Using survival analysis to predict septic shock onset in ICU patients.

PURPOSE: To determine the efficacy of survival analysis for predicting septic shock onset in ICU patients. MATERIALS AND METHODS: We performed a retrospective analysis on ICU cases from Mercy Hospital St. Louis from 2012 to 2016. As part of the procedure for inclusion in the Apache Outcomes database, each case is reviewed by critical care clinicians to identify septic shock patients as well as the time of septic shock onset. We used survival analysis to predict septic shock onset in these cases and employed lagging to compensate for uncertainties in septic shock onset time. RESULTS: Survival analysis was highly effective at predicting septic shock onset, producing AUC values of >0.87. The methodology was robust to lag times as well as the specific method of survival analysis used. CONCLUSIONS: This methodology has the potential to be implemented in the ICU for real time prediction and can be used as a building block to expand the approach to other hospital wards or care environments.

Virtual Care Supports the Common Good.

After completing my training program in critical care medicine more than 35 years ago, I joined Mercy Hospital St. Louis in Missouri, known then as St. John's Mercy Medical Center. Although I didn't fully appreciate it at the time, I was blessed to join a Catholic-based health care system built on the ideals of Catherine McAuley and the Sisters of Mercy.

Telemedicine intervention improves ICU outcomes.

Telemedicine for the intensive care unit (Tele-ICU) was founded as a means of delivering the clinical expertise of intensivists located remotely to hospitals with inadequate access to intensive care specialists. This was a retrospective pre- and postintervention study of adult patients admitted to a community hospital ICU. The patients in the preintervention period (n = 630) and during the Tele-ICU period (n = 2193) were controlled for baseline characteristics, acute physiologic scores (APS), and acute physiologic and health evaluation (APACHE IV) scores. Mean APS scores were 37.1 (SD, 22.8) and 37.7 (SD, 19.4) (P = 0.56), and mean APACHE IV scores were 49.7 (SD, 24.8) and 50.4 (SD, 21.0) (P = 0.53), respectively. ICU mortality was 7.9% during the preintervention period compared with 3.8% during the Tele-ICU period (odds ratio (OR) = 0.46, 95% confidence interval (CI), 0.32-0.66, P < 0.0001). ICU LOS in days was 2.7 (SD, 4.1) compared with 2.2 (SD, 3.4), respectively (hazard ratio (HR) = 1.16, 95% CI, 1.00-1.40, P = 0.01). Implementation of Tele-ICU intervention was associated with reduced ICU mortality and ICU LOS. This suggests that there are benefits of a closed Tele-ICU intervention beyond what is provided by daytime bedside physicians.

Therapeutic hypothermia for the management of intracranial hypertension in severe traumatic brain injury: a systematic review.

BACKGROUND: Traumatic brain injury (TBI) is a major source of death and severe disability worldwide. Raised Intracranial pressure (ICP) is an important predictor of mortality in patients with severe TBI and aggressive treatment of elevated ICP has been shown to reduce mortality and improve outcome. The acute post-injury period in TBI is characterized by several pathophysiologic processes that start in the minutes to hours following injury. All of these processes are temperature-dependent; they are all aggravated by fever and inhibited by hypothermia. METHODS: This study reviewed the current clinical evidence in support of the use of therapeutic hypothermia (TH) for the treatment of intracranial hypertension (ICH) in patients with severe TBI. RESULTS: This study identified a total of 18 studies involving hypothermia for control of ICP; 13 were randomized controlled trials (RCT) and five were observational studies. TH (32-34°C) was effective in controlling ICH in all studies. In the 13 RCT, ICP in the TH group was always significantly lower than ICP in the normothermia group. In the five observational studies, ICP during TH was always significantly lower than prior to inducing TH. CONCLUSIONS: Pending results from large multi-centre studies evaluating the effect of TH on ICH and outcome, TH should be included as a therapeutic option to control ICP in patients with severe TBI.

Silver-coated endotracheal tubes and incidence of ventilator-associated pneumonia: the NASCENT randomized trial.

