- Email: [email protected]
Critically ill patients with cancer and sepsis: Clinical course and prognostic factors
Journal of Critical Care (2012) 27, 301–307
Critically ill patients with cancer and sepsis: Clinical course and prognostic factors☆,☆☆,★ Maíra M. Rosolem MD a,b,1 , Lígia S.C.F. Rabello MD a,b,1 , Thiago Lisboa MD, PhD c , Pedro Caruso MD, PhD d , Ramon T. Costa MD d , Juliana V.R. Leal MD a , Jorge I.F. Salluh MD, PhD b,e , Márcio Soares MD, PhD b,e,⁎ a
ICU, Instituto Nacional de Câncer, Rio de Janeiro, Brazil 20230-130 Postgraduate Program, Instituto Nacional de Câncer, Rio de Janeiro, Brazil 20231-050 c ICU, Hospital de Clínicas, Universidade Federal do Rio Grande do Sul, Porto Alegre, Brazil 90035-903 d ICU, Hospital A. C. Camargo, São Paulo, Brazil 01509-010 e D'Or Institute for Research and Education, Rio de Janeiro RJ CEP 22281-100, Brazil b
Keywords: Cancer; Intensive care unit; Mortality; Sepsis; Outcome
Abstract Purpose: The purposes of this study were to evaluate the clinical course and to identify independent predictors of mortality in patients with cancer with sepsis. Materials and Methods: This is a secondary analysis of a prospective cohort study conducted at an oncological medical-surgical intensive care unit. Logistic regression was used to identify predictors of hospital mortality. Results: A total of 563 patients (77% solid tumor, 23% hematologic malignancies) were included over a 55month period. The most frequent sites of infection were the lung, abdomen, and urinary tract; 91% patients had severe sepsis/septic shock. Gram-negative bacteria were responsible for more than half of the episodes of infection; 38% of patients had polymicrobial infections. Intensive care unit, hospital, and 6-month mortality rates were 51%, 65%, and 72%, respectively. In multivariate analyses, sepsis in the context of medical complications; active disease; compromised performance status; presence of 3 to 4 systemic inflammatory response syndrome criteria; and the presence of respiratory, renal, and cardiovascular failures were associated with increased mortality. Adjusting for other covariates, patients with non–urinary tract infections, mostly represented by patients with pneumonia and abdominal infections, had worse outcomes. Conclusions: Sepsis remains a frequent complication in patients with cancer and associated with high mortality. Our results can be of help to assist intensivists in clinical decisions and to improve characterization and risk stratification in these patients. © 2012 Elsevier Inc. All rights reserved.
☆
This work was performed at the intensive care unit of the Instituto Nacional de Câncer, Brazil. Financial support: Dr Soares is supported in part by individual research grant from CNPq. ★ Conflicts of interest: All authors have no conflict of interest to declare. ⁎ Corresponding author. D'Or Institute for Research and Education, Rio de Janeiro RJ CEP 22281-100, Brazil. Tel.: +55 21 2538 3541; fax: +55 21 2294 8620. E-mail addresses: [email protected], [email protected] (M. Soares). 1 Drs Rosolem and Rabello contributed equally for the study. ☆☆
0883-9441/$ – see front matter © 2012 Elsevier Inc. All rights reserved. doi:10.1016/j.jcrc.2011.06.014
302
1. Introduction Patients with cancer are at increased risk for sepsis as a consequence of multiple mechanisms of immunosuppression imposed by the disease itself and aggressive treatments, including combined regimens of chemotherapy and radiation therapy, high dose of steroids, and hematopoietic stem cell transplantation [1-3]. In large databases, up to 1 in 5 patients admitted to intensive care units (ICUs) with sepsis have cancer [4-6], and along this line, sepsis is a leading reason for ICU admission in patients with cancer [7,8]. Moreover, despite the observation of improved survival rates over the last years, sepsis in patients with cancer remains associated with high morbidity, mortality, costs, and use of ICU resources, and information on this topic is limited [9-16]. Refinements in the information about the epidemiology, outcomes, and risk stratification of these patients are essential for future research to assist clinicians in clinical decision making, to allow better resource allocation, and for counseling patients and families. The aims of the present study were to evaluate the characteristics, clinical course, and outcomes and to identify independent predictors of mortality in a large cohort of patients with cancer admitted to the ICU with sepsis.
2. Patients and methods 2.1. Design and setting This was a secondary analysis of a prospective cohort study performed from January 2003 to July 2007 at the Instituto Nacional de Câncer, Rio de Janeiro, Brazil. The ICU is a 10-bed medical-surgical unit specialized in the care of patients with cancer [17]. As a rule, only patients with a chance of survival improvement are considered for ICU admission. However, some patients may occasionally be admitted while a full assessment of their cancer extent and therapeutic options is still ongoing. Decisions to withstand or withhold life-sustaining therapies are taken in patients who do not recover from the acute illness despite ICU care or if treatment for the cure or control of cancer cannot be given. This study was supported by institutional funds and did not interfere with clinical decisions related with patient care. The Ethics Committee of Instituto Nacional de Câncer approved the study (No. 10/2003), and the need for informed consent was waived.
2.2. Selection of participants, data collection, and definitions During the study period, every adult patient (≥18 years old) with a definite diagnosis of cancer and presenting with sepsis at ICU admission was evaluated. Patients in complete
M.M. Rosolem et al. remission longer than 5 years, those with an ICU stay of less than 24 hours, and readmissions were not considered. Infection was defined as the presence of a pathogenic microorganism in a sterile milieu (such as blood or cerebrospinal fluid) and/or clinically suspected infection that justified the administration of antibiotics [6,8]. Sepsis was diagnosed according to the consensus definitions [18]. Severe sepsis was defined as sepsis complicated by organ dysfunction, and septic shock was defined as severe sepsis with arterial hypotension (systolic pressure b90 mm Hg or mean arterial pressure b60 mm Hg or reduction in systolic pressure N40 mm Hg from baseline). Data related to patient's management were not collected. Demographic, clinical, and laboratory data were collected using standardized case report forms during the first day of ICU including main diagnosis for admission, the Simplified Acute Physiology Score (SAPS) II [19], the Sequential Organ Failure Assessment (SOFA) score [20], comorbidities, performance status (PS) (Eastern Cooperative Oncology Group scale) [21], and cancer- and treatment-related data. Cancer status was classified into 3 categories: remission/controlled (ie, patients in cancer remission or control who have undergone previous treatments, without evidence of recurrence according to the attending oncologist/ hematologist), recent diagnosis (patients with active disease diagnosed within the last 3 months with need for first-line anticancer treatment), and relapse/recurrence (patients with active disease with relapse or recurrence). For the purpose of the present study, individual organ failures were diagnosed in case of SOFA score of 2 points or greater in each domain [8]. In addition, patients receiving dialysis in the context of acute kidney injury and invasive mechanical ventilation on the first day of ICU were considered as having renal and respiratory failures regardless of the SOFA score, respectively. Comorbidities were evaluated using the Adult Comorbidity Evaluation 27 [22]. Neutropenia was defined as a neutrophil count less than 500/mm3. Vital status at hospital discharge was the outcome of interest, but the ICU and 6-month mortality rates were also assessed.
