 53 109 10 96 88 125 62 57 56 119 99

0.9 (0.6-1.95) 139 (119-181

Patients with solid tumors (n = 104) 1.2 (0.7-2.27) 135.5 (120.3-200.8)

P .087 .82

(45%) (17%) (15%) (30%) (33%) (67%) (6%) (59%) (54%) (77%) (38%) (35%) (35%) (73%) (61%)

26 (45%) 21 (36%) 12 (21%) 24 (41%) 22 (38%) 42 (72%) 10 (17%) 47 (81%) 31 (53%) 49 (84%) 25 (43%) 24 (41%) 21 (36%) 47 (81%) 37 (64%)

47 (45%) 6 (6%) 12 (12%) 24 (23%) 31 (30%) 67 (64%) 0 (0%) 49 (47%) 57 (55%) 76 (73%) 37 (36%) 33 (32%) 35 (34%) 72 (69%) 62 (60%)

1 .0001 .165 .019 .3 .383 .0001 .0001 .871 .119 .4 .234 .863 .137 .619

73 (45%) 89 (55%)

25 (43%) 33 (57%)

48 (46%) 56 (54%)

.744

Values are presented as median (interquartile range) or n (%). BMT indicates bone marrow transplantation; ALT, alanine aminotransferase. a Other solid tumors: renal cell cancer, nasopharyngeal cancer, laryngeal cancer, hepatocellular carcinoma, esophageal cancer, and glial brain tumors.

cancer and worse performance status. Respiratory failure and sepsis/ septic shock on admission were significantly more common in nonsurvivors. Being thrombocytopenic, on mechanical ventilation, having infection were significantly more frequent in nonsurvivors. Platelet count was found to be significantly higher in survivors. Some inflammatory markers such as procalcitonin and C-reactive protein were lower in survivors. Nosocomial infections and acute renal failure were significantly more common, and renal replacement therapy was required more often in nonsurvivors during ICU stay. Requirement for invasive monitoring and administration of steroids was significantly more frequent during ICU stay in the nonsurvivor group. Requirement for tracheostomy and blood/blood product transfusion was also significantly more common in nonsurvivors. Table 2 shows some characteristics of the survivor and nonsurvivor cancer patients in ICU. 3.4. Results of multivariate analysis in our study group Length of hospital stay before ICU admission, length of ICU stay, status of the underlying cancer, performance status, causes of ICU admission, source of ICU admission, some comorbidities and infections on ICU admission, APACHE II score, GCS, SOFA score, some laboratory abnormalities (inflammatory markers, in particular), and some morbidities during ICU stay (septic shock, renal replacement therapy, requirement of vasopressors and steroid, etc) are the variables that were found to be significantly associated with increased ICU mortality in cancer patients in univariate analysis. However, only 3 parameters, namely, APACHE II score on admission (odds ratio [OR], 1.12; 95% confidence interval [CI], 1.032-1.215; P = .007), sepsis/septic shock (OR, 8.94, 95% CI: 2.283-35.002; P = .002), and requirement of vasopressors during ICU stay (OR, 16.839; 95% CI, 3.98-71.235; P = .0001), were independently associated with higher ICU mortality. Complete or partial remission of the underlying cancer on admission to ICU was the only protective factor, with a markedly decreased risk of death (OR, 0.113; 95% CI, 0.027-0.483; P = .003) (Table 4).

stay, status of the underlying disease, performance status, source of ICU admission, some morbidities and infections on ICU admission, APACHE II score, GCS, SOFA score, and some morbidities during ICU stay (septic shock, renal replacement therapy, requirement of vasopressors and steroid, etc) were significantly associated with ICU mortality in patients with hematological malignancies (Table 3). Only 2 parameters, namely, APACHE II score on admission (OR, 1.303; 95% CI, 1.054-1.609; P = .014) and requirement of vasopressors during ICU stay (OR, 140.635; 95% CI, 3.592-5505.482; P = .008), were independently associated with higher ICU mortality in multivariate analysis. Admission from emergency service was the only protective factor, with a markedly decreased risk of death (OR, 0.005; 95% CI, 0.000-0.691; P = .035) (Table 4). Length of hospital stay before ICU admission, length of ICU stay, cancer status, some morbidities and infections on ICU admission, some laboratory abnormalities (lactate dehydrogenase [LDH], inflammatory markers, etc), APACHE II score, GCS, SOFA score, and some morbidities during ICU stay (septic shock, mechanical ventilation, nosocomial infections, blood/blood product transfusion, requirement of vasopressor and steroid, etc) were significantly associated with ICU mortality in patients with solid tumors in univariate analysis (Table 3). Only 3 parameters, namely, SOFA score (OR, 1.831; 95% CI, 1.289-2.601; P = .001), LDH level on admission (OR, 1.002; 95% CI, 1-1.005; P = .028), and sepsis/septic shock during ICU stay (OR, 138.406; 95% CI, 12.534-1528.4; P = .0001), were independently associated with higher ICU mortality in multivariate analysis; complete or partial remission of the underlying cancer on ICU admission was the only protective factor, with a markedly decreased risk of death (OR, 0.026; 95% CI, 0.002-0.304; P = .004) (Table 4). Our results indicate that the severity of critical illness (APACHE II score), presence of organ dysfunctions (SOFA score), development of some morbidities during ICU stay (sepsis/septic shock in particular), and the status of the underlying cancer (limited or progressive) predicted the ICU mortality in our cancer patients.

