Search

Background: Moderate to severe depressive symptoms occur in up to one-third of patients at 1 year following ICU discharge, negatively affecting patient outcomes. This study evaluated patient and caregiver factors associated with the development of these symptoms.Methods: This study used the Rehabilitation and Recovery in Patients after Critical Illness and Their Family Caregivers (RECOVER) Program (Phase 1) cohort of 391 patients from 10 medical/surgical university-affiliated ICUs across Canada. We determined the association between patient depressive symptoms (captured by using the Beck Depression Inventory II [BDI-II]), patient characteristics (age, sex, socioeconomic status, Charlson score, and ICU length of stay [LOS]), functional independence measure (FIM) motor subscale score, and caregiver characteristics (Caregiver Assistance Scale and Center for Epidemiologic Studies-Depression Scale) by using linear mixed models at time points 3, 6, and 12 months.Results: BDI-II data were available for 246 patients. Median age at ICU admission was 56 years (interquartile range, 45-65 years), 143 (58%) were male, and median ICU LOS was 19 days (interquartile range, 13-32 days). During the 12-month follow-up, 67 of 246 (27.2%) patients had a BDI-II score ≥ 20, indicating moderate to severe depressive symptoms. Mixed models showed worse depressive symptoms in patients with lower FIM motor subscale scores (1.1 BDI-II points per 10 FIM points), lower income status (by 3.7 BDI-II points; P = .007), and incomplete secondary education (by 3.8 BDI-II points; P = .009); a curvilinear relation with age (P = .001) was also reported, with highest BDI-II at ages 45 to 50 years. No associations were found between patient BDI-II and comorbidities (P = .92), sex (P = .25), ICU LOS (P = .51), or caregiver variables (Caregiver Assistance Scale [P = .28] and Center for Epidemiologic Studies Depression Scale [P = .74]).Conclusions: Increased functional dependence, lower income, and lower education are associated with increased severity of post-ICU depressive symptoms, whereas age has a curvilinear relation with symptom severity. Knowledge of risk factors may inform surveillance and targeted mental health follow-up. Early mobilization and rehabilitation aiming to improve function may serve to modify mood disorders. [ABSTRACT FROM AUTHOR]

Background: An increased risk of complications, including death, has been associated with stored red blood cell (RBC) units in observational studies but not in randomized trials. We aimed to evaluate for volume-dependent effects attributable to length of RBC storage in a secondary analysis of the Age of Blood Evaluation (ABLE) trial.Study Design and Methods: In the 2510 critically ill adults from the ABLE trial randomized to receive RBC units stored not more than 7 days or the oldest compatible RBC units, we estimated the hazard ratio (HR) for death by intensive care unit (ICU) and hospital discharge and by days 28, 90, and 180, within subgroups defined by the number of RBC units received. Extended Cox proportional hazards regression was used to model the HR.Results: A volume-dependent effect of storage age on survival was present for death by 90 and 180 days, but not earlier endpoints. The HR for death by 90 days was 0.55 (95% confidence interval [CI], 0.11-0.98, fresh vs standard) after transfusion of 6 RBC units but 1.45 (95% CI, 1.06-1.98) after transfusion of 1 RBC unit.Conclusion: In this exploratory analysis, volume-dependent effects related to RBC storage were documented in the ABLE trial. The harms associated with small volumes of fresh RBC units and large volumes of older RBC units should be further explored. [ABSTRACT FROM AUTHOR]