2.3. Data processing and statistical analysis Data entry was performed by a single investigator (MS), and consistency was assessed by a rechecking procedure of a 10% random sample of patients. Data were screened in detail by 4 investigators (MS, LSFR, MMR, and TL) for missing information and implausible and outlying values. Continuous variables were reported as mean ± SD or median (25%-75% interquartile range). Univariate and multivariate logistic regressions were used to identify factors associated with hospital mortality [23]. Linearity between each continuous variable and the dependent variable was demonstrated using locally weighted scatterplot smoothing [23]. In case of nonlinearity, the variable was stratified according to the inflection points and clinical
Patients with cancer and sepsis significance. For categorical variables with multiple levels, the reference level was attributed to the one with the lowest probability of the dependent variable. Variables yielding P b .2 by univariate analysis and those considered clinically relevant according to the current literature were entered into the multivariate analyses. Results were summarized as odds ratios and respective 95% confidence intervals (CIs). The statistical model was internally validated using the technique of bootstrap resampling. This technique is efficient and provides nearly unbiased estimates of the predictive accuracy of the model. We repeated the entire modeling process to validate the final model (1000 replications) and to adjust the estimated performance and regression coefficients. Possible interactions were tested. The area under the receiver operating characteristic curve (AROC) was used to assess the models' discrimination; an AROC of 1.0 denotes perfect, whereas a value close to 0.50 indicates no apparent accuracy [24]. The Hosmer-Lemeshow goodness-of-fit test was used to evaluate agreement between the observed and expected results across all strata of probabilities of the outcome of interest (calibration) [23]. With this test, P N .05 indicate a good fit for the model. Two-tailed P b .05 was considered statistically significant.
3. Results 3.1. Characterization of the study population Excluding scheduled surgical and those with ICU admission less than 24 hours, a total of 1332 patients were admitted to our ICU during the study period. Of them, 563 (42%) had sepsis fulfilling the eligibility criteria and constituted the study population. The sources of ICU admission were the operating room (27%), emergency department (23%), and wards (50%). There were 436 patients (77%) with solid tumors, and 127 patients (23%) had hematologic malignancies. The patients' main characteristics are depicted in Table 1. The primary sites of solid tumors were gastrointestinal (35%), head and neck (13%), urogenital (6%), lung (6%), breast (5%), and others (12%). The main hematologic malignancies were non-Hodgkin lymphomas (14%), leukemias (6%), multiple myeloma (2%), and others (1%). Previous anticancer treatments included surgery for tumor resections (42%), chemotherapy (40%), and radiation therapy (33%), which were used alone or in combination according to the hematologist/ oncologist responsible for each patient. Comorbidities were identified in 349 patients (62%), and the most frequent were arterial hypertension (38%), diabetes mellitus (14%), and chronic pulmonary disease (12%). Comparisons between patients with solid tumors and those with hematologic malignancies are depicted in e-Table 1 of Electronic Supplementary Material.
303 Table 1
Main patients' characteristics and outcomes
Variables Demographics Age (y) Male sex Comorbidity score (ACE-27) None-mild Moderate-severe Cancer-related characteristics Type of cancer Locoregional solid tumor Metastatic solid tumor Low-grade hematologic malignancy High-grade hematologic malignancy Cancer status Controlled/remission Active—newly diagnosed Active—recurrence/progression Performance status 0-1 2-4 Neutropenia Severity of acute illness at ICU admission SAPS II score (points) SAPS II–predicted mortality (%) SOFA on the first day of ICU (points) Acute organ failures (n) Mechanical ventilation Dialysis Vasopressors Outcome data ICU LOS (d) Hospital LOS (d) End-of-life decisions at the ICU ICU mortality Hospital mortality 6-mo mortality
All patients (n = 563) 59.2 ± 17.8 301 (54%) 424 (75%) 139 (25%)
348 (62%) 88 (16%) 39 (7%) 88 (16%) 260 (46%) 200 (36%) 103 (18%) 269 (48%) 294 (52%) 71 (13%) 51.0 ± 16.1 48.4 ± 26.8 8 (5-11) 2 (1-3) 489 (87%) 110 (20%) 372 (64%) 9 (4-18) 23 (11-43) 165 (29%) 289 (51%) 364 (65%) 407 (72%)
Results are expressed as mean ± SD, median (25%-75% interquartile range), and n (%). LOS indicates length of stay; ACE-27, Adult Comorbidity Evaluation—27.
3.2. Infection-related characteristics The main infection-related characteristics are reported in Table 2. Most of the patients had microbiologically proven infection (68%) and had severe sepsis/septic shock (91%). As expected, the most frequent sites of infection were the lung, abdomen, and urinary tract, and 24 patients (4%) had more than 1 site of infection. Gram-negative bacteria were responsible for more than half of the episodes of infection; 207 patients (38%) had polymicrobial (N1 infectious agent) infections.
3.3. Outcome analysis The overall ICU, hospital, and 6-month mortality rates were 51%, 65%, and 72%, respectively, and were higher in
304 Table 2
M.M. Rosolem et al. Infection-related characteristics
Variables Proof of infection Clinically suspected Microbiologically proven Severity Sepsis Severe sepsis Septic shock Acquisition Community acquired Nosocomial Positive blood cultures Pathogens a Gram-positive bacteria Enterococci Staphylococcus aureus Streptococci group D Pneumococci Other Gram-negative bacteria Escherichia coli Pseudomonas Klebsiella Enterobacter Proteus Morganella Stenotrophomonas maltophilia Other Fungi Candida species Other fungi Other infectious agents Multiresistant bacteria Site of infection Lung Abdomen Urinary tract Skin/soft tissue Primary bloodstream infection Central nervous system Other/unknown More than 1 site of infection
All patients (n = 563) 180 (32%) 383 (68%) 48 (9%) 143 (25%) 372 (66%) 227 (40%) 336 (60%) 114 (20%) 168 (30%) 92 (16%) 61 (11%) 16 (3%) 13 (2%) 12 (3%) 297 (53%) 91 (16%) 74 (13%) 72 (13%) 46 (8%) 34 (6%) 18 (4%) 14 (3%) 92 (16%) 43 (8%) 35 (6%) 8 (1%) 19 (3%) 81 (14%) 246 (44%) 172 (31%) 42 (8%) 35 (6%) 24 (4%) 11 (2%) 57 (10%) 24 (4%)
Results expressed as mean ± SD, median (25%–75% interquartile range), and n (%). a Three hundred eighty-three patients had microbiologically proven infections; patients could have more than 1 pathogen.