3.5. Results of multivariate analysis in our study subgroups

5. Discussion

We also evaluated dependent and independent risk factors associated with ICU mortality in patients with hematological malignancies and patients with solid tumors separately. Univariate analysis revealed that length of hospital stay before ICU, length of ICU

There is growing number of ICU admissions of cancer patients in parallel to the increase in patients living with cancer. Patients with cancer are more severely ill than the patients without cancer, resulting in higher ICU and hospital mortality rates. Similarly, our

Please cite this article as: Aygencel G, et al, Prognostic factors in critically ill cancer patients admitted to the intensive care unit, J Crit Care (2014), http://dx.doi.org/10.1016/j.jcrc.2014.01.014

G. Aygencel et al. / Journal of Critical Care xxx (2014) xxx–xxx

5

Table 2 Some characteristics of the survivor and nonsurvivor cancer patients in our ICU Variable Age (y) Male sex Length of hospital stay before ICU admission (d) Length of ICU stay (d) Characteristics of the cancer on ICU admission Newly diagnosed Complete or partial remission Relapsed or progressive disease End stage Patient's preadmission performance status Good (ECOG 0-2) Poor (ECOG 3 and 4) Cancer treatment status Chemotherapy Radiation therapy Cancer therapy within the last month before ICU admission BMT recipient Allogeneic BMT Comorbidities Chronic obstructive pulmonary disease Diabetes mellitus Hypertension Cardiovascular diseases Cause of ICU admission Sepsis/septic shock Acute respiratory failure Acute renal failure Source of admission Emergency service Morbidities on ICU admission Sepsis/septic shock Renal dysfunction Neutropenia Thrombocytopenia Requirement of mechanical ventilation (IMV or NIMV) Hepatic dysfunction Requirement of vasopressors treatment ≥2 organ dysfunctions Suspected or culture-proven infection Bacterial infection Fungal infection Pulmonary infection Bloodstream infection Urinary tract infection On ICU admission APACHE II score GCS SOFA score Hemoglobin level (g/dL) Leukocytes (/mm3) Thrombocytes (/mm3) Sodium (Na, mEq/L) AST (IU/L) LDH (IU/L) C-reactive protein Procalcitonin Calcium (mg/dL) Total protein (g/dL) Albumin (g/dL) Creatinine (mg/dL) Glucose (mg/dL) Morbidities during ICU stay Sepsis/septic shock Neutropenia Hepatic dysfunction Renal dysfunction Renal replacement therapy Requirement of mechanical ventilation (IMV or NIMV) Chemotherapy in ICU Blood/blood product transfusion Central venous catheterization Arterial catheterization Tracheostomy Suspected or culture-proven infection

Survivor (n = 73)

Nonsurvivor (n = 89)

P

62 (49-70) 44 (60%) 2 (1-6) 4 (3-6)

61 51 8 7

(47.5-72) (57%) (2-22.5) (3-16)

24 (33%) 27 (37%) 16 (22%) 6 (8%)

30 10 39 10

(34%) (11%) (44%) (11%)

59 (81%) 14 (19%)

58 (65%) 31 (35%)

.034

49 (67%) 19 (26%) 33 (45%) 7 (10%) 6 (8%)

66 22 51 11 10

(74%) (25%) (57%) (12%) (11%)

.385 .858 .155 .624 .603

7 (10%) 17 (23%) 20 (27%) 9 (12%)

10 16 16 15

(11%) (18%) (18%) (17%)

.801 .437 .185 .508

40 (55%) 39 (53%) 19 (26%)

68 (76%) 64 (72%) 32 (36%)

.004 .021 .234

34 (47%)

21 (24%)

.003

37 (51%) 32 (44%) 17 (23%) 29 (40%) 21 (29%) 22 (30%) 19 (26%) 24 (33%) 61 (84%) 21 (29%) 4 (5%) 41 (56%) 7 (10%) 10 (14%)

58 43 29 58 47 34 35 60 86 48 17 61 21 27

(65%) (48%) (33%) (65%) (53%) (38%) (39%) (67%) (97%) (54%) (19%) (69%) (24%) (30%)