Background: Blood transfusion is common in the resuscitation of patients with traumatic injury. However, the clinical impact of the length of storage of transfused blood is unclear in this population.Study Design and Methods: We undertook a prespecified nested analysis of 372 trauma victims of the 2510 critically ill patients from 64 centers treated as part of the Age of Blood Evaluation (ABLE) randomized controlled trial. Patients were randomized according to their trauma status to receive either a transfusion of fresh blood stored not more than 7 days or standard-issue blood. Our primary outcome was 90-day all-cause mortality.Results: Overall, 186 trauma patients received fresh blood and 186 received standard-issue blood. Adherence to transfusion protocol was 94% (915/971) for all fresh blood transfused and 100% (753/753) for all standard-issue blood transfused. Mean ± SD blood storage duration was 5.6 ± 3.8 days in the fresh group and 22.7 ± 8.4 days in the standard-issue group (p < 0.001). Ninety-day mortality in the fresh group was 21% (38/185), compared to 16% (29/184) in the standard-issue group, with an unadjusted absolute risk difference of 5% (95% confidence interval [CI], -3.1 to 12.6) and an adjusted absolute risk difference of 2% (95% CI, -3.5 to 6.8).Conclusion: In critically ill trauma patients, transfusion of fresh blood did not decrease 90-day mortality or secondary outcomes, a finding similar to the overall population of the ABLE trial. [ABSTRACT FROM AUTHOR]

Purpose: The intensity of care provided to critically ill patients has been shown to be associated with mortality. In patients with traumatic brain injury (TBI), specialized neurocritical care is often required, but whether it affects clinically significant outcomes is unknown. We aimed to determine the association of the intensity of care on mortality and the incidence of withdrawal of life-sustaining therapies in critically ill patients with severe TBI.Methods: We conducted a post hoc analysis of a multicentre retrospective cohort study of critically ill adult patients with severe TBI. We defined the intensity of care as a daily cumulative sum of interventions during the intensive care unit stay. Our outcome measures were all-cause hospital mortality and the incidence of withdrawal of life-sustaining therapies.Results: Seven hundred sixteen severe TBI patients were included in our study. Most were male (77%) with a mean (standard deviation) age of 42 (20.5) yr and a median [interquartile range] Glasgow Coma Scale score of 3 [3-6]. Our results showed an association between the intensity of care and mortality (hazard ratio [HR], 0.69; 95% confidence interval [CI], 0.63 to 0.74) and the incidence of withdrawal of life-sustaining therapy (HR, 0.73; 95% CI, 0.67 to 0.79).Conclusion: In general, more intense care was associated with fewer deaths and a lower incidence of withdrawal of life-sustaining therapies in critically ill patients with severe TBI. [ABSTRACT FROM AUTHOR]

Importance: There is limited evidence that the use of severity of illness scores in pediatric patients can facilitate timely admission to the intensive care unit or improve patient outcomes.Objective: To determine the effect of the Bedside Paediatric Early Warning System (BedsidePEWS) on all-cause hospital mortality and late admission to the intensive care unit (ICU), cardiac arrest, and ICU resource use.Design, Setting, and Participants: A multicenter cluster randomized trial of 21 hospitals located in 7 countries (Belgium, Canada, England, Ireland, Italy, New Zealand, and the Netherlands) that provided inpatient pediatric care for infants (gestational age ≥37 weeks) to teenagers (aged ≤18 years). Participating hospitals had continuous physician staffing and subspecialized pediatric services. Patient enrollment began on February 28, 2011, and ended on June 21, 2015. Follow-up ended on July 19, 2015.Interventions: The BedsidePEWS intervention (10 hospitals) was compared with usual care (no severity of illness score; 11 hospitals).Main Outcomes and Measures: The primary outcome was all-cause hospital mortality. The secondary outcome was a significant clinical deterioration event, which was defined as a composite outcome reflecting late ICU admission. Regression analyses accounted for hospital-level clustering and baseline rates.Results: Among 144 539 patient discharges at 21 randomized hospitals, there were 559 443 patient-days and 144 539 patients (100%) completed the trial. All-cause hospital mortality was 1.93 per 1000 patient discharges at hospitals with BedsidePEWS and 1.56 per 1000 patient discharges at hospitals with usual care (adjusted between-group rate difference, 0.01 [95% CI, -0.80 to 0.81 per 1000 patient discharges]; adjusted odds ratio, 1.01 [95% CI, 0.61 to 1.69]; P = .96). Significant clinical deterioration events occurred during 0.50 per 1000 patient-days at hospitals with BedsidePEWS vs 0.84 per 1000 patient-days at hospitals with usual care (adjusted between-group rate difference, -0.34 [95% CI, -0.73 to 0.05 per 1000 patient-days]; adjusted rate ratio, 0.77 [95% CI, 0.61 to 0.97]; P = .03).Conclusions and Relevance: Implementation of the Bedside Paediatric Early Warning System compared with usual care did not significantly decrease all-cause mortality among hospitalized pediatric patients. These findings do not support the use of this system to reduce mortality.Trial Registration: clinicaltrials.gov Identifier: NCT01260831. [ABSTRACT FROM AUTHOR]