patients with septic shock (62%, 74%, and 80%) than in patients with severe sepsis (36%, 55%, and 67%) and those with sepsis (15%, 25%, and 36%) (P b .001 for all comparisons). Unadjusted mortality rates for the most common primary sites of infection were higher in patients with pneumonia and bloodstream infections than in patients with urinary tract and abdominal infections. End-of-life decisions were taken in 165 patients (29%) after a median of 7 days (2-14 days) in the ICU. Main outcome data are depicted in Table 1. The SAPS II scores increased along with
severity of sepsis (33.8 ± 10.5 points in patients with sepsis; 45.8 ± 15.1 points in those with severe sepsis, and 56.8 ± 15.1 in those with septic shock, P b .001). The AROC of SAPS II score was 0.72 (95% CI, 0.68-0.77). The results of univariate analysis of predictive factors for hospital mortality in all studied patients are reported in eTable 2 of the electronic supplementary material. Sex, type of admission, type of cancer, cancer status, PS, nosocomial infection, abdominal infections, urinary tract infections, presence of 3 to 4 systemic inflammatory response syndrome (SIRS) criteria, and 5 individual organ failures (cardiovascular, respiratory, neurologic, hematologic, and renal) were entered into the multivariate model. The final model is presented in Table 3 and had both good calibration and discrimination (Table 3). Finally, patients were stratified
Table 3
Multivariate analysis of hospital mortality predictors
Variables
Coefficients OR (95% CI)
Type of admission Emergency surgery Medical
0.784
Cancer status Controlled/remission Active—newly diagnosed
0.565
Active—recurrence/ progression Performance status 0-1 2-4
0.884
1.00 2.19 (1.40-3.42) 1.00 1.76 (1.12-2.75) 2.42 (1.35-4.35)
P
.001
.013 .003
1.272
1.00 3.57 b.001 (2.36-5.97)
Site of infection Urinary tract infection Indeterminate or other site of infection SIRS criteria (n) At least 2 3 or 4
1.189
1.00 3.28 (1.57 6.86)
.002
0.589
1 1.80 (1.20-2.72)
.014
Organ dysfunctions Cardiovascular a
0.664
Respiratory a
0.827
Renal
0.750
Constant
1.94 (1.27-2.94) 2.29 (1.24-4.23) 2.12 (1.34-3.35)
.008 .002 .001
−3.568
n = 563. Area under receiver operating characteristic curve, 0.80 (95% CI, 0.760.84); Hosmer-Lemeshow goodness-of-fit (χ2 = 6.106, P = .635). OR indicates odds ratio. a For these variables, the OR are also affected by the associated interaction term: cardiovascular dysfunction × respiratory dysfunction (coefficient, 1.839; odds ratio, 6.29 [1.19-33.32]; P = .032).
Patients with cancer and sepsis
Fig. 1 Survival curves for patients with cancer and sepsis according to the number of independent risk factors (log-rank test, 143.26; P b .001). Patients were classified into categories of risk for death: low (1 factor), moderate (2 or 3 factors), high (4 or 5 factors), and very high risk (6-8 factors).
according to the number of independent risk factors reported in Table 3, and Kaplan-Meier curves were plotted (Fig. 1). The 6-month mortality rates were 8% (1/12), 47% (39/83), 69% (170/246), and 90% (197/219) in patients with 1, 2 to 3, 4 to 5, and 6 to 8 risk factors, respectively (P b .001).
4. Discussion This study is one of the largest cohorts of critically ill patients with cancer and sepsis published to date, with 91% of the patients presenting with severe sepsis and septic shock. We demonstrated that sepsis remains a frequent complication in patients with cancer and responsible for almost half of ICU admissions of nonscheduled surgical patients. Hospital and 6-month mortality rates were 65% and 72%, respectively, which are in accordance with current literature [8,11,12,14,16]. In addition, although the reported mortality rates were high, they are congruent with the degree of severity of organ dysfunctions presented by our patients and comparable with those reported in studies in general patients with severe sepsis and septic shock [25]. The present study also adds to current literature because cancer-related characteristics, which can potentially influence the outcomes in these patients, were also analyzed. We observed that adjusting for the type of ICU admission in multivariate analysis, mortality in these patients was
305 associated with a poor PS; active cancer; overt SIRS; and when cardiovascular, respiratory, or renal dysfunctions were present. Patients with non–urinary tract infections, mostly represented by patients with pneumonia and abdominal infections, had worse outcomes. Moreover, our study also reinforces that the type of cancer per se, recent exposure to anticancer treatments, and neutropenia are no longer major determinants of short-term mortality in critically ill patients with cancer [7,14,26]. Patients with cancer represent a considerable subgroup among septic patients. Recent studies showed that up to 1 in 6 patients with severe sepsis or septic shock has malignancies [4,15]; these patients present with up to 30% higher risk for death when compared with general patients [25,27]. Furthermore, critically ill patients with cancer with sepsis have longer duration of both ICU and hospital stays than general patients with sepsis, thus requiring the use of more resources [15]. For all those reasons, considering ICU admission for critically ill patients with cancer remained controversial for many years. However, the outcomes in these patients including those presenting with severe infectious complications seem to be improving. In a study, Pène et al [11] evaluated the temporal trends in ICU management and outcomes of patients with cancer with septic shock in 2 different periods, 1998 to 2001 and 2002 to 2005. Both 28-day and 6-month survival rates were significantly higher in patients admitted to the ICU during the later period. The authors ascribed the increases in survival to both a better selection of patients and improvements in the care and management, including the incorporation of new therapeutic strategies for sepsis [11]. In another study, Larché et al [12] evaluated retrospectively 88 critically ill patients with cancer admitted to ICU with septic shock over a 6-year period (1995-2000). They demonstrated that survival improved overtime and increased mortality was associated with a delay in administration antibiotic greater than 2 hours and persistent organ failure during the first ICU days. The nature and severity of organ failures seem to be the major determinants of outcome in critically ill patients with cancer. In a secondary analysis of the Sepsis Occurrence in Acutely Ill Patients study, Taccone et al [8] evaluated the characteristics and prognosis of critically ill patients with cancer. Hospital mortality rates in patients with solid cancer were comparable with general ICU patients (27%). However, taking into consideration only patients presenting with more than 3 organ failures, mortality was higher in patients with cancer [8]. In our study, organ dysfunctions were also associated with increased mortality. We can hypothesize that, in this particular high-risk group of patients, preemptive attitudes focused on early recognition of individuals before the onset or the worsening of organ failures and provision of close monitoring and support (including early ICU referral) are essential [11]. In addition, information provided in the present study has important clinical implications and can also help physicians to identify patients with cancer with infections at increased
306 risk for a more severe clinical course. For instance, approximately 50% of patients proceeded from wards, and most infections were acquired during hospitalization (60%). In this sense, microbiological surveillance and routine check for removal of unnecessary invasive devices must be encouraged to reduce the risk for infection, particularly in patients with previous clinical characteristics that are associated with worse outcomes such as those with poor compromised PS and active cancer. Nonetheless, some limitations should be considered when appraising our results. First, our study was performed at an ICU specialized in managing patients with cancer and severe complications. Over the last years, close collaboration among the ICU team, oncologists, and hematologists improved overtime supporting the broadening of ICU policies. Therefore, much caution is needed when transposing our finding to general ICUs, as potential biases regarding standards of care (eg, decisions to start or to withhold treatments and ICU admission/discharge policies) cannot be disregarded, particularly in a singlecenter study. Second, although we evaluated a large cohort of patients with different infectious complications, the study was underpowered to evaluate patients with opportunistic infections including those caused by invasive fungi and viruses. Moreover, patients with hematologic malignancies comprised only 23% of our population, and therefore, the present study may be also underpowered to evaluate them. Conversely, the 2 aforementioned characteristics can make our results more applicable to general ICUs caring for critically ill patients with cancer [7]. Third, we have not included information on novel biomarkers for sepsis nor on genetic polymorphisms; however, such information is not routinely available in clinical practice. Fourth, we have evaluated variables collected at the first day of ICU. However, sepsis is a dynamic process, and sequential data are expected to improve patient's characterization and risk stratification. Another important limitation was the lack of information on patients whose ICU admission was declined because they were considered either too sick or too well for benefit. Finally, because data on specific processes of care and adjunctive treatments for sepsis were not assessed, we were not able to evaluate response to treatments. In conclusion, sepsis remains a frequent complication in patients with cancer and associated with high mortality. A poor PS; active cancer; clinical admission; overt SIRS criteria; and the presence of cardiovascular, respiratory, or renal dysfunctions were relevant predictors of increased mortality in these patients. We believe that our results can be of help to assist intensivists and oncologists in clinical decisions and counseling of patients and families and to contribute with future research to improve characterization and risk stratification for them. Supplementary materials related to this article can be found online at doi:10.1016/j.jcrc.2011.06.014.