.078 .636 .22 .002 .002 .321 .094 .0001 .006 .001 .01 .141 .022 .014

(22-34) (7-15) (7-13) (8-10) (2310-16 775) (29 950-160 750) (134-142) (22.3-75) (268-909) (109-284) (1-26) (6.9-8.1) (4.4-5.8) (2.1-2.8) (0.7-2.48) (123-201)

.0001 .0001 .0001 .239 .268 .001 .036 .035 .007 .026 .003 .03 .029 .003 .611 .423

(74%) (24%) (26%) (48%) (45%) (84%) (7%) (75%) (78%) (91%) (19%) (56%)

.0001 .011 .0001 .0001 .0001 .0001

18 (14-22.5) 15 (14-15) 6 (3-8) 9.3 (8.05-11.05) 9140 (5275-14 100) 130 000 (59 600-225 500) 136 (132-140) 26 (17.3-45) 312 (210.5-507) 121.5 (51-218) 1.45 (0.28-4.9) 8 (7.25-8.5) 5.5 (4.8-6) 2.7 (2.4-3) 1 (0.7-1.9) 135 (118.3-172) 7 (10%) 6 (8%) 1 (1%) 5 (7%) 13 (18%) 34 (47%) 4 (5%) 29 (40%) 19 (26%) 44 (60%) 1 (1%) 12 (16%)

27.5 12.5 9 9.1 7800 66 000 137 35 475 168 4.5 7.7 5.1 2.4 1.15 138 66 21 23 43 40 75 6 67 69 81 17 50

.814 .75 .0001 .002 1 .0001 .004 .603

1 .0001 .0001 .0001 .0001 .0001 (continued on next page)

Please cite this article as: Aygencel G, et al, Prognostic factors in critically ill cancer patients admitted to the intensive care unit, J Crit Care (2014), http://dx.doi.org/10.1016/j.jcrc.2014.01.014

6

G. Aygencel et al. / Journal of Critical Care xxx (2014) xxx–xxx

Table 2 (continued) Variable

Survivor (n = 73)

Bacterial infection Fungal infection Pulmonary infection Urinary tract infect Bloodstream infection Steroid use Vasopressor use Types of cancer Solid tumors Hematological malignancies

Nonsurvivor (n = 89)

10 (14%) 2 (3%) 11 (15%) 3 (4%) 1 (1%) 38 (52%) 15 (21%)

47 18 45 18 19 81 84

(53%) (20%) (51%) (20%) (21%) (91%) (94%)

48 (66%) 25 (34%)

56 (63%) 33 (37%)

P .0001 .001 .0001 .002 .0001 .0001 .0001

Values are presented as median (interquartile range) or n (%). IMV indicates invasive mechanical ventilation; NIMV, noninvasive mechanical ventilation.

ICU mortality rate in cancer patients was 55% and higher than the overall mortality of 32%. Previous studies have reported ICU mortality between 10% and 70% in cancer patients. This large interstudy variation in the mortality rates might be explained by the variation in the composition of underlying disease, admission and discharge criteria, treatment decisions, and the implementation of end-of-life decisions [1–8]. The poor prognosis of patients with hematological malignancies who require ICU admission has been well documented, with ICU mortality rates up to 60%, increasing up to 80% when mechanical ventilation is required [9,10], whereas surgical admission is reported to be associated with better outcome in patients with solid tumors [1,11]. Although some studies claim higher ICU mortality rate in hematological malignancies [9,10,12], we did not confirm this result in our study. Intensive care unit mortality rates were similar in patients with solid tumors and hematological malignancies (53.8% vs 57%; P = .744), which might be explained by some features of the presented cohort. First of all, lymphomas were categorized within the solid tumors. Colon and breast cancers, which have a relatively better prognosis, consisted the majority of solid tumor group. Mortality rates could have been different when the lymphomas are removed from this subcategory. Another explanation is the high rate of patients with advanced cancer, with the rate of 43% in the solid tumor group, which might have caused overestimation of the mortality rate in this group of patients. Site and number of metastases, which have a significant impact on mortality in advanced cancer patients, were not recorded in our cohort and might be a limitation of the current study. Intensive care mortality could be significantly improved in critically ill cancer patients with rational patient selection during admission. As a general rule, patients with good performance status, who are at the initial phase of their malignant disease with available life-extending treatment options, should routinely be admitted to the ICU. Reliable objective criteria in cancer patients that can be assessed rapidly at the bedside to decide ICU admission are often lacking. In most of the studies, classic criteria such as the characteristics of the underlying malignancy are no longer reliable for assessing potential benefits of ICU admission [1–8]. The clinical judgment of physicians is inaccurate in up to 25% of cases according to a previous French study [13]. Severity of the acute illness on ICU admission is the main determinant of short-term mortality in noncancer critically ill patients. Severity scores and organ failure scores can be useful in predicting the outcome of cancer patients admitted to the ICU and thus help the medical decision making [14–17]. We demonstrated that one of the severity scores (APACHE II) in all cancer patient cohort and organ failure score (SOFA) in patients with solid tumors were independent risk factors for ICU mortality. Some studies in cancer patients have shown severity and organ failure scores (such as APACHE II, Simplified Acute Physiology Score II, SOFA, etc) on admission to be a valuable prognostic factor for ICU mortality, whereas some other studies have not [14–17]. Some studies have reported that daily assessments of organ failures and the course of the organ failure with time during the ICU stay showed good discrimination in cancer patients [14,16]. Up to 80% of cancer patients are admitted to the ICU because of sepsis/septic shock or acute respiratory failure. In other words, severe

