Septic shock is a subset of sepsis with underlying circulatory and cellular metabolism abnormalities resulting in high mortality rates (1). Cancer patients are at increased risk of septic shock due to impairment in immunity, exposure to new chemotherapeutic drugs targeting lymphocytes, use of extensive antibiotic prophylaxis, and to additional risk factors of infections (2–8).
Anemia and RBC transfusion may worsen outcomes and increase morbidity and mortality in cancer patients who develop septic shock. Anemia occurs in most patients undergoing surgery for cancer, largely because of blood loss during complex procedures, and due to poor nutritional condition, tumor stage, cancer-related anemia, and previous chemotherapy. RBC transfusion is frequently required for these patients (7). On the other hand, RBC transfusion is associated with severe complications such as transfusion-related immunomodulation, transfusion-related cardiovascular overload, and transfusion-related acute lung injury. There is additional concern of the association of transfusion with cancer progression or recurrence. In recent years, randomized trials and guidelines moved the approach toward a more restrictive strategy of RBC transfusions in critically ill patients (9–14).
Despite the fact that a restrictive policy of RBC transfusion is cost saving and safe in several clinical scenarios, there are no randomized data to guide transfusion decision in cancer patients with septic shock. We therefore conducted the Transfusion Requirements in Critically Ill Oncologic Patients (TRICOP) randomized study to elucidate whether the restrictive strategy of RBC transfusion reduces mortality when compared with the liberal strategy in cancer patients with septic shock.
MATERIALS AND METHODS
We performed a randomized, double-blind, parallel-group controlled, pragmatic clinical trial in a tertiary university hospital specialized in cancer treatment, the Cancer Institute of Faculty of Medicine of the University of Sao Paulo, in Brazil. The study was conducted in accordance with the International Conference on Harmonization Good Clinical Practice, and the protocol approved by the local ethics committee (the Comitê de Ética e Pesquisa, Faculdade de Medicina da Universidade de São Paulo). Written informed consent was obtained from all patients or their next of kin. The study protocol was registered at Clinicaltrials.gov as NCT01648946.
We screened patients 18 years old or older with a diagnosis of solid cancer and fulfilling the criteria for septic shock (15–17) (definition in the supplemental data, Supplemental Digital Content 1, http://links.lww.com/CCM/C375) in the first 6 hours after ICU admission. We excluded patients less than 18 years old, those who refused to participate, who had inability to receive transfusion of blood components (Jehovah Witnesses, history of allergy, or hypersensitivity reactions to blood components), and those with a too high expected mortality or transfusion rate (supplemental data, Supplemental Digital Content 1, http://links.lww.com/CCM/C375).
Intervention and Hemoglobin Measurement
Patients were randomly assigned to the liberal strategy of RBC transfusion (patients received one unit of RBC each time Hb < 9 g/dL) or to the restrictive strategy (one unit of RBC each time Hb < 7 g/dL) during ICU stay with transfusion thresholds based on previous studies (9, 10). In our study, patients had hemoglobin levels assessed after ICU admission, twice a day during ICU stay, and after each RBC transfusion. When the hemoglobin concentration after transfusion was below the preestablished threshold, a new RBC transfusion was administered, but ICU staff was instructed to transfuse only one RBC unit per time. Transfusion decisions were not performed blindly. All RBC units were leukodepleted. The intervention was performed only during patient stay in the ICU. After ICU discharge, patients received RBC transfusion at the discretion of the medical staff in the regular wards who was blinded to the treatment arm.
Physicians were instructed to follow the protocol according to the predefined transfusion threshold; however, in life-threatening situations, they could perform a RBC transfusion out of the protocol if they judged that there was an indication for it. These transfusions and any other transfusion performed against the study protocol were recorded and considered as a protocol deviation.
Data Collection and Outcomes
A team of three investigators collected baseline demographic and cancer characteristics together with outcome data (supplemental data, Supplemental Digital Content 1, http://links.lww.com/CCM/C375). The primary outcome, in accordance with most large scale clinical trials in ICU, was all-cause mortality by 28 days after randomization. Secondary outcomes included need of advanced organ support (invasive mechanical ventilation, inotropic therapy, or renal replacement therapy), cerebral ischemia (diagnosed by imaging and new focal deficit), acute myocardial infarction (supplemental data, Supplemental Digital Content 1, http://links.lww.com/CCM/C375), mesenteric ischemia, limb ischemia, and serious adverse reactions (hemolytic transfusion reactions, anaphylaxis, transfusion-associated lung injury, or transfusion-associated circulatory overload) in the 28 days after randomization. Other secondary outcomes were ICU and hospital length of stay, ICU readmission, and death by 60 and 90 days after randomization. We also evaluated predictive factors for 28-, 60-, and 90-day mortality.
