Development and internal validation of the multivariable CIPHER (Collaborative Integrated Pregnancy High-dependency Estimate of Risk) clinical risk prediction model.

BACKGROUND: Intensive care unit (ICU) outcome prediction models, such as Acute Physiology And Chronic Health Evaluation (APACHE), were designed in general critical care populations and their use in obstetric populations is contentious. The aim of the CIPHER (Collaborative Integrated Pregnancy High-dependency Estimate of Risk) study was to develop and internally validate a multivariable prognostic model calibrated specifically for pregnant or recently delivered women admitted for critical care. METHODS: A retrospective observational cohort was created for this study from 13 tertiary facilities across five high-income and six low- or middle-income countries. Women admitted to an ICU for more than 24 h during pregnancy or less than 6 weeks post-partum from 2000 to 2012 were included in the cohort. A composite primary outcome was defined as maternal death or need for organ support for more than 7 days or acute life-saving intervention. Model development involved selection of candidate predictor variables based on prior evidence of effect, availability across study sites, and use of LASSO (Least Absolute Shrinkage and Selection Operator) model building after multiple imputation using chained equations to address missing data for variable selection. The final model was estimated using multivariable logistic regression. Internal validation was completed using bootstrapping to correct for optimism in model performance measures of discrimination and calibration. RESULTS: Overall, 127 out of 769 (16.5%) women experienced an adverse outcome. Predictors included in the final CIPHER model were maternal age, surgery in the preceding 24 h, systolic blood pressure, Glasgow Coma Scale score, serum sodium, serum potassium, activated partial thromboplastin time, arterial blood gas (ABG) pH, serum creatinine, and serum bilirubin. After internal validation, the model maintained excellent discrimination (area under the curve of the receiver operating characteristic (AUROC) 0.82, 95% confidence interval (CI) 0.81 to 0.84) and good calibration (slope of 0.92, 95% CI 0.91 to 0.92 and intercept of -0.11, 95% CI -0.13 to -0.08). CONCLUSIONS: The CIPHER model has the potential to be a pragmatic risk prediction tool. CIPHER can identify critically ill pregnant women at highest risk for adverse outcomes, inform counseling of patients about risk, and facilitate bench-marking of outcomes between centers by adjusting for baseline risk.

The influence of rewarming after therapeutic hypothermia on outcome after cardiac arrest.

INTRODUCTION: Treatment with hypothermia has been shown to improve outcome after cardiac arrest (CA). Current consensus is to rewarm at 0.25-0.5 °C/h and avoid fever. The aim of this study was to investigate whether active rewarming, the rate of rewarming or development of fever after treatment with hypothermia after CA was correlated with poor outcome. METHODS: This retrospective cohort study included adult patients treated with hypothermia after CA and admitted to the intensive care unit between January 2006 and January 2009. The average rewarming rate from end of hypothermia treatment (passive rewarming) or start active rewarming until 36 °C was dichotomized in a high (≥ 0.5 °C/h) or normal rate (<0.5 °C/h). Fever was defined as >38 °C within 72 h after admission. Poor outcome was defined as death, vegetative state, or severe disability after 6 months. RESULTS: From 128 included patients, 56% had a poor outcome. Actively rewarmed patients (38%) had a higher risk for poor outcome, OR 2.14 (1.01-4.57), p<0.05. However, this effect disappeared after adjustment for the confounders age and initial rhythm, OR 1.51 (0.64-3.58). A poor outcome was found in 15/21 patients (71%) with a high rewarming rate, compared to 54/103 patients (52%) with a normal rewarming rate, OR 2.61 (0.88-7.73), p = 0.08. Fever was not associated with outcome, OR 0.64 (0.31-1.30), p = 0.22. CONCLUSIONS: This study showed that patients who needed active rewarming after therapeutic hypothermia after CA did not have a higher risk for a poor outcome. In addition, neither speed of rewarming, nor development of fever had an effect on outcome.