infections and organ failures are the main reasons for ICU admission in cancer patients [3–5,8]. Similarly, sepsis/septic shock (67%) and respiratory failure (64%) were the leading causes of ICU admission, whereas respiratory system (63%) was the leading site of infections in our study. Acute kidney injury is known to be among the frequent organ failures on ICU admission and approximately one third of cancer patients have acute kidney injury, half of whom require renal replacement therapy [11,12,18,19]. Nevertheless, 31.5% of our cancer patients were admitted to our ICU with acute renal failure, and 32.7% of them required renal replacement therapy during ICU stay. The main objectives of the current study were to determine the outcome of our cancer patients admitted to ICU over a 3.5-year period and to identify factors influencing ICU mortality. The ICU mortality rate was 55% and was negatively influenced by high APACHE II score on admission, vasopressor use, and development of septic shock during ICU stay in multivariate analysis. We also confirmed that presence of organ failure on ICU admission predicts poor ICU outcome in patients with solid tumors. Achievement of complete or partial remission had significantly better ICU outcome in our cohort. Although several previous studies deny that response to chemotherapy, stage of malignancy, and other characteristics of the cancer have little or no impact on short-term survival, some other reports, in accordance with ours, claim that some characteristics of the underlying cancer, such as its extent or phase influenced the ICU mortality [4,6,9,20]. Some studies have reported that patient level of functioning before the admission to ICU influenced the ICU mortality and morbidities [3,5,20,21]. We confirmed these results in our study. Poor performance status predicted poor ICU outcome; however, this result was not an independent risk factor for ICU mortality in multivariate analysis. Different studies have showed various independent risk factors for ICU mortality in cancer patients. Some of the previously defined risk factors are older age, male sex, progressive underlying malignancy, respiratory insufficiency, high severity of illness scores (such as APACHE II and III and Simplified Acute Physiology Score II), the number and the severity of organ failures (especially multiorgan failure), requirement for invasive mechanical ventilation, renal dysfunction and requirement of renal replacement therapy, hepatic dysfunction, neutropenia, allogenic HSCT, presence of graft-versushost disease, sepsis on admission and during ICU stay, infections (especially invasive fungal infection or bacteremia), use of vasopressors, length of hospital stay before ICU admission, source of admission, performance status, administration of chemotherapy in ICU, and some laboratory abnormalities (including anemia, thrombocytopenia, neutropenia, high bilirubin and creatinine levels, low albumin levels, etc) [1–29]. Several variables were shown to be related with ICU mortality in univariate analysis of our data (length of hospital stay before ICU admission, length of ICU stay, relapsed or progressive cancer, admitted with sepsis/septic shock, illness severity score and organ failure score, thrombocytopenia, some laboratory abnormalities in aspartate aminotransferase (AST), LDH, albumin, inflammatory markers, organ dysfunctions during ICU stay, infection during ICU

Please cite this article as: Aygencel G, et al, Prognostic factors in critically ill cancer patients admitted to the intensive care unit, J Crit Care (2014), http://dx.doi.org/10.1016/j.jcrc.2014.01.014

G. Aygencel et al. / Journal of Critical Care xxx (2014) xxx–xxx

7

Table 3 Some characteristics of the survivor and nonsurvivor hematological malignancies and solid tumors patients in our ICU Variable

Patients with hematological malignancies Survivor (n = 25)

Age (y) Male sex Length of hospital stay before ICU admission (d) Length of ICU stay (d) Characteristics of the cancer on ICU admission Newly diagnosed Complete or partial remission Relapsed or progressive disease End stage Patient's preadmission performance status Good (ECOG 0-2) Poor (ECOG 3 and 4) Cancer treatment status Chemotherapy Radiation therapy Cancer therapy within the last month before ICU admission BMT recipient Allogeneic BMT Comorbidities Chronic obstructive pulmonary disease Diabetes mellitus Hypertension Cardiovascular diseases Cause of ICU admission Sepsis/septic shock Acute respiratory failure Acute renal failure Neurologic disorders Source of admission Emergency service Morbidities on ICU admission Sepsis/septic shock Renal dysfunction Neutropenia Thrombocytopenia Requirement of mechanical ventilation (IMV or NIMV) Hepatic dysfunction Requirement of vasopressors treatment ≥2 organ dysfunctions Suspected or culture-proven infection Bacterial infection Fungal infection Pulmonary infection Bloodstream infection Urinary tract infection On ICU admission APACHE II score GCS SOFA score Hemoglobin level (g/dL) Leukocytes (/mm3) Thrombocytes (/mm3) Sodium (Na, mEq/L) AST (IU/L) LDH (IU/L) C-reactive protein Procalcitonin Calcium (mg/dL) Total protein (g/dL) Albumin (g/dL) Creatinine (mg/dL) Glucose (mg/dL) Morbidities during ICU stay Sepsis/septic shock Neutropenia Hepatic dysfunction Renal dysfunction Renal replacement therapy