Patients and investigators who collected outcomes were blinded to transfusion data.
Randomization and Masking
Eligible patients were randomly assigned to the liberal or restrictive RBC transfusion strategy by means of an Internet-based system that concealed assignments. Physicians and nurses of the ICU were aware of the groups of treatment.
Two blinded investigators assessed primary and secondary outcomes by patient records review or by telephone call (long-term survival). There was no identification of transfusion strategy group on patients, patient records, or patient beds. Patients and investigators who collected outcomes had no access to transfusion data and were unaware of the group assignment.
Sample size was based on a published 28-day mortality rate for patients with cancer and septic shock of 50% (5). Since there were no previous randomized studies addressing transfusion in cancer patients, we hypothesized, based on retrospective data, that the restrictive RBC transfusion strategy would reduce the absolute risk of the primary outcome by 16% when compared with the liberal strategy (18, 19). We calculated that a sample size of 300 patients would be required for the study to have 80% power to detect this difference in a two-sided test, at a 5% level of significance.
Data were analyzed on an intention-to-treat basis according to the randomized study group assignment. Continuous variables were analyzed using a t test or the Mann-Whitney U test, and categorical variables were compared using Pearson chi-square test, Fisher exact test, or a likelihood ratio test. Hemoglobin levels were compared between the groups using a mixed-design analysis of variance. The model was constructed using the nadir hemoglobin of each day during the first 8 days of ICU stay.
Continuous data are expressed as means with SDs or medians with interquartile ranges. The primary outcome analysis was analyzed with the use of logistic regression presented as hazard ratio, and secondary outcomes are presented as Mantel-Haenszel odds ratio with 95% CI. An unadjusted Kaplan-Meier survival estimates were calculated, dividing patients according to the transfusion strategy. A multivariate Cox proportional hazards model was performed to estimate predictive factors for mortality. Relevant variables identified in the univariate analysis (p ≤ 0.10) were included in the multivariate model. A two-sided p value of less than 0.05 was considered statistically significant. Analyses were performed using SPSS version 18.0 (SPSS, Chicago, IL).
Population of the Study
Between June 2012 and May 2014, 1,658 patients were screened for eligibility and 300 enrolled (149 to the liberal and 151 to the restrictive strategy group; Supplemental Fig. 1, Supplemental Digital Content 2, http://links.lww.com/CCM/C376). There were no exclusions after randomization or loss of follow-up, and all enrolled patients were analyzed for the primary outcome.
Baseline characteristics and severity scores were well-balanced between groups (Table 1). Patients were 61.5 ± 13.2 years old, and 154 (51%) were male. Pneumonia was the most common infection, followed by intra-abdominal and urinary infection. Gastrointestinal neoplasm was the most common cancer, followed by lung and genitourinary cancer. One fourth of the patients had received chemotherapy within 4 weeks before being enrolled in the study, and one tenth had received radiotherapy (Table 1).
Hemoglobin Concentration and Blood Transfusion
Hemoglobin at randomization was 9.7 ± 2.1 g/dL in the liberal group versus 9.6 ± 2.1 g/dL in the restrictive group. Hemoglobin nadir concentration in the liberal group was significantly higher (by approximately 0.7 g/dL) than in the restrictive group after randomization and during the ICU stay (p = 0.038) (Fig. 1A). In transfused patients, hemoglobin nadir concentration in the liberal group was significantly higher (by approximately 1.0–1.5 g/dL) than in the restrictive group after randomization and during the ICU stay (p < 0.001) (Fig. 1B).
Ninety-one patients (61%) in the liberal group versus 62 patients (41%) in the restrictive group were transfused (p < 0.001) (Table 2). Few protocol deviations were noted (supplemental data, Supplemental Digital Content 1, http://links.lww.com/CCM/C375)
The number of overall RBC units transfused was higher in the liberal group when compared with the restrictive group (314 vs 212 units). Patients in the liberal group received more RBC transfusion than those in the restrictive group (1 [0–3] vs 0 [0–2] unit; p < 0.001). There were no differences between the groups regarding requirements of platelets, fresh frozen plasma, or cryoprecipitate (Table 2).
