Center Effects in Hospital Mortality of Critically Ill Patients With Hematologic Malignancies

Development and Validation of a Prediction Model for 1-Year Mortality in Patients With a Hematologic Malignancy Admitted to the ICU

OBJECTIVES

To develop and validate a prediction model for 1-year mortality in patients with a hematologic malignancy acutely admitted to the ICU.

DESIGN

A retrospective cohort study.

SETTING

Five university hospitals in the Netherlands between 2002 and 2015.

PATIENTS

A total of 1097 consecutive patients with a hematologic malignancy were acutely admitted to the ICU for at least 24 h.

INTERVENTIONS

None.

MEASUREMENTS AND MAIN RESULTS

We created a 13-variable model from 22 potential predictors. Key predictors included active disease, age, previous hematopoietic stem cell transplantation, mechanical ventilation, lowest platelet count, acute kidney injury, maximum heart rate, and type of malignancy. A bootstrap procedure reduced overfitting and improved the model's generalizability. This involved estimating the optimism in the initial model and shrinking the regression coefficients accordingly in the final model. We assessed performance using internal-external cross-validation by center and compared it with the Acute Physiology and Chronic Health Evaluation II model. Additionally, we evaluated clinical usefulness through decision curve analysis. The overall 1-year mortality rate observed in the study was 62% (95% CI, 59-65). Our 13-variable prediction model demonstrated acceptable calibration and discrimination at internal-external validation across centers (C-statistic 0.70; 95% CI, 0.63-0.77), outperforming the Acute Physiology and Chronic Health Evaluation II model (C-statistic 0.61; 95% CI, 0.57-0.65). Decision curve analysis indicated overall net benefit within a clinically relevant threshold probability range of 60-100% predicted 1-year mortality.

CONCLUSIONS

Our newly developed 13-variable prediction model predicts 1-year mortality in hematologic malignancy patients admitted to the ICU more accurately than the Acute Physiology and Chronic Health Evaluation II model. This model may aid in shared decision-making regarding the continuation of ICU care and end-of-life considerations.

Safety and Effectiveness of Intensive Treatments Administered Outside the Intensive Care Unit to Hematological Critically Ill Patients: An Intensive Care without Walls Trial

Historically, the admission of hematological patients in the ICU shortly after the start of a critical illness is associated with better survival rates. Early intensive interventions administered by MET could play a role in the management of hematological critically ill patients, eventually reducing the ICU admission rate. In this retrospective and monocentric study, we evaluate the safety and effectiveness of intensive treatments administered by the MET in a medical ward frame. The administered interventions were mainly helmet CPAP and pharmacological cardiovascular support. Frequent reassessment by the MET at least every 8 to 12 h was guaranteed. We analyzed data from 133 hematological patients who required MET intervention. In-hospital mortality was 38%; mortality does not increase in patients not immediately transferred to the ICU. Only three patients died without a former admission to the ICU; in these cases, mortality was not related to the acute illness. Moreover, 37% of patients overcame the critical episode in the hematological ward. Higher SOFA and MEWS scores were associated with a worse survival rate, while neutropenia and pharmacological immunosuppression were not. The MET approach seems to be safe and effective. SOFA and MEWS were confirmed to be effective tools for prognostication.

[Invasive mechanical ventilation in patients with solid tumor or hematological malignancy]

Invasive mechanical ventilation in onco-hematology patients has become relatively routine, and is now part and parcel of their care pathway. Nevertheless, specific complications and subsequent therapeutic possibilities require discussion. To a greater extent than with regard to other patient populations, cooperation between specialist and ICU physician is mandatory, the objective being to more comprehensively assess a therapeutic project before or during the period of invasive mechanical ventilation. After an overview of recent results concerning ventilated patients in intensive care, this review aims to describe the specific complications and factors associated with mortality in this population.

Insights from patients screened but not randomised in the HYPERION trial

BACKGROUND

Few data are available about outcomes of patients screened for, but not enrolled in, randomised clinical trials.

METHODS

We retrospectively reviewed patients who had non-inclusion criteria for the HYPERION trial comparing 33 °C to 37 °C in patients comatose after cardiac arrest in non-shockable rhythm, due to any cause. A good neurological outcome was defined as a day-90 Cerebral Performance Category score of 1 or 2.

RESULTS

Of the 1144 patients with non-inclusion criteria, 1130 had day-90 information and, among these, 158 (14%) had good functional outcomes, compared to 7.9% overall in the HYPERION trial (10.2% with and 5.7% without hypothermia). Considerable centre-to-centre variability was found in the proportion of non-included patients who received hypothermia (0% to 83.8%) and who had good day-90 functional outcomes (0% to 31.3%). The proportion of patients with a good day-90 functional outcome was significantly higher with than without hypothermia (18.5% vs. 11.9%, P = 0.003).