Nonsurvivor (n = 33)

Patients with solid tumors P

Survivor (n = 48)

53 (37.5-63.5) 20 (80%) 3 (1-9.5) 3 (3-6)

52 20 19 7

(28-61) (61%) (5.5-30) (4-12)

.55 .155 .001 .027

10 (40%) 9 (36%) 4 (16%) 2 (8%)

11 1 21 0

(33%) (3%) (64%) (0%)

.783 .001 .0001 .063

64.5 (57-74.8) 24 (50%) 1.5 (0-4.8) 4.5 (3-6)

Nonsurvivor (n = 56)

P

66 (57-74.8) 31 (55%) 6 (1-16) 7 (3-16.8)

.904 .694 .001 .025

14 (29%) 18 (38%) 12 (25%) 4 (8%)

19 (34%) 9 (16%) 18 (32%) 10 (18%)

.675 .015 .516 .249

4 (16%)

15 (45%)

.024

10 (21%)

16 (29%)

.496

20 (80%) 3 (12%) 16 (64%)

25 (76%) 4 (12%) 23 (70%)

.76 .989 .779

29 (60%) 16 (33%) 17 (35%)

41 (73%) 18 (32%) 28 (50%)

.209 1 .166

6 (24%) 5 (20%)

8 (24%) 7 (21%)

1 (2.1%) 1 (2%)

3 (5.4%) 3 (5%)

.374 .374

0 (0%) 3 (12%) 1 (4%) 3 (12%)

2 1 3 2

(6%) (3%) (9%) (6.1%)

.129 .179 .436 .427

7 (15%) 14 (29%) 19 (40%) 6 (13%)

8 (14%) 15 (27%) 13 (23%) 13 (23%)

1 .829 .09 .206

16 (64%) 17 (68%) 4 (16%) 3 (12%)

26 23 12 15

(79%) (70%) (36%) (45%)

.247

24 (50%) 22 (46%) 15 (31%)

42 (75%) 41 (73%) 20 (36%)

.014 .005 .681

.002

26 (54%)

20 (36%)

.078

8 (32%)

1 1

1 .138 .009

1 (3%)

15 (60%) 9 (36%) 14 (56%) 17 (68%) 6 (24%)

23 15 23 31 23

(70%) (45%) (70%) (94%) (70%)

.578 .592 .408 .009 .001

22 (46%) 23 (48%) 3 (6%) 12 (25%) 15 (31%)

35 (63%) 28 (50%) 6 (11%) 27 (48%) 24 (43%)

.114 .847 .414 .016 .310

6 (24%) 8 (32%) 15 (60%) 23 (92%) 9 (36%) 3 (12%) 17 (68%) 5 (20%) 2 (8%)

11 15 26 32 23 10 22 12 7

(33%) (45%) (79%) (97%) (70%) (30%) (67%) (36%) (21%)

.564 .417 .151 .572 .016 .122

16 (33%) 11 (23%) 9 (19%) 38 (79%) 12 (25%) 1 (2%) 24 (50%) 2 (4%) 8 (17%)

23 (41%) 20 (36%) 34 (61%) 54 (96%) 25 (45%) 7 (13%) 39 (70%) 9 (16%) 20 (36%)

.543 .198 .0001 .011 .042 .034 .047 .060 .045

18 (12.5-21.5) 15 (14.5-15) 8 (5-9) 8.4 (6.9-9.75) 8800 (1625-20 300) 55200 (28500-135450) 136 (134.5-138.5) 32 (16-43) 334 (201-682) 154 (92-266.3) 3.7 (0.6-23) 7.5 (7.2-8) 5.5 (4.7-6.4) 2.5 (2.3-2.9) 0.9 (0.65-1.65) 142 (120-177) 3 (12%) 5 (20%) 0 (0%) 2 (8%) 3 (12%)

30 (24-38.5) 10 (4.5-14.5) 10.5(7.3-13.8) 8.2 (6.5-9.4) 2960 (695-16 825) 37650 (18000-65750) 137.5(135-141.5) 29(16.3-69.8) 418.5 (263-1026) 205 (106-327) 15.5 (0.9-73.5) 7.4 (6.63-7.78) 4.75 (4.3-5.6) 2.4 (2.2-2.98) 0.9 (0.6-2.08) 138 (117-200) 23 16 12 22 19