CONTEXT: Ventilator-associated pneumonia (VAP) causes substantial morbidity. A silver-coated endotracheal tube has been designed to reduce VAP incidence by preventing bacterial colonization and biofilm formation. OBJECTIVE: To determine whether a silver-coated endotracheal tube would reduce the incidence of microbiologically confirmed VAP. DESIGN, SETTING, AND PARTICIPANTS: Prospective, randomized, single-blind, controlled study conducted in 54 centers in North America. A total of 9417 adult patients (> or = 18 years) were screened between 2002 and 2006. A total of 2003 patients expected to require mechanical ventilation for 24 hours or longer were randomized. INTERVENTION: Patients were assigned to undergo intubation with 1 of 2 high-volume, low-pressure endotracheal tubes, similar except for a silver coating on the experimental tube. MAIN OUTCOME MEASURES: Primary outcome was VAP incidence based on quantitative bronchoalveolar lavage fluid culture with 10(4) colony-forming units/mL or greater in patients intubated for 24 hours or longer. Other outcomes were VAP incidence in all intubated patients, time to VAP onset, length of intubation and duration of intensive care unit and hospital stay, mortality, and adverse events. RESULTS: Among patients intubated for 24 hours or longer, rates of microbiologically confirmed VAP were 4.8% (37/766 patients; 95% confidence interval [CI], 3.4%-6.6%) in the group receiving the silver-coated tube and 7.5% (56/743; 95% CI, 5.7%-9.7%) (P = .03) in the group receiving the uncoated tube (all intubated patients, 3.8% [37/968; 95% CI, 2.7%-5.2%] and 5.8% [56/964; 95% CI, 4.4%-7.5%] [P = .04]), with a relative risk reduction of 35.9% (95% CI, 3.6%-69.0%; all intubated patients, 34.2% [95% CI, 1.2%-67.9%]). The silver-coated endotracheal tube was associated with delayed occurrence of VAP (P = .005). No statistically significant between-group differences were observed in durations of intubation, intensive care unit stay, and hospital stay; mortality; and frequency and severity of adverse events. CONCLUSION: Patients receiving a silver-coated endotracheal tube had a statistically significant reduction in the incidence of VAP and delayed time to VAP occurrence compared with those receiving a similar, uncoated tube. TRIAL REGISTRATION: clinicaltrials.gov Identifier: NCT00148642.

Red blood cell transfusions and nosocomial infections in critically ill patients.

OBJECTIVE: A previous retrospective evaluation of Project Impact data demonstrated an association between red blood cell transfusions, nosocomial infections, and poorer outcomes in critically ill patients, independent of survival probability or patient age. The objective of this study was to determine whether transfused patients, independent of survival probability based on Mortality Prediction Model scores, have higher nosocomial infection rates, longer intensive care unit and hospital lengths of stay, and higher mortality rates than nontransfused patients. DESIGN: Prospective, observational, cohort study. SETTING: A single-center, mixed medical/surgical, closed intensive care unit. PATIENTS: : Adults admitted to St. John's Mercy Medical Center between August 2001 and June 2003 (n = 2,085) were enrolled using Project Impact software. Both nonoperative and postoperative populations were represented, and transfusion decisions were made independently of patient study inclusion. Patients whose nosocomial infection was diagnosed before transfusion were counted as nontransfused. INTERVENTIONS: : None. MEASUREMENTS AND MAIN RESULTS: Nosocomial infections, mortality rates, and intensive care unit and hospital length of stay were the main outcome measures. Of the 2,085 patients enrolled, 21.5% received red blood cell transfusions. The posttransfusion nosocomial infection rate was 14.3% in 428 evaluable patients, significantly higher than that observed in nontransfused patients (5.8%; p < .0001, chi-square). In a multivariate analysis controlling for patient age, maximum storage age of red blood cells, and number of red blood cell transfusions, only the number of transfusions was independently associated with nosocomial infection (odds ratio 1.097; 95% confidence interval 1.028-1.171; p = .005). When corrected for survival probability, the risk of nosocomial infection associated with red blood cell transfusions remained statistically significant (p < .0001). Leukoreduction tended to reduce the nosocomial infection rate but not significantly. Mortality and length of stay (intensive care unit and hospital) were significantly higher in transfused patients, even when corrected for illness severity. CONCLUSIONS: Red blood cell transfusions should be used sparingly, bearing in mind the potential risks of infection and poor outcomes in critically ill patients.