Background: Children with traumatic brain injury (TBI) are frequently at risk of long-term impairments of attention and executive functioning but these problems are difficult to predict. Although deficits have been reported to vary with injury severity, age at injury and sex, prognostication of outcome remains imperfect at a patient-specific level. The objective of this proof of principle study was to evaluate a variety of patient variables, along with six brain-specific and inflammatory serum protein biomarkers, as predictors of long-term cognitive outcome following paediatric TBI.Method: Outcome was assessed in 23 patients via parent-rated questionnaires related to attention deficit hyperactivity disorder (ADHD) and executive functioning, using the Conners 3rd Edition Rating Scales (Conners-3) and Behaviour Rating Inventory of Executive Function (BRIEF) at a mean time since injury of 3.1 years. Partial least squares (PLS) analyses were performed to identify factors measured at the time of injury that were most closely associated with outcome on (1) the Conners-3 and (2) the Behavioural Regulation Index (BRI) and (3) Metacognition Index (MI) of the BRIEF.Results: Higher levels of neuron specific enolase (NSE) and lower levels of soluble neuron cell adhesion molecule (sNCAM) were associated with higher scores on the inattention, hyperactivity/impulsivity and executive functioning scales of the Conners-3, as well as working memory and initiate scales of the MI from the BRIEF. Higher levels of NSE only were associated with higher scores on the inhibit scale of the BRI.Conclusions: NSE and sNCAM show promise as reliable, early predictors of long-term attention-related and executive functioning problems following paediatric TBI. [ABSTRACT FROM AUTHOR]

Rationale: Recruitment in critical care research differs from other contexts in important ways: patients lack decision-making capacity, uncertainty exists regarding patient prognosis, and critical illnesses are often associated with appreciable morbidity and mortality.Objectives: We aimed to describe the experiences of surrogate decision makers (SDMs) in being approached for consent for critically ill patients to participate in research.Methods: A multicenter, qualitative study involving semistructured interviews with 26 SDMs, who provided or declined surrogate consent for research participation, at 5 Canadian centers nested within a multicenter observational study of research recruitment practices. Transcripts were reviewed by three qualitative researchers, and data were analyzed using grounded theory and a narrative critical analysis.Measurements and Main Results: SDMs were guided by an overarching desire for the patient to live. Surrogate research decision-making involved three sequential stages: (1) being approached; (2) reflecting on participation; and (3) making a decision. In stage 1, SDMs identified factors (their expectations, how they were approached, the attributes of the person approaching, and study risks and benefits) that characterized their consent encounter and affirmed a preference to be approached in person. If SDMs perceived the risk of participation to be too high or felt patients may not benefit from participation, they did not contemplate further. In stage 2, SDMs who knew the patient's wishes or had a deeper understanding of research prioritized the patient's wishes and the perceived benefits of participation. Without this information, SDMs prioritized obtaining more and better care for the patient, considered what was in their mutual best interests, and valued healthcare professional's knowledge. Trust in healthcare professionals was essential to proceeding further. In stage 3, SDMs considered six factors in rendering decisions.Conclusions: SDMs engaged in three sequential stages and considered six factors in making surrogate decisions for research participation. Surrogates' assessments of the risks and benefits of participation and their trust in healthcare professionals were critical factors in research decision-making. By conceptualizing surrogate decision-making for research in stages, future research can develop and test procedures to enhance the surrogate research decision-making process. [ABSTRACT FROM AUTHOR]