(70%) (46%) (36%) (67%) (58%)

1 .247 .155 .0001 .0001 .001 .568 .143 .038 .131 .711 .446 .39 .212 .363 .032 .378 .917 .792 .0001 .031 .0001 .0001 .0001

18.5 (14.25-23.75) 15 (14-15) 5 (3-7) 10 (8.45-11.65) 9930 (6242-13 575) 166500 (97850-249750) 135 (131.25-140) 24 (17-45) 308 (223.5-417) 89 (36.25-198.25) 0.9 (0.245-3.5) 8.15 (7.5-8.8) 5.6 (4.9-6) 2.8 (2.4-3.1) 1.2 (0.7-2.125) 134 (118-172) 4 (8%) 1 (2%) 1 (2%) 3 (6%) 10 (21%)

25 (21-32) 13 (7-15) 9 (6-12) 9.5 (8.53-10.7) 10 300 (4300-16 775) 114500 (35375-210725) 136.5 (133-143) 38 (24.25-80.25) 501 (265.5-875.5) 149.5 (111-223.5) 3.9 (1.15-20) 7.8 (7.125-8.3) 5.4 (4.6-5.83) 2.4 (2-2.78) 1.2 (0.8-2.6) 139.5 (123-201) 43 (77%) 5 (9%) 11 (20%) 21 (38%) 21 (38%)

.0001 .0001 .0001 .215 .907 .016 .113 .019 .004 .026 .003 .018 .28 .002 .509 .455 .0001 .117 .005 .0001 .086

(continued on next page)

Please cite this article as: Aygencel G, et al, Prognostic factors in critically ill cancer patients admitted to the intensive care unit, J Crit Care (2014), http://dx.doi.org/10.1016/j.jcrc.2014.01.014

8

G. Aygencel et al. / Journal of Critical Care xxx (2014) xxx–xxx

Table 3 (continued) Variable

Patients with hematological malignancies Survivor (n = 25)

Requirement of mechanical ventilation (IMV or NIMV) Chemotherapy in ICU Blood/blood product transfusion Central venous catheterization Arterial catheterization Tracheostomy Suspected or culture-proven infection Bacterial infection Fungal infection Pulmonary infection Urinary tract infection Bloodstream infection Steroid use Vasopressor use

Nonsurvivor (n = 33)

Patients with solid tumors P

Survivor (n = 48)

Nonsurvivor (n = 56)

P

13 (52%)

29 (88%)

.003

21 (44%)

46 (82%)

.0001

4 (16%) 15 (60%) 6 (24%) 19 (76%) 1 (4%) 7 (28%) 6 (24%) 0 (0%) 6 (24%) 2 (8%) 1 (4%) 16 (64%) 6 (24%)

6 (18%) 32 (97%) 25 (76%) 30 (91%) 7 (21%) 18 (55%) 18 (55%) 5 (15%) 15 (45%) 3 (9%) 9 (27%) 31 (94%) 31 (94%)

.827 .0001 .0001 .121 .045 .062 .031 .014 .107 .883 .012 .006 .0001

14 (29%) 13 (27%) 25 (52%) 0 (0%) 5 (10%) 4 (8%) 2 (4%) 5 (10%) 1 (2%) 0 (0%) 22 (46%) 9 (19%)

35 44 51 10 32 29 13 30 15 10 50 53

.001 .0001 .0001 .0001 .0001 .0001 .01 .0001 .001 .0001 .0001 .0001

(63%) (79%) (91%) (18%) (57%) (52%) (23%) (54%) (27%) (18%) (89%) (95%)

Values are presented as median (interquartile range) or n (%).

stay, sepsis/septic shock during ICU stay, etc). However, only 3 independent parameters were identified to be associated with higher ICU mortality rate in multivariate analysis. These were APACHE II score, requirement of vasopressors, and sepsis/septic shock during ICU stay in all cancer patients. In addition, complete or partial remission of the underlying cancer was identified as a good prognostic factor. Although some studies evaluated the independent risk factors for mortality in cancer groups in general, some other studies evaluated risk factors for cancer subgroups specifically [9–12,19– 21]. We, therefore, analyzed our data in all the cohort and categorized as subgroup of patients with solid and hematological malignancies. Acute Physiology and Chronic Health Evaluation II score and requirement of vasopressors during ICU stay were independent risk factors for mortality in patients with hematological malignancies, whereas SOFA score, LDH level, and sepsis/septic shock during ICU stay were the independent risk factors for ICU mortality in solid tumor patients. Admission from emergency department for patients with hematological malignancies and complete or partial remission status for patients with solid tumors were good prognostic factors. Delayed ICU admission is a very well-known risk factor for higher mortality rates in noncancer patients. Furthermore, some studies have reported that late ICU admission of cancer patients may be associated with poor survival [21,28,29]. We believe that some factors such as renal dysfunction or respiratory failure could have been modified or managed easily by earlier admission to ICU, whereas some other factors such as progressive underlying cancer are among the ones that are unmodifiable. We, therefore, suggest that at least the subgroup of