On-ward participation of a hospital pharmacist in a Dutch intensive care unit reduces prescribing errors and related patient harm: an intervention study.

INTRODUCTION: Patients admitted to an intensive care unit (ICU) are at high risk for prescribing errors and related adverse drug events (ADEs). An effective intervention to decrease this risk, based on studies conducted mainly in North America, is on-ward participation of a clinical pharmacist in an ICU team. As the Dutch Healthcare System is organized differently and the on-ward role of hospital pharmacists in Dutch ICU teams is not well established, we conducted an intervention study to investigate whether participation of a hospital pharmacist can also be an effective approach in reducing prescribing errors and related patient harm (preventable ADEs) in this specific setting. METHODS: A prospective study compared a baseline period with an intervention period. During the intervention period, an ICU hospital pharmacist reviewed medication orders for patients admitted to the ICU, noted issues related to prescribing, formulated recommendations and discussed those during patient review meetings with the attending ICU physicians. Prescribing issues were scored as prescribing errors when consensus was reached between the ICU hospital pharmacist and ICU physicians. RESULTS: During the 8.5-month study period, medication orders for 1,173 patients were reviewed. The ICU hospital pharmacist made a total of 659 recommendations. During the intervention period, the rate of consensus between the ICU hospital pharmacist and ICU physicians was 74%. The incidence of prescribing errors during the intervention period was significantly lower than during the baseline period: 62.5 per 1,000 monitored patient-days versus 190.5 per 1,000 monitored patient-days, respectively (P < 0.001). Preventable ADEs (patient harm, National Coordinating Council for Medication Error Reporting and Prevention severity categories E and F) were reduced from 4.0 per 1,000 monitored patient-days during the baseline period to 1.0 per 1,000 monitored patient-days during the intervention period (P = 0.25). Per monitored patient-day, the intervention itself cost €3, but might have saved €26 to €40 by preventing ADEs. CONCLUSIONS: On-ward participation of a hospital pharmacist in a Dutch ICU was associated with significant reductions in prescribing errors and related patient harm (preventable ADEs) at acceptable costs per monitored patient-day. TRIAL REGISTRATION NUMBER: ISRCTN92487665.

Spontaneous hypothermia on intensive care unit admission is a predictor of unfavorable neurological outcome in patients after resuscitation: an observational cohort study.

INTRODUCTION: A large number of patients resuscitated for primary cardiac arrest arrive in the intensive care unit (ICU) with a body temperature < 35.0 degrees C. The aim of this observational cohort study was to determine the association between ICU admission temperature and neurological outcome in this patient group. METHODS: Demographics and parameters influencing neurological outcome were retrieved from the charts of all patients resuscitated for primary cardiac arrest and treated with induced mild hypothermia in our ICU from January 2006 until January 2008. Patients were divided into two groups according to their body temperature on ICU admission: a hypothermia group (< 35.0 degrees C) and a non-hypothermia group (>or=35.0 degrees C). Neurological outcome after six months was assessed by means of the Glasgow Outcome Score (GOS), with GOS 1 to 3 defined as unfavorable and GOS 4 to 5 as favorable. A logistic regression model was used to analyze the influence of the different parameters on neurological outcome. RESULTS: The data of 105 consecutive patients resuscitated for primary cardiac arrest and treated with induced mild hypothermia were analyzed. Median ICU admission temperature was 35.1 degrees C (interquartile range (IQR) 34.3 to 35.7). After six months, 61% of the patients had an unfavorable outcome (59% died and 2% were severely disabled), whereas 39% had a favorable outcome (moderate disability or good recovery). Among patients with spontaneous hypothermia on ICU admission, the percentage with unfavorable outcome was higher (69% versus 50%, P = 0.05). Logistic regression showed that age, acute physiology and chronic health evaluation (APACHE) II and sequential organ failure assessment (SOFA) scores and spontaneous hypothermia on ICU admission all had an increased odds ratio (OR) for an unfavorable outcome after six months. Spontaneous hypothermia had the strongest association with unfavorable outcome (OR 2.6, 95% CI (confidence interval) 1.1 to 5.9), which became even stronger after adjustment for age, presenting heart rhythm, APACHE II and SOFA scores (OR 3.8, CI 1.3 to 11.0). CONCLUSIONS: In this observational cohort study, spontaneous hypothermia on ICU admission was the strongest predictor of an unfavorable neurological outcome in patients resuscitated for primary cardiac arrest.