Background: Mechanically ventilated patients may receive more sedation during the night than during the day, potentially delaying extubation. We compared nighttime and daytime benzodiazepine and opioid administration in adult patients enrolled in a multicenter sedation trial comparing protocolized sedation alone or protocolized sedation combined with daily sedation interruption; and we evaluated whether nighttime and daytime doses were associated with liberation from mechanical ventilation.Methods: This is a secondary analysis of a randomized trial which was conducted in 16 North American medical-surgical ICUs. In all 423 patients, nurses applied a validated sedation scale hourly to titrate benzodiazepine and opioid infusions to achieve a light level of sedation. Using fentanyl equivalents and midazolam equivalents, we compared dosages administered during night (19:00 to 07:00) and day (07:00 to 19:00) shifts. Using multivariable logistic regression we evaluated the association between nighttime and daytime opioid and sedative doses, and spontaneous breathing trial (SBT) conduct, SBT success, and extubation.Results: Nighttime benzodiazepine and opioid doses were significantly higher than daytime doses (mean difference midazolam equivalents 23.3 mg, 95 % CI 12.9, 33.8, p < 0.0001; mean difference fentanyl equivalents 356 mcg, 95 % CI 130, 582, p = 0.0021). Mean Sedation Agitation Scale score was similar between night and day, and was at target (3.2 vs 3.3, 95 % CI -0.05, 0.02, p = 0.35). Self-reported nurse workload was similar during the night and day. Patients were more often restrained during day shifts (76.3 % vs 73.7 %, p < 0.0001), and there were more unintentional device removals during the day compared with night (15.9 % vs 9.1 %, p < 0.0001). Increases in nighttime drug doses were independently associated with failure to meet SBT screening criteria, SBT failure, and the decision not to extubate the patient despite successful SBT.Conclusion: Patients received higher doses of opioids and benzodiazepines at night. Higher nighttime doses were associated with SBT failure and delayed extubation.Trial Registration: ClinicalTrials.gov NCT00675363 . Registered 7 May 2008. [ABSTRACT FROM AUTHOR]

Background: Despite the high mortality in patients with pneumonia admitted to an ICU, data on risk factors for death remain limited.Methods: In this secondary analysis of PROTECT (Prophylaxis for Thromboembolism in Critical Care Trial), we focused on the patients admitted to ICU with a primary diagnosis of pneumonia. The primary outcome for this study was 90-day hospital mortality and the secondary outcome was 90-day ICU mortality. Cox regression model was conducted to examine the relationship between baseline and time-dependent variables and hospital and ICU mortality.Results: Six hundred sixty seven patients admitted with pneumonia (43.8 % females) were included in our analysis, with a mean age of 60.7 years and mean APACHE II score of 21.3. During follow-up, 111 patients (16.6 %) died in ICU and in total, 149 (22.3 %) died in hospital. Multivariable analysis demonstrated significant independent risk factors for hospital mortality including male sex (hazard ratio (HR) = 1.5, 95 % confidence interval (CI): 1.1 - 2.2, p-value = 0.021), higher APACHE II score (HR = 1.2, 95 % CI: 1.1 - 1.4, p-value < 0.001 for per-5 point increase), chronic heart failure (HR = 2.9, 95 % CI: 1.6 - 5.4, p-value = 0.001), and dialysis (time-dependent effect: HR = 2.7, 95 % CI: 1.3 - 5.7, p-value = 0.008). Higher APACHE II score (HR = 1.2, 95 % CI: 1.1 - 1.4, p-value = 0.002 for per-5 point increase) and chronic heart failure (HR = 2.6, 95 % CI: 1.3 - 5.0, p-value = 0.004) were significantly related to risk of death in the ICU.Conclusion: In this study using data from a multicenter thromboprophylaxis trial, we found that male sex, higher APACHE II score on admission, chronic heart failure, and dialysis were independently associated with risk of hospital mortality in patients admitted to ICU with pneumonia. While high illness severity score, presence of a serious comorbidity (heart failure) and need for an advanced life support (dialysis) are not unexpected risk factors of mortality, male sex might necessitate further exploration. More studies are warranted to clarify the effect of these risk factors on survival in critically ill patients admitted to ICU with pneumonia.Trial Registration: ClinicalTrials.gov Identifier: NCT00182143 . [ABSTRACT FROM AUTHOR]