cancer patients already inpatient could be saved by earlier evaluation and intervention by the ICU team when they have less severe disease. Earlier ICU admission rather than refusal due to poor outcome during admission seems to be a more promising and rational approach, which could be investigated in prospective trials. The retrospective nature and being a single-center study are the major limitations of the current report. Furthermore, the relatively small sample size leads us to refrain from drawing solid conclusions. It is also rather difficult to compare crude mortality in different studies because of variations in the underlying disease composition, ICU admission, and discharge criteria and treatment decision criteria. Our local treatment decisions may not necessarily be applicable to all institutions. Putting admission criteria for cancer patients at the beginning, categorizing lymphomas in solid tumor group, absence of site, and number of metastases in cancer patients were the other limitations of our study. Finally, we did not collect data on long-term outcomes or quality of life after discharge from ICU in our cohort, which might have increased the impact of the study. In conclusion, our study indicates that a reasonable number of cancer patients might survive their ICU stay. We attribute this 45% survival rate to better selection during ICU admission. Acute Physiology and Chronic Health Evaluation II score, the status of underlying cancer, and requirement of vasopressors can be used as selection criteria on admission to ICU. Earlier ICU intervention and admission of the at-risk patients might further improve the ICU outcome of cancer patients. The role of earlier ICU evaluation in improving ICU outcome seems to be an attractive topic for further prospective multicenter trials.

Table 4 Independent risk factors for ICU mortality in cancer patients and their subgroups in multivariate analysis

References

Variable

OR (95% CI)

All cancer patients APACHE II score 1.12 (1.032-1.215) Sepsis/septic shock during ICU stay 8.94 (2.283-35.002) Vasopressor use during ICU stay 16.839 (3.98-71.235) Complete or partial remission of the 0.113 (0.027-0.483) underlying cancer on ICU admission Hematological malignancies APACHE II score 1.303 (1.054-1.609) Vasopressor use during ICU stay 140.635 (3.592-5505.482) Admission from emergency services 0.005 (0.000-0.691) Solid tumors SOFA score 1.831 (1.289-2.601) LDH level on admission 1.002 (1-1.005) Sepsis/septic shock during ICU stay 138.406 (12.534-1528.4) Complete or partial remission of the 0.026 (0.002-0.304) underlying cancer on ICU admission

P .007 .002 .0001 .003

.014 .008 .035 .001 .028 .0001 .004

[1] Staudinger T, Stoiser B, Müllner M, Locker GJ, Laczika K, Knapp S, et al. Outcome and prognostic factors in critically ill cancer patients admitted to the intensive care unit. Crit Care Med 2000;28(5):1322–8. [2] Mokart D, Etienne A, Esterni B, Brun JP, Chow-Chine L, Sannini A, et al. Critically ill cancer patients in the intensive care unit: short-term outcome and 1-year mortality. Acta Anaesthesiol Scand 2012;56:178–89. [3] Azoulay E, Afessa B. The intensive care support of patients with malignancy: do everything that can be done. Intensive Care Med 2006;32:3–5. [4] Taccone FS, Artigas AA, Sprung CL, Moreno R, Sakr Y, Sakr Y, Vincent JL, et al. Characteristics and outcomes of cancer patients in European ICUs. Crit Care 2009;13(1):R15. [5] Azoulay E, Soares M, Darmon M, Benoit D, Pastores S, Pastores S, Afessa B, et al. Intensive care of the cancer patient: recent achievements and remaining challenges. Ann Intensive Care 2011;1(1):5. [6] Mc Grath S, Chatterjee F, Whiteley C, Ostermann M. ICU and 6-month outcome of oncology patients in the intensive care unit. Q J Med 2010;103:397–403. [7] Darmon M, Azoulay E. Critical care management of cancer patients: cause for optimism and need for objectivity. Curr Opin Oncol 2009;21:318–26. [8] Kress JP, Christenson J, Pohlman AS, Linkin DR, Hall JB. Outcomes of critically ill cancer patients in a university hospital setting. Am J Respir Crit Care Med 1999;160:1957–61.