CONCLUSION

Our finding of better functional outcomes without than with inclusion in the HYPERION trial, despite most non-inclusion criteria being of adverse prognostic significance (e.g., long no-flow and low-flow times and haemodynamic instability), raises important questions about the choice of patient selection criteria and the applicability of trial results to everyday practice. At present, reserving hypothermia for patients without predictors of poor prognosis seems open to criticism.

Oxygenation strategy during acute respiratory failure in immunocompromised patients

Acute respiratory failure (ARF) in immunocompromised patients remains challenging to treat. A large number of case require admission to intensive care unit (ICU) where mortality remains high. Oxygenation without intubation is important in this setting. This review summarizes recent studies assessing oxygenation devices for immunocompromised patients. Previous studies showed that non-invasive ventilation (NIV) has been associated with lower intubation and mortality rates. Indeed, in recent years, the outcomes of immunocompromised patients admitted to the ICU have improved. In the most recent randomized controlled trials, including immunocompromised patients admitted to the ICU with ARF, neither NIV nor high-flow nasal oxygen (HFNO) could reduce the mortality rate. In this setting, other strategies need to be tested to decrease the mortality rate. Early admission strategy and avoiding late failure of oxygenation strategy have been assessed in retrospective studies. However, objective criteria are still lacking to clearly discriminate time to admission or time to intubation. Also, diagnosis strategy may have an impact on intubation or mortality rates. On the other hand, lack of diagnosis has been associated with a higher mortality rate. In conclusion, improving outcomes in immunocompromised patients with ARF may include strategies other than the oxygenation strategy alone. This review discusses other unresolved questions to decrease mortality after ICU admission in such patients.

Critical illness in patients with hematologic malignancy: a population-based cohort study

PURPOSE

To describe the modern incidence and predictors of ICU admission for adult patients newly diagnosed with a hematologic malignancy.

METHODS

We conducted a population-based cohort study of adults with a new diagnosis of hematologic malignancy (April 1, 2006-March 31, 2017) in Ontario, Canada. We described the baseline demographic, clinical and laboratory predictors of ICU admission and subsequent mortality. The primary outcome was the incidence of ICU admission within 1 year of hematologic malignancy diagnosis. We assessed the predictors of ICU admission using Cox-proportional models that accounted for the competing risk of death and reported as subdistribution hazard ratios (sHR) with 95% confidence intervals (CI).

RESULTS

A total of 87,965 patients (mean [SD] age, 67.8 (15.7) years) were included. The 1-year incidence of ICU admission was 13.9% (median time 35 days), ranging from 7.3% (indolent lymphoma) to 22.5% (acute myeloid leukemia). After multivariable adjustment, compared to indolent lymphoma, acute myeloid leukemia (sHR, 3.09; 95% CI 2.84-3.35), aggressive non-Hodgkin lymphoma (sHR, 2.47; 95% CI 2.31-2.65) and acute lymphoblastic leukemia (sHR, 2.46; 95% CI 2.15-2.80) had the highest risk of ICU admission. Comorbidities such as cardiovascular disease (sHR, 2.09; 95% CI 2.01-2.19), chronic obstructive pulmonary disease (sHR, 1.33; 95% CI 1.26-1.39) and baseline laboratory abnormalities (anemia, thrombocytopenia and high creatinine) were also associated with ICU admission. Among ICU patients, 36.7% required invasive mechanical ventilation and in-hospital mortality was 31%.

CONCLUSION

Critical illness in patients with a newly diagnosed hematologic malignancy is frequent, occurring early after diagnosis. Certain baseline characteristics can help identify those patients at the highest risk.

Outcomes in 1096 patients with severe thrombotic thrombocytopenic purpura before the Caplacizumab era

INTRODUCTION

Thrombotic thrombocytopenic purpura (TTP) is a diagnostic and therapeutic emergency. Therapeutic plasma exchange (TPE) combined with immunosuppression has been the cornerstone of the initial management. To produce optimal benefits, emerging treatments must be used against a background of best standard of care. Clarifying current uncertainties is therefore crucial.

METHODS

The objective of this study was to analyze a large high-quality database (Marketscan) of TTP patients managed between 2005 and 2014, in the pre-caplacizumab era, in order to assess the impact of time to first TPE and use of first-line rituximab on mortality, and whether mortality declines over time.