Background: Physicians often administer corticosteroids for the treatment of fluid and vasoactive infusion dependent pediatric shock. This use of corticosteroids is controversial, however, and has never been studied in a pediatric randomized controlled trial (RCT). This pilot trial will determine the feasibility of a larger RCT on the role of corticosteroids in pediatric shock.Methods/design: Steroids in Fluid and/or Vasoactive Infusion Dependent Pediatric Shock (STRIPES) is a pragmatic, seven-center, double-blind, pilot RCT. We aim to randomize 72 pediatric patients with fluid and vasoactive infusion dependent shock to receive either hydrocortisone or a saline placebo for 7 days or until clinical stability, whichever occurs first. The primary outcome of this pilot trial is the feasibility of recruitment, defined as the number of patients enrolled over a 1-year period. Secondary outcomes include the frequency of, and reasons for, open-label steroid use, protocol adherence, incidence of mortality and corticosteroid-associated adverse events, time to discontinuation of inotropes, and feasibility of blood sampling.Discussion: Corticosteroids are used for the treatment of pediatric shock without sufficient evidence to support this practice. While there is a scientific rationale and limited data supporting their use in this setting, there is also evidence from other populations suggesting potential harm. The STRIPES pilot study will assess the feasibility of a larger, much needed trial powered for clinically important outcomes.Trial Registration: ClinicalTrials.gov: NCT02044159. [ABSTRACT FROM AUTHOR]

Background: Patients with 2009 pandemic influenza A(H1N1pdm09)-related critical illness were frequently treated with systemic corticosteroids. While observational studies have reported significant corticosteroid-associated mortality after adjusting for baseline differences in patients treated with corticosteroids or not, corticosteroids have remained a common treatment in subsequent influenza outbreaks, including avian influenza A(H7N9). Our objective was to describe the use of corticosteroids in these patients and investigate predictors of steroid prescription and clinical outcomes, adjusting for both baseline and time-dependent factors.Methods: In an observational cohort study of adults with H1N1pdm09-related critical illness from 51 Canadian ICUs, we investigated predictors of steroid administration and outcomes of patients who received and those who did not receive corticosteroids. We adjusted for potential baseline confounding using multivariate logistic regression and propensity score analysis and adjusted for potential time-dependent confounding using marginal structural models.Results: Among 607 patients, corticosteroids were administered to 280 patients (46.1%) at a median daily dose of 227 (interquartile range, 154-443) mg of hydrocortisone equivalents for a median of 7.0 (4.0-13.0) days. Compared with patients who did not receive corticosteroids, patients who received corticosteroids had higher hospital crude mortality (25.5% vs 16.4%, p = 0.007) and fewer ventilator-free days at 28 days (12.5 ± 10.7 vs 15.7 ± 10.1, p < 0.001). The odds ratio association between corticosteroid use and hospital mortality decreased from 1.85 (95% confidence interval 1.12-3.04, p = 0.02) with multivariate logistic regression, to 1.71 (1.05-2.78, p = 0.03) after adjustment for propensity score to receive corticosteroids, to 1.52 (0.90-2.58, p = 0.12) after case-matching on propensity score, and to 0.96 (0.28-3.28, p = 0.95) using marginal structural modeling to adjust for time-dependent between-group differences.Conclusions: Corticosteroids were commonly prescribed for H1N1pdm09-related critical illness. Adjusting for only baseline between-group differences suggested a significant increased risk of death associated with corticosteroids. However, after adjusting for time-dependent differences, we found no significant association between corticosteroids and mortality. These findings highlight the challenges and importance in adjusting for baseline and time-dependent confounders when estimating clinical effects of treatments using observational studies. [ABSTRACT FROM AUTHOR]