Please cite this article as: Aygencel G, et al, Prognostic factors in critically ill cancer patients admitted to the intensive care unit, J Crit Care (2014), http://dx.doi.org/10.1016/j.jcrc.2014.01.014

G. Aygencel et al. / Journal of Critical Care xxx (2014) xxx–xxx [9] Yeo CD, Kim JW, Kim SC, Kim YK, Kim KH, et al. Prognostic factors in critically ill patients with hematologic malignancies admitted to the intensive care unit. J Crit Care 2012;27:739.e1-6. [10] Ferra C, Marcos P, Misis M, Morgades M, Bordeje ML, Oriol A, et al. Outcome and prognostic factors in patients with hematologic malignancies admitted to the intensive care unit: a single center experience. Int J Hematol 2007;85(3):195–202. [11] Azoulay E, Moreau D, Alberti C, Leleu G, Adrie C, Barboteu M, et al. Predictors of short-term mortality in critically ill patients with solid malignancies. Intensive Care Med 2000;26:1817–23. [12] Benoit DD, Vandewoude KH, Decruyenaere JM, Hoste EA, Colardyn FA. Outcome and early prognostic indicators in patients with a hematologic malignancy admitted to the intensive care unit for a life-threatening complication. Crit Care Med 2003;31(1):104–12. [13] Thiery G, Azoulay E, Darmon M, Ciroldi M, De Miranda S, Levy V, et al. Outcome of cancer patients considered for intensive care unit admission. A hospital-wide prospective study. J Clin Oncol 2005;23:4406–13. [14] Lamia B, Hellot MF, Girault C, Tamion F, Dachraoui F, Lenain P, et al. Changes in severity and organ failure scores as prognostic factors in onco-hematological malignancy patients admitted to the ICU. Intensive Care Med 2006;32(10):1560–8. [15] Sculier JP, Paesmons M, Markiewicz E, Berghmans T. Scoring systems in cancer patients admitted for an acute complication in a medical intensive care unit. Crit Care Med 2000;28(8):2786–92. [16] Soares M, Fontes F, Dantas J, Gadelha D, Cariello P, Nardes F, et al. Performance of six severity-of-illness scores in cancer patients requiring admission to the intensive care unit: a prospective observational study. Crit Care 2004;8:R194–203. [17] Schellongowski P, Benesch M, Lang T, Traunmuller F, Zauner C, Laczika K, et al. Comparison of three severity scores for critically ill cancer patients. Intensive Care Med 2004;30:430–6. [18] Soares M, Salluh JI, Carvalho MS, Darmon M, Rocco JR. Prognosis of critically ill patients with cancer and acute renal dysfunction. J Clin Oncol 2006;24:4003–10.

9

[19] Bird GT, Farquhar-Smith P, Wigmore T, Potter M, Gruber PC. Outcomes and prognostic factors in patients with haematological malignancy admitted to a specialist cancer intensive care unit: a 5 yr study. Br J Anaesth 2012;108(3):452–9. [20] Soares M, Salluh JI, Spector N, Rocco JR. Characteristics and outcomes of cancer patients requiring mechanical ventilatory support for N24 hrs. Crit Care Med 2005;3:520–6. [21] Moran JL, Solomon PJ, Williams PJ. Assessment of outcome over a 10-year period of patients admitted to a multidisciplinary adult intensive care unit with haematological and solid tumors. Anaesth Intensive Care 2005;33:26–35. [22] Markou N, Demopoulou E, Myrianthefs P. The critically ill patient with cancerindications for intensive care unit admission and outcomes. J BUON 2008;13(4):469–78. [23] Namendys-Silva SA, Texcocano-Becerra J, Herrera-Gómez A. Prognostic factors in critically ill patients with solid tumor admitted to an oncological intensive care unit. Anaesth Intensive Care 2010;38:317–24. [24] Soares M, Carvalho MS, Salluh JI, Ferreira CG, Luizz RR, Rocco JR, et al. Effect of age on survival of critically ill patients with cancer. Crit Care Med 2006;34:715–21. [25] Benoit DD, Depuydt PO, Vandewoude KH, Offner FC, Boterberg T, De Cock CA, et al. Outcome in severely ill patients with hematological malignancies who received intravenous chemotherapy in the intensive care unit. Intensive Care Med 2006;32:93–9. [26] Massion PB, Dive AM, Doyen C, Bulpa P, Jamart J, Bosly A, et al. Prognosis of hematologic malignancies does not predict intensive care unit mortality. Crit Care Med 2002;30:2260–70. [27] Chalfin DB, Carlon GC. Age and utilization of intensive care unit resources of critically ill cancer patients. Crit Care Med 1990;18:694–8. [28] Soares M, Caruso P, Silva E, Teles JM, Lobo SM, Brazilian Research in Intensive Care Network (BRICNet), Friedman G, et al. Characteristics and outcomes of patients with cancer requiring admission to intensive care units: a prospective multicenter study. Crit Care Med 2010;38(1):9–15. [29] Mokart D, Lambert J, Schnell D, Fouche L, Rabbat A, Kouatchet A, et al. Delayed intensive care unit admission is associated with increased mortality in patients with cancer with acute respiratory failure. Leuk Lymphoma 2013;54(8):1724–9.

Please cite this article as: Aygencel G, et al, Prognostic factors in critically ill cancer patients admitted to the intensive care unit, J Crit Care (2014), http://dx.doi.org/10.1016/j.jcrc.2014.01.014