At 28 days after randomization, we observed a trend toward an increased mortality rate in the liberal group (45%, 67 patients) when compared with the restrictive group (56%, 84 patients; hazard ratio, 0.74; 95% CI, 0.53–1.04; p = 0.08). The main cause of death in both groups was multiple organ dysfunction syndrome (Table 3). Withdrawal of active treatment/nonresuscitation orders was applied to 17 patients in the liberal group and 18 patients in the restrictive group (p = 0.89).
There were no differences between the groups regarding requirements of advanced organ support, ischemic events, serious adverse reactions, length of ICU and hospital stay, ICU readmission, and 60-day mortality (Table 3).
Regarding mortality rate at 90 days after randomization, the liberal group presented a lower mortality rate when compared with the restrictive group (59.1% [88 patients] vs 70.2% [106 patients]; hazard ratio, 0.72; 95% CI, 0.53–0.97; p = 0.03) (Table 3) with multivariate analyses (supplemental data, Supplemental Digital Content 1, http://links.lww.com/CCM/C375) and an unadjusted Kaplan-Meier analysis (p = 0.047) that supported this trend toward a difference in mortality over time (Fig. 2). Fifty patients in the liberal group and 66 patients in the restrictive group died in the ICU (p = 0.071).
In the TRICOP trial, we tested the hypothesis that in cancer patients with septic shock, the restrictive strategy of RBC transfusion would reduce mortality when compared with the liberal strategy. We observed a survival trend favoring the liberal transfusion strategy in the primary outcome of 28-day survival and an improved survival in the secondary endpoint of 90-day mortality again favoring the liberal transfusion strategy group. These results went in the opposite direction of our a priori hypothesis, of existing guidelines and of other trials in the field and should be considered of limited external generalizability and only hypothesis generating.
We focused on cancer patients because 15% of all ICU admissions due to septic shock are related to an underlying malignancy, because cancer patients are at high risk of anemia and transfusion and were not well represented in previous randomized trials (20, 21). Our hypothesis was that the restrictive RBC transfusion strategy would decrease 28-day mortality rate due to a reduced exposure to transfusion-related complications. The majority of enrolled patients developed anemia during ICU stay, and transfusion management was different between groups with few protocol deviations, and more RBC units given to the liberal group than to the restrictive group.
Our results are similar to those of a recent meta-analysis of randomized trials involving critically ill patients (22): clinical outcomes in patients who received transfusions with restrictive hemoglobin thresholds were similar to those in patients who received transfusion with liberal thresholds, and there was a trend toward an increased mortality in patients receiving a restrictive threshold in the perioperative period. A restrictive threshold for transfusion is usually preferred because it requires the use of fewer units of allogeneic red cells. However, the results of our secondary analyses (we observed a higher frequency of death in the restrictive group in 90 d) create concern regarding the safety of a restrictive strategy of RBC transfusion in cancer patients with septic shock. We should be cautious in interpreting the results because these were secondary analyses and because this finding had a fragility index (the number of deaths on which the statistical significance depends) of only 1 (23). We hypothesize that lower hemoglobin levels attributable to the restrictive threshold may have resulted in an increased number of deaths. However, we could not establish a cause-and-effect relationship analyzing the deaths. This hypothesis should be tested in future trials.
Even if current guidelines are suggesting the restrictive strategy for transfusion (13, 24), recent randomized trials suggested an increased mortality with this strategy at least in the specific setting of perioperative patients (25–27). These findings could be attributed to a better definition of specific subset of patients who may present harm when exposed to anemia (e.g., perioperative and oncologic patients may suffer more with the restrictive strategy).
Observational studies with a substantial number of patients and covariates had already suggested that patients with septic shock might benefit from blood transfusion. Park et al (28), in a propensity-matched analysis of a prospective observational database of 1,054 patients with severe sepsis and septic shock, reported that RBC transfusion was associated with a lower risk of 7-day, 28-day, and in-hospital mortality. Vincent et al (29) in the Sepsis Occurrence in Acutely Ill Patients propensity-matched study reported that transfused patients had a higher 30-day survival rate than nontransfused patients. In both studies, cancer patients accounted for approximately 15% of the participants.