Objective: To examine the incidence and predictors of clinician discomfort with life support plans for ICU patients.Design and Setting: Prospective cohort in 13 medical-surgical ICUs in four countries.Patients: 657 mechanically ventilated adults expected to stay in ICU at least 72 h.Measurements and Results: Daily we documented the life support plan for mechanical ventilation, inotropes and dialysis, and clinician comfort with these plans. If uncomfortable, clinicians stated whether the plan was too technologically intense (the provision of too many life support modalities or the provision of any modality for too long) or not intense enough, and why. At least one clinician was uncomfortable at least once for 283 (43.1%) patients, primarily because plans were too technologically intense rather than not intense enough (93.9% vs. 6.1%). Predictors of discomfort because plans were too intense were: patient age, medical admission, APACHE II score, poor prior functional status, organ dysfunction, dialysis in ICU, plan to withhold dialysis, plan to withhold mechanical ventilation, first week in the ICU, clinician, and city.Conclusions: Clinician discomfort with life support perceived as too technologically intense is common, experienced mostly by nurses, variable across centers, and is more likely for older, severely ill medical patients, those with acute renal failure, and patients lacking plans to forgo reintubation and ventilation. Acknowledging the sources of discomfort could improve communication and decision making. [ABSTRACT FROM AUTHOR]

Background: Current pediatric septic shock resuscitation guidelines from the American College of Critical Care Medicine focus on the early and goal-directed administration of intravascular fluid followed by vasoactive medication infusions for persistent and fluid-refractory shock. However, accumulating adult and pediatric data suggest that excessive fluid administration is associated with worse patient outcomes and even increased risk of death. The optimal amount of intravascular fluid required in early pediatric septic shock resuscitation prior to the initiation of vasoactive support remains unanswered.Methods/design: The SQUEEZE Pilot Trial is a pragmatic, two-arm, parallel-group, open-label, prospective pilot randomized controlled trial. Participants are children aged 29 days to under 18 years with suspected or confirmed septic shock and a need for ongoing resuscitation. Eligible participants are enrolled under an exception to consent process and randomly assigned via concealed allocation to either the Usual Care (control) or Fluid Sparing (intervention) resuscitation strategy. The primary objective of this pilot trial is to determine feasibility, based on the ability to enroll participants and to adhere to the study protocol. The primary outcome measure by which success will be determined is participant enrollment rate ("pass" defined as at least two participants/site/month, recognizing that enrollment may be slower during the run-in phase). Secondary objectives include assessing (1) appropriateness of eligibility criteria, and (2) completeness of clinical outcomes to inform the endpoints for the planned multisite trial. To support the nested translational study, SQUEEZE-D, we will also evaluate the feasibility of describing cell-free DNA (a procoagulant molecule with prognostic utility) in blood samples obtained from children enrolled into the SQUEEZE Pilot Trial at baseline and at 24 h.Discussion: The optimal degree of fluid resuscitation and the timing of initiation of vasoactive support in order to achieve recommended therapeutic targets in children with septic shock remains unanswered. No prospective study to date has examined this important question for children in developed countries including Canada. Recruitment for the SQUEEZE Pilot Trial opened on 6 January 2014. Findings will inform the feasibility of the planned multicenter trial to answer our overall research question.Trial Registration: ClinicalTrials.gov Identifier NCT01973907 , registered on 23 October 2013. [ABSTRACT FROM AUTHOR]