M.M. Rosolem et al.
Acknowledgments Authors' contributions: Study concept and design: Marcio Soares, Thiago Lisboa, and Jorge Salluh. Acquisition of data: Marcio Soares, Maíra Rosolem, Ligia Rabello, Juliana Leal, Pedro Caruso, and Ramon Costa. Analysis and interpretation of data: Marcio Soares, Thiago Lisboa, Jorge Salluh, Maira Rosolem, and Ligia Rabello. Drafting of the manuscript: Marcio Soares, Thiago Lisboa, Jorge Salluh, and Maira Rosolem. Critical revision of the manuscript for important intellectual content: Marcio Soares, Thiago Lisboa, Juliana Leal, Pedro Caruso, Ramon Costa, Jorge Salluh, Maira Rosolem, and Ligia Rabello. Statistical expertise: Márcio Soares and Thiago Lisboa. Study supervision: Márcio Soares and Jorge Salluh. Final approval of the version to be published: Marcio Soares, Thiago Lisboa, Juliana Leal, Pedro Caruso, Ramon Costa, Jorge Salluh, Maira Rosolem, and Ligia Rabello.
References [1] Thirumala R, Ramaswamy M, Chawla S. Diagnosis and management of infectious complications in critically ill patients with cancer. Crit Care Cli 2010;26:59-91. [2] Danai PA, Moss M, Mannino DM, et al. The epidemiology of sepsis in patients with malignancy. Chest 2006;129:1432-40. [3] Fox AC, Robertson CM, Belt B, et al. Cancer causes increased mortality and is associated with altered apoptosis in murine sepsis. Crit Care Med 2010;38:886-93. [4] 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. [5] Annane D, Aegerter P, Jars-Guincestre MC, et al. Current epidemiology of septic shock: the CUB-Rea Network. Am J Respir Crit Care Med 2003;168:165-72. [6] Vincent JL, Sakr Y, Sprung CL, et al. Sepsis in European intensive care units: results of the SOAP study. Crit Care Med 2006;34: 344-53. [7] Soares M, Caruso P, Silva E, et al. Characteristics and outcomes of patients with cancer requiring admission to intensive care units: a prospective multicenter study. Crit Care Med 2010;38:9-15. [8] Taccone FS, Artigas AA, Sprung CL, et al. Characteristics and outcomes of cancer patients in European ICUs. Crit Care 2009;13:R15. [9] Benoit DD, Vandewoude KH, Decruyenaere JM, et al. 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;31:104-12. [10] Maschmeyer G, Bertschat FL, Moesta KT, et al. Outcome analysis of 189 consecutive cancer patients referred to the intensive care unit as emergencies during a 2-year period. Eur J Cancer 2003;39:783-92. [11] Pène F, Percheron S, Lemiale V, et al. Temporal changes in management and outcome of septic shock in patients with malignancies in the intensive care unit. Crit Care Med 2008;36:690-6. [12] Larché J, Azoulay E, Fieux F, et al. Improved survival of critically ill cancer patients with septic shock. Intensive Care Med 2003;29: 1688-95. [13] Regazzoni CJ, Irrazabal C, Luna CM, et al. Cancer patients with septic shock: mortality predictors and neutropenia. Support Care Cancer 2004;12:833-9. [14] Vandijck DM, Benoit DD, Depuydt PO, et al. Impact of recent intravenous chemotherapy on outcome in severe sepsis and septic
Patients with cancer and sepsis
[15]
[16]
[17]
[18]
[19]
[20]
shock patients with hematological malignancies. Intensive Care Med 2008;34:847-55. Williams MD, Braun LA, Cooper LM, et al. Hospitalized cancer patients with severe sepsis: analysis of incidence, mortality, and associated costs of care. Crit Care 2004;8:R291-8. Hampshire PA, Welch CA, McCrossan LA, et al. Admission factors associated with hospital mortality in patients with haematological malignancy admitted to UK adult, general critical care units: a secondary analysis of the ICNARC Case Mix Programme Database. Crit Care 2009;13:R137. Soares M, Silva UV, Teles JM, et al. Validation of four prognostic scores in patients with cancer admitted to Brazilian intensive care units: results from a prospective multicenter study. Intensive Care Med 2010;36:1188-95. Levy MM, Fink MP, Marshall JC, et al. 2001 SCCM/ESICM/ACCP/ATS/SIS International Sepsis Definitions Conference. Crit Care Med 2003;31:1250-6. Le Gall JR, Lemeshow S, Saulnier F. A new simplified acute physiology score (SAPS II) based on a European/North American multicenter study. JAMA 1993;270:2957-63. Vincent JL, Moreno R, Takala J, et al. The SOFA (Sepsis-related Organ Failure Assessment) score to describe organ dysfunction/failure. On
307
[21]
[22]
[23] [24] [25]
[26]
[27]
behalf of the Working Group on Sepsis-Related Problems of the European Society of Intensive Care Medicine. Intensive Care Med 1996;22:707-10. Zubrod CG, Schneiderman M, Frei III E, et al. Appraisal of methods for the study of chemotherapy of cancer in man: comparative therapeutic trial of nitrogen mustard and triethylene thiophosphoramide. J Chron Dis 1960;11:7-33. Piccirillo JF. Adult Comorbidity Evaluation—27. Comorbidity data collection form. Available from URL: http://oto.wustl.edu/clinepi/ downloads.html: 2001 [accessed November 15, 2008]. Hosmer DW, Lemeshow S. Applied logistic regression. 2nd ed. New York (NY): Wiley-Interscience; 2000. Hanley JA, McNeil BJ. The meaning and use of the area under receiver operating characteristic (ROC) curve. Radiology 1982;143:29-36. Alberti C, Brun-Buisson C, Burchardi H, et al. Epidemiology of sepsis and infection in ICU patients from an international multicentre cohort study. Intensive Care Med 2002;28:108-21. Souza-Dantas VC, Salluh JI, Soares M. Impact of neutropenia on the outcomes of critically ill patients with cancer: a matched case-control study. Ann Oncol 2011 [Epub ahead of print]. Angus DC, Linde-Zwirble WT, Lidicker J, et al. Epidemiology of severe sepsis in the United States: analysis of incidence, outcome and associated costs of care. Crit Care Med 2001;29:1303-10.