RESULTS

Among the 1096 included patients (median age 46 [IQR 35-55], 70% female), 28.8% received TPE before day 2 in the ICU. Hospital mortality was 7.6% (83 deaths). Mortality was independently associated with older age (hazard ratio [HR], 1.024/year; 95% confidence interval [95%CI], [1.009-1.040]), diagnosis of sepsis (HR, 2.360; 95%CI [1.552-3.588]), and the need for mechanical ventilation (HR, 4.103; 95%CI, [2.749-6.126]). Factors independently associated with lower mortality were TPE at ICU admission (HR, 0.284; 95%CI, [0.112-0.717]), TPE within one day after ICU admission (HR, 0.449; 95%CI, [0.275-0.907]), and early rituximab therapy (HR, 0.229; 95% CI, [0.111-0.471]). Delayed TPE was associated with significantly higher costs.

CONCLUSIONS

Immediate TPE and early rituximab are associated with improved survival in TTP patients. Improved treatments have led to a decline in mortality over time, and alternate outcome variables such as the use of hospital resources or longer term outcomes therefore need to be considered.

A 12-year study evaluating the outcomes and predictors of mortality in critically ill cancer patients admitted with septic shock

Background

Though sepsis is common in patients with cancer, there are limited studies that evaluated sepsis and septic shock in this patient population. The objective of this study was to evaluate the outcomes and to identify predictors of mortality in cancer patients admitted to the intensive care unit (ICU) with septic shock.

Methods

This was a retrospective study conducted at a medical-surgical oncologic ICU of a comprehensive cancer center. Adult cancer patients admitted with septic shock between January 1, 2008 and December 31, 2019 were enrolled. Septic shock was defined as an ICU admission diagnosis of sepsis that required initiating vasopressors within 24 h of admission. Patient baseline characteristics, ICU length of stay and ICU and hospital mortality were recorded. Univariate analysis and logistic regression were performed to identify predictors associated with ICU and hospital mortality.

Results

During the study period, 1408 patients met the inclusion criteria. The mean age was 56.8 ± 16.1 (SD) years and mean Acute Physiology and Chronic Health Evaluation (APACHE) II was 23.0 ± 7.91 (SD). Among the enrolled patients, 67.8% had solid tumors while the remaining had hematological malignancies. Neutropenia and thrombocytopenia were reported in 19.3 and 39.5% of the patients, respectively, and mechanical ventilation was required for 42% of the patients. Positive cultures were reported in 836 (59.4%) patients, most commonly blood (33%) and respiratory (26.6%). Upon admission, about half the patients had acute kidney injury, while elevated total bilirubin and lactic acid levels were reported in 13.8 and 65.2% of the patients, respectively. The median ICU length of stay was 4 days (IQR 3–8), and ICU and hospital mortality were reported in 688 (48.9%) and 914 (64.9%) patients, respectively. Mechanical ventilation, APACHE II, thrombocytopenia, positive cultures, elevated bilirubin and lactic acid levels were significantly associated with both ICU and hospital mortality.

Conclusions

In a relatively large cohort of patients with solid and hematological malignancies admitted to the ICU with septic shock, hospital mortality was reported in about two-third of the patients. Mechanical ventilation, APACHE II, thrombocytopenia, positive cultures, elevated bilirubin and lactic acid levels were significant predictors of mortality.

Sepsis and Septic Shock in Patients With Malignancies: A Groupe de Recherche Respiratoire en Réanimation Onco-Hématologique Study

OBJECTIVES

Cancer affects up to 20% of critically ill patients, and sepsis is one of the leading reasons for ICU admission in this setting. Early signals suggested that survival might be increasing in this population. However, confirmation studies have been lacking. The goal of this study was to assess trends in survival rates over time in cancer patients admitted to the ICU for sepsis or septic shock over the last 2 decades.

DATA SOURCE

Seven European ICUs.

STUDY SELECTION

A hierarchical model taking into account the year of admission and the source dataset as random variables was used to identify risk factors for day 30 mortality.

DATA EXTRACTION

Data from cancer patients admitted to ICUs for sepsis or septic shock were extracted from the Groupe de Recherche Respiratoire en Réanimation Onco-Hématologique database (1994-2015).

DATA SYNTHESIS

Overall, 2,062 patients (62% men, median [interquartile range] age 59 yr [48-67 yr]) were included in the study. Underlying malignancies were solid tumors (n = 362; 17.6%) or hematologic malignancies (n = 1,700; 82.4%), including acute leukemia (n = 591; 28.7%), non-Hodgkin lymphoma (n = 461; 22.3%), and myeloma (n = 244; 11.8%). Two-hundred fifty patients (12%) underwent allogeneic hematopoietic stem cell transplantation and 640 (31.0%) were neutropenic at ICU admission. Day 30 mortality was 39.9% (823 deaths). The year of ICU admission was associated with significant decrease in day 30 mortality over time (odds ratio, 0.96; 95% CI, 0.93-0.98; p = 0.001). Mechanical ventilation (odds ratio, 3.25; 95% CI, 2.52-4.19; p < 0.01) and vasopressors use (odds ratio, 1.42; 95% CI, 1.10-1.83; p < 0.01) were independently associated with day 30 mortality, whereas underlying malignancy, allogeneic hematopoietic stem cell transplantation, and neutropenia were not.