Introduction: A decreased frequency of upper gastrointestinal bleeding and a possible association of proton pump inhibitor use with Clostridium difficile and ventilator-associated pneumonia have raised concerns recently. The Reevaluating the Inhibition of Stress Erosions Pilot Trial determined the feasibility of undertaking a larger trial investigating the efficacy and safety of withholding proton pump inhibitors in critically ill patients.Methods: In 10 ICUs, we randomized adult ICU patients anticipated to be mechanically ventilated for greater than or equal to 48 hours to receive 40 mg of IV pantoprazole daily or placebo. We excluded patients who had acute or recent gastrointestinal bleed, used dual antiplatelet agents, had a medical condition requiring proton pump inhibitor treatment, or had already received more than one dose of acid suppression daily. Patients, families, clinicians, and research staff were blinded. We conducted a systematic review and meta-analysis of similar trials.Main Results: Ninety-one patients (49 pantoprazole and 42 placebo) from 10 centers in Canada, Saudi Arabia, and Australia were enrolled. All feasibility goals were met: 1) recruitment rate was 2.6 patients per month; 2) consent rate was 77.8%; and 3) protocol adherence was 97.7%. Upper gastrointestinal bleeding developed in 6.1% of patients in the pantoprazole group and 4.8% in the placebo group (p = 1.0). Ventilator-associated pneumonia developed in 20.4% of patients in the pantoprazole group and 14.3% in the placebo group (p = 0.58). C. difficile was identified in 4.1% pantoprazole patients and in 2.4% placebo patients (p = 1.0). We meta-analyzed five trials (604 patients) of proton pump inhibitors versus placebo; there was no statistically significant difference in the risk of upper gastrointestinal bleeding, infections, or mortality.Conclusions: Our results support the feasibility of a larger trial to evaluate the safety of withholding stress ulcer prophylaxis. Although the results are imprecise, there was no alarming increase in the risk of upper gastrointestinal bleeding; the effect of proton pump inhibitors on ventilator-associated pneumonia and C. difficile remain unclear. [ABSTRACT FROM AUTHOR]

Objectives: The optimum duration of antimicrobial treatment for patients with bacteremia is unknown. Our objectives were to determine duration of antimicrobial treatment provided to patients who have bacteremia in ICUs, to assess pathogen/patient factors related to treatment duration, and to assess the relationship between treatment duration and survival.Design: Retrospective cohort study.Settings: Fourteen ICUs across Canada.Patients: Patients with bacteremia and were present in the ICU at the time culture reported positive.Interventions: Duration of antimicrobial treatment for patients who had bacteremia in ICU.Measurements and Main Results: Among 1,202 ICU patients with bacteremia, the median duration of treatment was 14 days, but with wide variability (interquartile range, 9-17.5). Most patient characteristics were not associated with treatment duration. Coagulase-negative staphylococci were the only pathogens associated with shorter treatment (odds ratio, 2.82; 95% CI, 1.51-5.26). The urinary tract was the only source of infection associated with a trend toward lower likelihood of shorter treatment (odds ratio, 0.67; 95% CI, 0.42-1.08); an unknown source of infection was associated with a greater likelihood of shorter treatment (odds ratio, 2.14; 95% CI, 1.17-3.91). The association of treatment duration and survival was unstable when analyzed based on timing of death.Conclusions: Critically ill patients who have bacteremia typically receive long courses of antimicrobials. Most patient/pathogen characteristics are not associated with treatment duration; survivor bias precludes a valid assessment of the association between treatment duration and survival. A definitive randomized controlled trial is needed to compare shorter versus longer antimicrobial treatment in patients who have bacteremia. [ABSTRACT FROM AUTHOR]