Critically ill patients with cancer and sepsis: Clinical course and prognostic factors☆,☆☆,★ Maíra M. Rosolem MD a,b,1 , Lígia S.C.F. Rabello MD a,b,1 , Thiago Lisboa MD, PhD c , Pedro Caruso MD, PhD d , Ramon T. Costa MD d , Juliana V.R. Leal MD a , Jorge I.F. Salluh MD, PhD b,e , Márcio Soares MD, PhD b,e,⁎ a
ICU, Instituto Nacional de Câncer, Rio de Janeiro, Brazil 20230-130 Postgraduate Program, Instituto Nacional de Câncer, Rio de Janeiro, Brazil 20231-050 c ICU, Hospital de Clínicas, Universidade Federal do Rio Grande do Sul, Porto Alegre, Brazil 90035-903 d ICU, Hospital A. C. Camargo, São Paulo, Brazil 01509-010 e D'Or Institute for Research and Education, Rio de Janeiro RJ CEP 22281-100, Brazil b
Keywords: Cancer; Intensive care unit; Mortality; Sepsis; Outcome
Abstract Purpose: The purposes of this study were to evaluate the clinical course and to identify independent predictors of mortality in patients with cancer with sepsis. Materials and Methods: This is a secondary analysis of a prospective cohort study conducted at an oncological medical-surgical intensive care unit. Logistic regression was used to identify predictors of hospital mortality. Results: A total of 563 patients (77% solid tumor, 23% hematologic malignancies) were included over a 55month period. The most frequent sites of infection were the lung, abdomen, and urinary tract; 91% patients had severe sepsis/septic shock. Gram-negative bacteria were responsible for more than half of the episodes of infection; 38% of patients had polymicrobial infections. Intensive care unit, hospital, and 6-month mortality rates were 51%, 65%, and 72%, respectively. In multivariate analyses, sepsis in the context of medical complications; active disease; compromised performance status; presence of 3 to 4 systemic inflammatory response syndrome criteria; and the presence of respiratory, renal, and cardiovascular failures were associated with increased mortality. Adjusting for other covariates, patients with non–urinary tract infections, mostly represented by patients with pneumonia and abdominal infections, had worse outcomes. Conclusions: Sepsis remains a frequent complication in patients with cancer and associated with high mortality. Our results can be of help to assist intensivists in clinical decisions and to improve characterization and risk stratification in these patients. © 2012 Elsevier Inc. All rights reserved.
☆
This work was performed at the intensive care unit of the Instituto Nacional de Câncer, Brazil. Financial support: Dr Soares is supported in part by individual research grant from CNPq. ★ Conflicts of interest: All authors have no conflict of interest to declare. ⁎ Corresponding author. D'Or Institute for Research and Education, Rio de Janeiro RJ CEP 22281-100, Brazil. Tel.: +55 21 2538 3541; fax: +55 21 2294 8620. E-mail addresses: [email protected], [email protected] (M. Soares). 1 Drs Rosolem and Rabello contributed equally for the study. ☆☆
0883-9441/$ – see front matter © 2012 Elsevier Inc. All rights reserved. doi:10.1016/j.jcrc.2011.06.014
302
1. Introduction Patients with cancer are at increased risk for sepsis as a consequence of multiple mechanisms of immunosuppression imposed by the disease itself and aggressive treatments, including combined regimens of chemotherapy and radiation therapy, high dose of steroids, and hematopoietic stem cell transplantation [1-3]. In large databases, up to 1 in 5 patients admitted to intensive care units (ICUs) with sepsis have cancer [4-6], and along this line, sepsis is a leading reason for ICU admission in patients with cancer [7,8]. Moreover, despite the observation of improved survival rates over the last years, sepsis in patients with cancer remains associated with high morbidity, mortality, costs, and use of ICU resources, and information on this topic is limited [9-16]. Refinements in the information about the epidemiology, outcomes, and risk stratification of these patients are essential for future research to assist clinicians in clinical decision making, to allow better resource allocation, and for counseling patients and families. The aims of the present study were to evaluate the characteristics, clinical course, and outcomes and to identify independent predictors of mortality in a large cohort of patients with cancer admitted to the ICU with sepsis.
2. Patients and methods 2.1. Design and setting This was a secondary analysis of a prospective cohort study performed from January 2003 to July 2007 at the Instituto Nacional de Câncer, Rio de Janeiro, Brazil. The ICU is a 10-bed medical-surgical unit specialized in the care of patients with cancer [17]. As a rule, only patients with a chance of survival improvement are considered for ICU admission. However, some patients may occasionally be admitted while a full assessment of their cancer extent and therapeutic options is still ongoing. Decisions to withstand or withhold life-sustaining therapies are taken in patients who do not recover from the acute illness despite ICU care or if treatment for the cure or control of cancer cannot be given. This study was supported by institutional funds and did not interfere with clinical decisions related with patient care. The Ethics Committee of Instituto Nacional de Câncer approved the study (No. 10/2003), and the need for informed consent was waived.
2.2. Selection of participants, data collection, and definitions During the study period, every adult patient (≥18 years old) with a definite diagnosis of cancer and presenting with sepsis at ICU admission was evaluated. Patients in complete
M.M. Rosolem et al. remission longer than 5 years, those with an ICU stay of less than 24 hours, and readmissions were not considered. Infection was defined as the presence of a pathogenic microorganism in a sterile milieu (such as blood or cerebrospinal fluid) and/or clinically suspected infection that justified the administration of antibiotics [6,8]. Sepsis was diagnosed according to the consensus definitions [18]. Severe sepsis was defined as sepsis complicated by organ dysfunction, and septic shock was defined as severe sepsis with arterial hypotension (systolic pressure b90 mm Hg or mean arterial pressure b60 mm Hg or reduction in systolic pressure N40 mm Hg from baseline). Data related to patient's management were not collected. Demographic, clinical, and laboratory data were collected using standardized case report forms during the first day of ICU including main diagnosis for admission, the Simplified Acute Physiology Score (SAPS) II [19], the Sequential Organ Failure Assessment (SOFA) score [20], comorbidities, performance status (PS) (Eastern Cooperative Oncology Group scale) [21], and cancer- and treatment-related data. Cancer status was classified into 3 categories: remission/controlled (ie, patients in cancer remission or control who have undergone previous treatments, without evidence of recurrence according to the attending oncologist/ hematologist), recent diagnosis (patients with active disease diagnosed within the last 3 months with need for first-line anticancer treatment), and relapse/recurrence (patients with active disease with relapse or recurrence). For the purpose of the present study, individual organ failures were diagnosed in case of SOFA score of 2 points or greater in each domain [8]. In addition, patients receiving dialysis in the context of acute kidney injury and invasive mechanical ventilation on the first day of ICU were considered as having renal and respiratory failures regardless of the SOFA score, respectively. Comorbidities were evaluated using the Adult Comorbidity Evaluation 27 [22]. Neutropenia was defined as a neutrophil count less than 500/mm3. Vital status at hospital discharge was the outcome of interest, but the ICU and 6-month mortality rates were also assessed.