Our trial might be compared with the Transfusion Requirements in Critical Care (TRICC) trial that randomized 838 ICU patients receiving RBC transfusion based on hemoglobin thresholds of 7.0 and 10 g/dL (9). No difference between groups was observed regarding 30-day mortality and organ dysfunction. The study concluded that the restrictive strategy of RBC transfusion was at least as safe as the liberal strategy in critically ill patients (9). Nevertheless, it is hard to transfer the results of the TRICC trial to a population with cancer and septic shock, since only a fourth of the participants had infection and the number of patients with cancer was not mentioned. In a more recent study, Holst et al (20) compared different hemoglobin transfusion thresholds (≤ 7 g/dL [lower threshold] vs ≤ 9 g/dL [higher threshold]) in 998 patients with septic shock (Transfusion Requirements in Septic Shock [TRISS] trial). There were no differences between groups regarding the primary outcome of 90-day mortality, ischemic events, and requirements of life support. No difference in 28-day mortality was noted. Although all patients had septic shock, only 7.5% of patients had a hematologic malignancy and the number of patients with solid tumor was not mentioned, limiting the external validity of the TRISS trial to cancer population.
Our study failed in proving the hypothesis that the restrictive strategy of RBC transfusion is superior to the liberal strategy. We observed no harm and a trend toward a survival benefit with an excess transfusion of a median of 1 unit of blood in the liberal group. Whether this was due to the reduced number of transfused units in the liberal group (61% of patients were transfused) or to the use of leukoreduced blood cannot be assessed, but results similar to ours were observed in the Functional Outcomes in Cardiovascular Patients Undergoing Surgical Hip Fracture Repair (FOCUS) trial (30) and in the TRISS trial (20), in which the majority of patients also received leukoreduced blood. The main limitation of our trial was our inability to keep healthcare staff unaware of the group assignments. However, the use of objective endpoints and the blinding of outcomes by researchers who were unaware of the group protected against bias. The nature of nonadherence to protocol was low and similar between groups, affecting the overall transfusion rate in only a small percentage of participants. Another limitation of our trial is that it was performed in a single referral center for cancer and had several exclusion criteria, which could compromise the generalizability of our findings. However, it reduces the noise. Differently from previous trials which included only patients with hemoglobin levels lower than 9 g/dL during ICU stay, our study included cancer patients with septic shock at ICU admission independently of hemoglobin levels. Our patients had similar levels of inclusion hemoglobin comparing to other studies—9.7 g/dL in the liberal and 9.6 g/dL in the restrictive group. We used a pragmatic randomized design to show the feasibility of implementation of ICU-wide transfusion policies for cancer patients with sepsis, resulting in a significant reduction in blood use without compromising clinical outcomes. Hemoglobin difference between groups after randomization was less than 1 g/dL. We also acknowledge that transfusion threshold trials are criticized for being “fixed,” that is, not allowing day-to-day variation in critically ill patient who gets sicker or somewhat better (more or fewer organ failures, on/off ventilator or pressors) over time.
In conclusion, the TRICOP trial showed a trend toward a survival benefit in the liberal transfusion threshold group when compared with the restrictive transfusion threshold group. Since these findings were unexpected, went in the opposite direction of the study hypothesis, and is not in agreement with current guidelines and expert opinion, the results should be considered hypothesis generating and will require to be confirmed in future trials.
1. Singer M, Deutschman CS, Seymour CW, et al. The Third International Consensus Definitions for Sepsis and Septic Shock (Sepsis-3). JAMA 2016; 315:801–810
2. de Montmollin E, Tandjaoui-Lambiotte Y, Legrand M, et al. Outcomes in critically ill cancer patients with septic shock of pulmonary origin. Shock 2013; 39:250–254
3. 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; 36:690–696
4. Regazzoni CJ, Irrazabal C, Luna CM, et al. Cancer patients with septic shock: Mortality predictors and neutropenia. Support Care Cancer 2004; 12:833–839
5. Rosolem MM, Rabello LS, Lisboa T, et al. Critically ill patients with cancer and sepsis: Clinical course and prognostic factors. J Crit Care 2012; 27:301–307