CONCLUSIONS

Survival in critically ill oncology and hematology patients with sepsis improved significantly over time. As outcomes improve, clinicians should consider updating admission policies and goals of care in this population.

Diagnostic Performance of Hemophagocytic Lymphohistiocytosis Criteria and HScore in Critically Ill Patients With Severe Hemophagocytic Syndrome

OBJECTIVES

To assess whether critically ill hematologic patients without diagnosis of hemophagocytic lymphohistiocytosis may have features mimicking hemophagocytic lymphohistiocytosis according to both diagnostic scores.

DESIGN

A retrospective case-control study.

SETTING

Hemophagocytic syndrome diagnosis was standardized and based on a consensual diagnosis by at least two experts of a university hospital which is a reference center for hemophagocytic syndrome.

PATIENTS

Cases (hemophagocytic syndrome+) consisted in a group of consecutive patients (n = 150) admitted in our ICU between 2007 and 2018. Control group (hemophagocytic syndrome-) consisted in patients included in a prospective multicenter cohort of hematologic patients in whom three independent experts ruled out the diagnosis of hemophagocytic syndrome (n = 1011).

MEASUREMENTS AND MAIN RESULTS

Overall, 1,161 patients were included. Hospital mortality was 45.8% in hemophagocytic syndrome- patients (n = 66) and 38.8% in control patients (n = 392; p = 0.126). Median HScore was 235 (205-262) in hemophagocytic syndrome+ and 42 (18-62) in hemophagocytic syndrome- patients (p < 0.001); number of hemophagocytic lymphohistiocytosis criteria was 4 (4-5) vs 1 (0-1), respectively (p < 0.001). Diagnostic performances of both scores were excellent with area under receiver operating characteristic curve of 0.99 (95% CI, 0.99-0.99) and 0.99 (95% CI, 0.99-0.99) for hemophagocytic lymphohistiocytosis and HScore, respectively. After propensity score matching (n = 144 × 2), the median HScore was 234 (205-262) in hemophagocytic syndrome+ patients versus 49 (18-71) in hemophagocytic syndrome- patients (p < 0.001). Median number of hemophagocytic lymphohistiocytosis criteria was 4 (4-5) in hemophagocytic syndrome+ and 1 (0-1) in hemophagocytic syndrome- patients (p < 0.001). Area under receiver operating characteristic curve was then of 0.98 (95% CI, 0.96-0.99) for hemophagocytic lymphohistiocytosis criteria and 0.99 (95% CI, 0.99-1) for HScore.

CONCLUSIONS

In ICU patients, several conditions share some similarities with hemophagocytic syndrome, explaining the poor predictive value of isolated biological markers such as ferritin level. Despite these potential confounding factors, our study suggests HScore and hemophagocytic lymphohistiocytosis criteria to be highly discriminant identifying hemophagocytic syndrome in critically ill patients.

Center effect in intubation risk in critically ill immunocompromised patients with acute hypoxemic respiratory failure

BACKGROUND

Acute respiratory failure is the leading reason for intensive care unit (ICU) admission in immunocompromised patients, and the need for invasive mechanical ventilation has become a major clinical endpoint in randomized controlled trials (RCTs). However, data are lacking on whether intubation is an objective criteria that is used unbiasedly across centers. This study explores how this outcome varies across ICUs.

METHODS

Hierarchical models and permutation procedures for testing multiple random effects were applied on both data from an observational cohort (the TRIAL-OH study: 703 patients, 17 ICUs) and a randomized controlled trial (the HIGH trial: 776 patients, 31 ICUs) to characterize ICU variation in intubation risk across centers.

RESULTS

The crude intubation rate varied across ICUs from 29 to 80% in the observational cohort and from 0 to 86% in the RCT. This center effect on the mean ICU intubation rate was statistically significant, even after adjustment on individual patient characteristics (observational cohort: p value = 0.013, median OR 1.48 [1.30-1.72]; RCT: p value 0.004, median OR 1.51 [1.36-1.68]). Two ICU-level characteristics were associated with intubation risk (the annual rate of intubation procedure per center and the time from respiratory symptoms to ICU admission) and could partly explain this center effect. In the RCT that controlled for the use of high-flow oxygen therapy, we did not find significant variation in the effect of oxygenation strategy on intubation risk across centers, despite a significant variation in the need for invasive mechanical ventilation.

CONCLUSION

Intubation rates varied considerably among ICUs, even after adjustment on individual characteristics.