2.3. Data processing and statistical analysis Data entry was performed by a single investigator (MS), and consistency was assessed by a rechecking procedure of a 10% random sample of patients. Data were screened in detail by 4 investigators (MS, LSFR, MMR, and TL) for missing information and implausible and outlying values. Continuous variables were reported as mean ± SD or median (25%-75% interquartile range). Univariate and multivariate logistic regressions were used to identify factors associated with hospital mortality [23]. Linearity between each continuous variable and the dependent variable was demonstrated using locally weighted scatterplot smoothing [23]. In case of nonlinearity, the variable was stratified according to the inflection points and clinical
Patients with cancer and sepsis significance. For categorical variables with multiple levels, the reference level was attributed to the one with the lowest probability of the dependent variable. Variables yielding P b .2 by univariate analysis and those considered clinically relevant according to the current literature were entered into the multivariate analyses. Results were summarized as odds ratios and respective 95% confidence intervals (CIs). The statistical model was internally validated using the technique of bootstrap resampling. This technique is efficient and provides nearly unbiased estimates of the predictive accuracy of the model. We repeated the entire modeling process to validate the final model (1000 replications) and to adjust the estimated performance and regression coefficients. Possible interactions were tested. The area under the receiver operating characteristic curve (AROC) was used to assess the models' discrimination; an AROC of 1.0 denotes perfect, whereas a value close to 0.50 indicates no apparent accuracy [24]. The Hosmer-Lemeshow goodness-of-fit test was used to evaluate agreement between the observed and expected results across all strata of probabilities of the outcome of interest (calibration) [23]. With this test, P N .05 indicate a good fit for the model. Two-tailed P b .05 was considered statistically significant.
3. Results 3.1. Characterization of the study population Excluding scheduled surgical and those with ICU admission less than 24 hours, a total of 1332 patients were admitted to our ICU during the study period. Of them, 563 (42%) had sepsis fulfilling the eligibility criteria and constituted the study population. The sources of ICU admission were the operating room (27%), emergency department (23%), and wards (50%). There were 436 patients (77%) with solid tumors, and 127 patients (23%) had hematologic malignancies. The patients' main characteristics are depicted in Table 1. The primary sites of solid tumors were gastrointestinal (35%), head and neck (13%), urogenital (6%), lung (6%), breast (5%), and others (12%). The main hematologic malignancies were non-Hodgkin lymphomas (14%), leukemias (6%), multiple myeloma (2%), and others (1%). Previous anticancer treatments included surgery for tumor resections (42%), chemotherapy (40%), and radiation therapy (33%), which were used alone or in combination according to the hematologist/ oncologist responsible for each patient. Comorbidities were identified in 349 patients (62%), and the most frequent were arterial hypertension (38%), diabetes mellitus (14%), and chronic pulmonary disease (12%). Comparisons between patients with solid tumors and those with hematologic malignancies are depicted in e-Table 1 of Electronic Supplementary Material.
303 Table 1
Main patients' characteristics and outcomes
Variables Demographics Age (y) Male sex Comorbidity score (ACE-27) None-mild Moderate-severe Cancer-related characteristics Type of cancer Locoregional solid tumor Metastatic solid tumor Low-grade hematologic malignancy High-grade hematologic malignancy Cancer status Controlled/remission Active—newly diagnosed Active—recurrence/progression Performance status 0-1 2-4 Neutropenia Severity of acute illness at ICU admission SAPS II score (points) SAPS II–predicted mortality (%) SOFA on the first day of ICU (points) Acute organ failures (n) Mechanical ventilation Dialysis Vasopressors Outcome data ICU LOS (d) Hospital LOS (d) End-of-life decisions at the ICU ICU mortality Hospital mortality 6-mo mortality
All patients (n = 563) 59.2 ± 17.8 301 (54%) 424 (75%) 139 (25%)
348 (62%) 88 (16%) 39 (7%) 88 (16%) 260 (46%) 200 (36%) 103 (18%) 269 (48%) 294 (52%) 71 (13%) 51.0 ± 16.1 48.4 ± 26.8 8 (5-11) 2 (1-3) 489 (87%) 110 (20%) 372 (64%) 9 (4-18) 23 (11-43) 165 (29%) 289 (51%) 364 (65%) 407 (72%)
Results are expressed as mean ± SD, median (25%-75% interquartile range), and n (%). LOS indicates length of stay; ACE-27, Adult Comorbidity Evaluation—27.
3.2. Infection-related characteristics The main infection-related characteristics are reported in Table 2. Most of the patients had microbiologically proven infection (68%) and had severe sepsis/septic shock (91%). As expected, the most frequent sites of infection were the lung, abdomen, and urinary tract, and 24 patients (4%) had more than 1 site of infection. Gram-negative bacteria were responsible for more than half of the episodes of infection; 207 patients (38%) had polymicrobial (N1 infectious agent) infections.
3.3. Outcome analysis The overall ICU, hospital, and 6-month mortality rates were 51%, 65%, and 72%, respectively, and were higher in
304 Table 2
M.M. Rosolem et al. Infection-related characteristics
Variables Proof of infection Clinically suspected Microbiologically proven Severity Sepsis Severe sepsis Septic shock Acquisition Community acquired Nosocomial Positive blood cultures Pathogens a Gram-positive bacteria Enterococci Staphylococcus aureus Streptococci group D Pneumococci Other Gram-negative bacteria Escherichia coli Pseudomonas Klebsiella Enterobacter Proteus Morganella Stenotrophomonas maltophilia Other Fungi Candida species Other fungi Other infectious agents Multiresistant bacteria Site of infection Lung Abdomen Urinary tract Skin/soft tissue Primary bloodstream infection Central nervous system Other/unknown More than 1 site of infection
All patients (n = 563) 180 (32%) 383 (68%) 48 (9%) 143 (25%) 372 (66%) 227 (40%) 336 (60%) 114 (20%) 168 (30%) 92 (16%) 61 (11%) 16 (3%) 13 (2%) 12 (3%) 297 (53%) 91 (16%) 74 (13%) 72 (13%) 46 (8%) 34 (6%) 18 (4%) 14 (3%) 92 (16%) 43 (8%) 35 (6%) 8 (1%) 19 (3%) 81 (14%) 246 (44%) 172 (31%) 42 (8%) 35 (6%) 24 (4%) 11 (2%) 57 (10%) 24 (4%)
Results expressed as mean ± SD, median (25%–75% interquartile range), and n (%). a Three hundred eighty-three patients had microbiologically proven infections; patients could have more than 1 pathogen.