6. Danai PA, Moss M, Mannino DM, et al. The epidemiology of sepsis in patients with malignancy. Chest 2006; 129:1432–1440
7. Staudinger T, Pène F. Current insights into severe sepsis in cancer patients. Rev Bras Ter Intensiva 2014; 26:335–338
8. Mayr FB, Yende S, Angus DC. Epidemiology of severe sepsis. Virulence 2014; 5:4–11
9. Hébert PC, Wells G, Blajchman MA, et al. A multicenter, randomized, controlled clinical trial of transfusion requirements in critical care. Transfusion Requirements in Critical Care Investigators, Canadian Critical Care Trials Group. N Engl J Med 1999; 340:409–417
10. Hajjar LA, Vincent JL, Galas FR, et al. Transfusion requirements after cardiac surgery: The TRACS randomized controlled trial. JAMA 2010; 304:1559–1567
11. Villanueva C, Colomo A, Bosch A, et al. Transfusion strategies for acute upper gastrointestinal bleeding. N Engl J Med 2013; 368:11–21
12. Vincent JL. Indications for blood transfusions: Too complex to base on a single number? Ann Intern Med 2012; 157:71–72
13. Carson JL, Carless PA, Hebert PC. Transfusion thresholds and other strategies for guiding allogeneic red blood cell transfusion. Cochrane Database Syst Rev 2012; 4:CD002042
14. Chatterjee S, Wetterslev J, Sharma A, et al. Association of blood transfusion with increased mortality in myocardial infarction: A meta-analysis and diversity-adjusted study sequential analysis. JAMA Intern Med 2013; 173:132–139
15. Dellinger RP, Levy MM, Rhodes A, et al; Surviving Sepsis Campaign Guidelines Committee including the Pediatric Subgroup: Surviving sepsis campaign: International guidelines for management of severe sepsis and septic shock: 2012. Crit Care Med 2013; 41:580–637
16. Angus DC. The search for effective therapy for sepsis: Back to the drawing board? JAMA 2011; 306:2614–2615
17. Levy MM, Fink MP, Marshall JC, et al. 2001 SCCM/ESICM/ACCP/ATS/SIS International Sepsis Definitions Conference. Crit Care Med 2003; 31:1250–1256
18. Corwin HL, Gettinger A, Pearl RG, et al. The CRIT Study: Anemia and blood transfusion in the critically ill–current clinical practice in the United States. Crit Care Med 2004; 32:39–52
19. Vincent JL, Baron JF, Reinhart K, et al; ABC (Anemia and Blood Transfusion in Critical Care) Investigators: Anemia and blood transfusion in critically ill patients. JAMA 2002; 288:1499–1507
20. Holst LB, Haase N, Wetterslev J, et al; TRISS Trial Group; Scandinavian Critical Care Trials Group: Lower versus higher hemoglobin threshold for transfusion in septic shock. N Engl J Med 2014; 371:1381–1391
21. Soares M, Caruso P, Silva E, et al; Brazilian Research in Intensive Care Network (BRICNet): Characteristics and outcomes of patients with cancer requiring admission to intensive care units: A prospective multicenter study. Crit Care Med 2010; 38:9–15
22. Hovaguimian F, Myles PS. Restrictive versus liberal transfusion strategy in the perioperative and acute care settings: A context-specific systematic review and meta-analysis of randomized controlled trials. Anesthesiology 2016; 125:46–61
23. Ridgeon EE, Young PJ, Bellomo R, et al. The fragility index in multicenter randomized controlled critical care trials. Crit Care Med 2016; 44:1278–1284
24. Holst LB, Petersen MW, Haase N, et al. Restrictive versus liberal transfusion strategy for red blood cell transfusion: Systematic review of randomised trials with meta-analysis and trial sequential analysis. BMJ 2015; 350:h1354
25. de Almeida JP, Vincent JL, Galas FR, et al. Transfusion requirements in surgical oncology patients: A prospective, randomized controlled trial. Anesthesiology 2015; 122:29–38
26. Murphy GJ, Pike K, Rogers CA, et al; TITRe2 Investigators: Liberal or restrictive transfusion after cardiac surgery. N Engl J Med 2015; 372:997–1008
27. Gregersen M, Borris LC, Damsgaard EM. Postoperative blood transfusion strategy in frail, anemic elderly patients with hip fracture: The TRIFE randomized controlled trial. Acta Orthop 2015; 86:363–372
28. Park DW, Chun BC, Kwon SS, et al. Red blood cell transfusions are associated with lower mortality in patients with severe sepsis and septic shock: A propensity-matched analysis. Crit Care Med 2012; 40:3140–3145
29. Vincent JL, Sakr Y, Sprung C, et al; Sepsis Occurrence in Acutely Ill Patients (SOAP) Investigators: Are blood transfusions associated with greater mortality rates? Results of the Sepsis Occurrence in Acutely Ill Patients study. Anesthesiology 2008; 108:31–39
30. Carson JL, Terrin ML, Noveck H, et al; FOCUS Investigators: Liberal or restrictive transfusion in high-risk patients after hip surgery. N Engl J Med 2011; 365:2453–2462
critically ill oncology; intensive care; randomized controlled trial; transfusion