A positive fluid balance is associated with a worse outcome in patients with acute renal failure.

INTRODUCTION: Despite significant improvements in intensive care medicine, the prognosis of acute renal failure (ARF) remains poor, with mortality ranging from 40% to 65%. The aim of the present observational study was to analyze the influence of patient characteristics and fluid balance on the outcome of ARF in intensive care unit (ICU) patients. METHODS: The data were extracted from the Sepsis Occurrence in Acutely Ill Patients (SOAP) study, a multicenter observational cohort study to which 198 ICUs from 24 European countries contributed. All adult patients admitted to a participating ICU between 1 and 15 May 2002, except those admitted for uncomplicated postoperative surveillance, were eligible for the study. For the purposes of this substudy, patients were divided into two groups according to whether they had ARF. The groups were compared with respect to patient characteristics, fluid balance, and outcome. RESULTS: Of the 3,147 patients included in the SOAP study, 1,120 (36%) had ARF at some point during their ICU stay. Sixty-day mortality rates were 36% in patients with ARF and 16% in patients without ARF (P < 0.01). Oliguric patients and patients treated with renal replacement therapy (RRT) had higher 60-day mortality rates than patients without oliguria or the need for RRT (41% versus 33% and 52% versus 32%, respectively; P < 0.01). Independent risk factors for 60-day mortality in the patients with ARF were age, Simplified Acute Physiology Score II (SAPS II), heart failure, liver cirrhosis, medical admission, mean fluid balance, and need for mechanical ventilation. Among patients treated with RRT, length of stay and mortality were lower when RRT was started early in the course of the ICU stay. CONCLUSION: In this large European multicenter study, a positive fluid balance was an important factor associated with increased 60-day mortality. Outcome among patients treated with RRT was better when RRT was started early in the course of the ICU stay.

Individual and joint expert judgments as reference standards in artifact detection.

OBJECTIVE: To investigate the agreement among clinical experts in their judgments of monitoring data with respect to artifacts, and to examine the effect of reference standards that consist of individual and joint expert judgments on the performance of artifact filters. DESIGN: Individual judgments of four physicians, a majority vote judgment, and a consensus judgment were obtained for 30 time series of three monitoring variables: mean arterial blood pressure (ABPm), central venous pressure (CVP), and heart rate (HR). The individual and joint judgments were used to tune three existing automated filtering methods and to evaluate the performance of the resulting filters. MEASUREMENTS: The interrater agreement was calculated in terms of positive specific agreement (PSA). The performance of the artifact filters was quantified in terms of sensitivity and positive predictive value (PPV). RESULTS: PSA values between 0.33 and 0.85 were observed among clinical experts in their selection of artifacts, with relatively high values for CVP data. Artifact filters developed using judgments of individual experts were found to moderately generalize to new time series and other experts; sensitivity values ranged from 0.40 to 0.60 for ABPm and HR filters (PPV: 0.57-0.84), and from 0.63 to 0.80 for CVP filters (PPV: 0.71-0.86). A higher performance value for the filters was found for the three variable types when joint judgments were used for tuning the filtering methods. CONCLUSION: Given the disagreement among experts in their individual judgment of monitoring data with respect to artifacts, the use of joint reference standards obtained from multiple experts is recommended for development of automatic artifact filters.