patients with septic shock (62%, 74%, and 80%) than in patients with severe sepsis (36%, 55%, and 67%) and those with sepsis (15%, 25%, and 36%) (P b .001 for all comparisons). Unadjusted mortality rates for the most common primary sites of infection were higher in patients with pneumonia and bloodstream infections than in patients with urinary tract and abdominal infections. End-of-life decisions were taken in 165 patients (29%) after a median of 7 days (2-14 days) in the ICU. Main outcome data are depicted in Table 1. The SAPS II scores increased along with
severity of sepsis (33.8 ± 10.5 points in patients with sepsis; 45.8 ± 15.1 points in those with severe sepsis, and 56.8 ± 15.1 in those with septic shock, P b .001). The AROC of SAPS II score was 0.72 (95% CI, 0.68-0.77). The results of univariate analysis of predictive factors for hospital mortality in all studied patients are reported in eTable 2 of the electronic supplementary material. Sex, type of admission, type of cancer, cancer status, PS, nosocomial infection, abdominal infections, urinary tract infections, presence of 3 to 4 systemic inflammatory response syndrome (SIRS) criteria, and 5 individual organ failures (cardiovascular, respiratory, neurologic, hematologic, and renal) were entered into the multivariate model. The final model is presented in Table 3 and had both good calibration and discrimination (Table 3). Finally, patients were stratified
Table 3
Multivariate analysis of hospital mortality predictors
Variables
Coefficients OR (95% CI)
Type of admission Emergency surgery Medical
0.784
Cancer status Controlled/remission Active—newly diagnosed
0.565
Active—recurrence/ progression Performance status 0-1 2-4
0.884
1.00 2.19 (1.40-3.42) 1.00 1.76 (1.12-2.75) 2.42 (1.35-4.35)
P
.001
.013 .003
1.272
1.00 3.57 b.001 (2.36-5.97)
Site of infection Urinary tract infection Indeterminate or other site of infection SIRS criteria (n) At least 2 3 or 4
1.189
1.00 3.28 (1.57 6.86)
.002
0.589
1 1.80 (1.20-2.72)
.014
Organ dysfunctions Cardiovascular a
0.664
Respiratory a
0.827
Renal
0.750
Constant
1.94 (1.27-2.94) 2.29 (1.24-4.23) 2.12 (1.34-3.35)
.008 .002 .001
−3.568
n = 563. Area under receiver operating characteristic curve, 0.80 (95% CI, 0.760.84); Hosmer-Lemeshow goodness-of-fit (χ2 = 6.106, P = .635). OR indicates odds ratio. a For these variables, the OR are also affected by the associated interaction term: cardiovascular dysfunction × respiratory dysfunction (coefficient, 1.839; odds ratio, 6.29 [1.19-33.32]; P = .032).
Patients with cancer and sepsis
Fig. 1 Survival curves for patients with cancer and sepsis according to the number of independent risk factors (log-rank test, 143.26; P b .001). Patients were classified into categories of risk for death: low (1 factor), moderate (2 or 3 factors), high (4 or 5 factors), and very high risk (6-8 factors).
according to the number of independent risk factors reported in Table 3, and Kaplan-Meier curves were plotted (Fig. 1). The 6-month mortality rates were 8% (1/12), 47% (39/83), 69% (170/246), and 90% (197/219) in patients with 1, 2 to 3, 4 to 5, and 6 to 8 risk factors, respectively (P b .001).
4. Discussion This study is one of the largest cohorts of critically ill patients with cancer and sepsis published to date, with 91% of the patients presenting with severe sepsis and septic shock. We demonstrated that sepsis remains a frequent complication in patients with cancer and responsible for almost half of ICU admissions of nonscheduled surgical patients. Hospital and 6-month mortality rates were 65% and 72%, respectively, which are in accordance with current literature [8,11,12,14,16]. In addition, although the reported mortality rates were high, they are congruent with the degree of severity of organ dysfunctions presented by our patients and comparable with those reported in studies in general patients with severe sepsis and septic shock [25]. The present study also adds to current literature because cancer-related characteristics, which can potentially influence the outcomes in these patients, were also analyzed. We observed that adjusting for the type of ICU admission in multivariate analysis, mortality in these patients was
305 associated with a poor PS; active cancer; overt SIRS; and when cardiovascular, respiratory, or renal dysfunctions were present. Patients with non–urinary tract infections, mostly represented by patients with pneumonia and abdominal infections, had worse outcomes. Moreover, our study also reinforces that the type of cancer per se, recent exposure to anticancer treatments, and neutropenia are no longer major determinants of short-term mortality in critically ill patients with cancer [7,14,26]. Patients with cancer represent a considerable subgroup among septic patients. Recent studies showed that up to 1 in 6 patients with severe sepsis or septic shock has malignancies [4,15]; these patients present with up to 30% higher risk for death when compared with general patients [25,27]. Furthermore, critically ill patients with cancer with sepsis have longer duration of both ICU and hospital stays than general patients with sepsis, thus requiring the use of more resources [15]. For all those reasons, considering ICU admission for critically ill patients with cancer remained controversial for many years. However, the outcomes in these patients including those presenting with severe infectious complications seem to be improving. In a study, Pène et al [11] evaluated the temporal trends in ICU management and outcomes of patients with cancer with septic shock in 2 different periods, 1998 to 2001 and 2002 to 2005. Both 28-day and 6-month survival rates were significantly higher in patients admitted to the ICU during the later period. The authors ascribed the increases in survival to both a better selection of patients and improvements in the care and management, including the incorporation of new therapeutic strategies for sepsis [11]. In another study, Larché et al [12] evaluated retrospectively 88 critically ill patients with cancer admitted to ICU with septic shock over a 6-year period (1995-2000). They demonstrated that survival improved overtime and increased mortality was associated with a delay in administration antibiotic greater than 2 hours and persistent organ failure during the first ICU days. The nature and severity of organ failures seem to be the major determinants of outcome in critically ill patients with cancer. In a secondary analysis of the Sepsis Occurrence in Acutely Ill Patients study, Taccone et al [8] evaluated the characteristics and prognosis of critically ill patients with cancer. Hospital mortality rates in patients with solid cancer were comparable with general ICU patients (27%). However, taking into consideration only patients presenting with more than 3 organ failures, mortality was higher in patients with cancer [8]. In our study, organ dysfunctions were also associated with increased mortality. We can hypothesize that, in this particular high-risk group of patients, preemptive attitudes focused on early recognition of individuals before the onset or the worsening of organ failures and provision of close monitoring and support (including early ICU referral) are essential [11]. In addition, information provided in the present study has important clinical implications and can also help physicians to identify patients with cancer with infections at increased
306 risk for a more severe clinical course. For instance, approximately 50% of patients proceeded from wards, and most infections were acquired during hospitalization (60%). In this sense, microbiological surveillance and routine check for removal of unnecessary invasive devices must be encouraged to reduce the risk for infection, particularly in patients with previous clinical characteristics that are associated with worse outcomes such as those with poor compromised PS and active cancer. Nonetheless, some limitations should be considered when appraising our results. First, our study was performed at an ICU specialized in managing patients with cancer and severe complications. Over the last years, close collaboration among the ICU team, oncologists, and hematologists improved overtime supporting the broadening of ICU policies. Therefore, much caution is needed when transposing our finding to general ICUs, as potential biases regarding standards of care (eg, decisions to start or to withhold treatments and ICU admission/discharge policies) cannot be disregarded, particularly in a singlecenter study. Second, although we evaluated a large cohort of patients with different infectious complications, the study was underpowered to evaluate patients with opportunistic infections including those caused by invasive fungi and viruses. Moreover, patients with hematologic malignancies comprised only 23% of our population, and therefore, the present study may be also underpowered to evaluate them. Conversely, the 2 aforementioned characteristics can make our results more applicable to general ICUs caring for critically ill patients with cancer [7]. Third, we have not included information on novel biomarkers for sepsis nor on genetic polymorphisms; however, such information is not routinely available in clinical practice. Fourth, we have evaluated variables collected at the first day of ICU. However, sepsis is a dynamic process, and sequential data are expected to improve patient's characterization and risk stratification. Another important limitation was the lack of information on patients whose ICU admission was declined because they were considered either too sick or too well for benefit. Finally, because data on specific processes of care and adjunctive treatments for sepsis were not assessed, we were not able to evaluate response to treatments. In conclusion, sepsis remains a frequent complication in patients with cancer and associated with high mortality. A poor PS; active cancer; clinical admission; overt SIRS criteria; and the presence of cardiovascular, respiratory, or renal dysfunctions were relevant predictors of increased mortality in these patients. We believe that our results can be of help to assist intensivists and oncologists in clinical decisions and counseling of patients and families and to contribute with future research to improve characterization and risk stratification for them. Supplementary materials related to this article can be found online at doi:10.1016/j.jcrc.2011.06.014.