Ordering errors by first-year residents: evidence of learning from mistakes.

OBJECTIVE AND DESIGN: To study patterns of ordering errors, we reviewed charts of patients cared for by first-year residents at our community-based teaching hospital. RESULTS: In month one, 39 of 1248 orders (3.1%) were rated as errors. In contrast, 9 of 1072 orders (0.84%) in month six were rated as errors (p < 0.001). Of the 48 errors identified, 36 (75%) related to medications (choice, dose, route, frequency, or duration). Over half (53%) of these errors involved antimicrobials, pulmonary medications, or diabetes-related drugs. For all errors, half (24/48) were identified by attending physicians, with hospital support staff noting most of the remaining errors (43.8%). Errors were ultimately corrected by the first-year resident in over half of the cases (52%); attending physicians corrected 16 of 39 errors (41%) in month one but only one of nine errors (11%) in month six. Errors by first-year residents were primarily attributed to inattention (45.8%) or deficits in clinical knowledge (43.8%). No adverse effects resulted from any of the ordering errors identified. Primary consequences of errors included inconvenience to staff (50%), delay in treatment or diagnosis (31.3%), or receiving unneeded medication (18.8%). CONCLUSIONS: The number of medical errors decreases with experience. Our results highlight the importance of clinical supervision during the initial months of training.

Impact of allogenic packed red blood cell transfusion on nosocomial infection rates in the critically ill patient.

OBJECTIVE: To determine whether critically ill patients who receive allogenic packed red blood cell transfusions are at increased risk of developing nosocomial infections during hospitalization. DESIGN: Retrospective database study utilizing Project IMPACT. SETTING: A 40-bed medical-surgical-trauma intensive care unit in an 825-bed tertiary referral teaching hospital. PATIENTS: One thousand seven hundred and seventeen patients admitted to the medical-surgical-trauma intensive care unit. MEASUREMENTS AND MAIN RESULTS: Data were collected by using the Project IMPACT database. Nosocomial infection rates were compared among three groups: the entire cohort, the transfusion group, and the nontransfusion group. We determined the nosocomial infection rates in these groups while adjusting for probability of survival by using Mortality Prediction Model (MPM-0) scores, age, gender, and number of units of packed red blood cells transfused. The average number of units transfused per patient was 4.0. The nosocomial infection rate for the entire cohort was 5.94%. The nosocomial infection rates for the transfusion group (n = 416) and the nontransfusion group (n = 1301) were 15.38% and 2.92%, respectively (p <.005 chi-square). Transfusion of packed red blood cells was related to the occurrence of nosocomial infection, and there was a dose-response pattern (the more units of packed red blood cells transfused, the greater the chance of nosocomial infection; p< 0.0001 chi-square). The transfusion group was six times more likely to develop nosocomial infection compared with the nontransfusion group. In addition, for each unit of packed red blood cells transfused, the odds of developing nosocomial infection were increased by a factor of 1.5. A subgroup analysis of nosocomial infection rates adjusted for probability of survival by using MPM-0 scores showed nosocomial infection to occur at consistently higher rates in transfused patients vs. nontransfused patients. A second subgroup analysis adjusted for patient age showed a statistically significant increase in rates of nosocomial infection for transfused patients regardless of age. CONCLUSIONS: Transfusion of packed red blood cells is associated with nosocomial infection. This association continues to exist when adjusted for probability of survival and age. In addition, mortality rates and length of intensive care unit and hospital stay are significantly increased in transfused patients.