M.M. Rosolem et al.
Acknowledgments Authors' contributions: Study concept and design: Marcio Soares, Thiago Lisboa, and Jorge Salluh. Acquisition of data: Marcio Soares, Maíra Rosolem, Ligia Rabello, Juliana Leal, Pedro Caruso, and Ramon Costa. Analysis and interpretation of data: Marcio Soares, Thiago Lisboa, Jorge Salluh, Maira Rosolem, and Ligia Rabello. Drafting of the manuscript: Marcio Soares, Thiago Lisboa, Jorge Salluh, and Maira Rosolem. Critical revision of the manuscript for important intellectual content: Marcio Soares, Thiago Lisboa, Juliana Leal, Pedro Caruso, Ramon Costa, Jorge Salluh, Maira Rosolem, and Ligia Rabello. Statistical expertise: Márcio Soares and Thiago Lisboa. Study supervision: Márcio Soares and Jorge Salluh. Final approval of the version to be published: Marcio Soares, Thiago Lisboa, Juliana Leal, Pedro Caruso, Ramon Costa, Jorge Salluh, Maira Rosolem, and Ligia Rabello.
References [1] Thirumala R, Ramaswamy M, Chawla S. Diagnosis and management of infectious complications in critically ill patients with cancer. Crit Care Cli 2010;26:59-91. [2] Danai PA, Moss M, Mannino DM, et al. The epidemiology of sepsis in patients with malignancy. Chest 2006;129:1432-40. [3] Fox AC, Robertson CM, Belt B, et al. Cancer causes increased mortality and is associated with altered apoptosis in murine sepsis. Crit Care Med 2010;38:886-93. [4] 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. [5] Annane D, Aegerter P, Jars-Guincestre MC, et al. Current epidemiology of septic shock: the CUB-Rea Network. Am J Respir Crit Care Med 2003;168:165-72. [6] Vincent JL, Sakr Y, Sprung CL, et al. Sepsis in European intensive care units: results of the SOAP study. Crit Care Med 2006;34: 344-53. [7] Soares M, Caruso P, Silva E, et al. Characteristics and outcomes of patients with cancer requiring admission to intensive care units: a prospective multicenter study. Crit Care Med 2010;38:9-15. [8] Taccone FS, Artigas AA, Sprung CL, et al. Characteristics and outcomes of cancer patients in European ICUs. Crit Care 2009;13:R15. [9] Benoit DD, Vandewoude KH, Decruyenaere JM, et al. 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;31:104-12. [10] Maschmeyer G, Bertschat FL, Moesta KT, et al. Outcome analysis of 189 consecutive cancer patients referred to the intensive care unit as emergencies during a 2-year period. Eur J Cancer 2003;39:783-92. [11] Pène F, Percheron S, Lemiale V, et al. Temporal changes in management and outcome of septic shock in patients with malignancies in the intensive care unit. Crit Care Med 2008;36:690-6. [12] Larché J, Azoulay E, Fieux F, et al. Improved survival of critically ill cancer patients with septic shock. Intensive Care Med 2003;29: 1688-95. [13] Regazzoni CJ, Irrazabal C, Luna CM, et al. Cancer patients with septic shock: mortality predictors and neutropenia. Support Care Cancer 2004;12:833-9. [14] Vandijck DM, Benoit DD, Depuydt PO, et al. Impact of recent intravenous chemotherapy on outcome in severe sepsis and septic
Patients with cancer and sepsis
[15]
[16]
[17]
[18]
[19]
[20]
shock patients with hematological malignancies. Intensive Care Med 2008;34:847-55. Williams MD, Braun LA, Cooper LM, et al. Hospitalized cancer patients with severe sepsis: analysis of incidence, mortality, and associated costs of care. Crit Care 2004;8:R291-8. Hampshire PA, Welch CA, McCrossan LA, et al. Admission factors associated with hospital mortality in patients with haematological malignancy admitted to UK adult, general critical care units: a secondary analysis of the ICNARC Case Mix Programme Database. Crit Care 2009;13:R137. Soares M, Silva UV, Teles JM, et al. Validation of four prognostic scores in patients with cancer admitted to Brazilian intensive care units: results from a prospective multicenter study. Intensive Care Med 2010;36:1188-95. Levy MM, Fink MP, Marshall JC, et al. 2001 SCCM/ESICM/ACCP/ATS/SIS International Sepsis Definitions Conference. Crit Care Med 2003;31:1250-6. Le Gall JR, Lemeshow S, Saulnier F. A new simplified acute physiology score (SAPS II) based on a European/North American multicenter study. JAMA 1993;270:2957-63. Vincent JL, Moreno R, Takala J, et al. The SOFA (Sepsis-related Organ Failure Assessment) score to describe organ dysfunction/failure. On
307
[21]
[22]
[23] [24] [25]
[26]
[27]
behalf of the Working Group on Sepsis-Related Problems of the European Society of Intensive Care Medicine. Intensive Care Med 1996;22:707-10. Zubrod CG, Schneiderman M, Frei III E, et al. Appraisal of methods for the study of chemotherapy of cancer in man: comparative therapeutic trial of nitrogen mustard and triethylene thiophosphoramide. J Chron Dis 1960;11:7-33. Piccirillo JF. Adult Comorbidity Evaluation—27. Comorbidity data collection form. Available from URL: http://oto.wustl.edu/clinepi/ downloads.html: 2001 [accessed November 15, 2008]. Hosmer DW, Lemeshow S. Applied logistic regression. 2nd ed. New York (NY): Wiley-Interscience; 2000. Hanley JA, McNeil BJ. The meaning and use of the area under receiver operating characteristic (ROC) curve. Radiology 1982;143:29-36. Alberti C, Brun-Buisson C, Burchardi H, et al. Epidemiology of sepsis and infection in ICU patients from an international multicentre cohort study. Intensive Care Med 2002;28:108-21. Souza-Dantas VC, Salluh JI, Soares M. Impact of neutropenia on the outcomes of critically ill patients with cancer: a matched case-control study. Ann Oncol 2011 [Epub ahead of print]. Angus DC, Linde-Zwirble WT, Lidicker J, et al. Epidemiology of severe sepsis in the United States: analysis of incidence, outcome and associated costs of care. Crit Care Med 2001;29:1303-10.