Showing papers in "Intensive Care Medicine in 2001"

Intensive Care Delirium Screening Checklist: evaluation of a new screening tool

Abstract: Objective: Delirium in the intensive care unit is poorly defined. Clinical evaluation is difficult in the setting of unstable, often intubated patients. A screening tool may improve the detection of delirium. Method: We created a screening checklist of eight items based on DSM criteria and features of delirium: altered level of consciousness, inattention, disorientation, hallucination or delusion, psychomotor agitation or retardation, inappropriate mood or speech, sleep/wake cycle disturbance, and symptom fluctuation. During 3 months, all patients admitted to a busy medical/surgical intensive care unit were evaluated, and the scale score was compared to a psychiatric evaluation. Results: In 93 patients studied, 15 developed delirium. Fourteen (93%) of them had a score of 4 points or more. This score was also present in 15 (19%) of patients without delirium, 14 of whom had a known psychiatric illness, dementia, a structural neurological abnormality or encephalopathy. A ROC analysis was used to determine the sensitivity and specificity of the screening tool. The area under the ROC curve is 0.9017. Predicted sensitivity is 99% and specificity is 64%. Conclusion: This study suggests that the Intensive Care Delirium Screening Checklist can easily be applied by a clinician or a nurse in a busy critical care setting to screen all patients even when communication is compromised. The tool can be utilized quickly and helps to identify delirious patients. Earlier diagnosis may lead to earlier intervention and better patient care.

The impact of delirium in the intensive care unit on hospital length of stay

Abstract: Study objective: To determine the relationship between delirium in the intensive care unit (ICU) and outcomes including length of stay in the hospital. Design: A prospective cohort study. Setting: The adult medical ICU of a tertiary care, university-based medical center. Participants: The study population consisted of 48 patients admitted to the ICU, 24 of whom received mechanical ventilation. Measurements: All patients were evaluated for the development and persistence of delirium on a daily basis by a geriatric or psychiatric specialist with expertise in delirium assessment using the Diagnostic Statistical Manual IV (DSM-IV) criteria of the American Psychiatric Association, the reference standard for delirium ratings. Primary outcomes measured were length of stay in the ICU and hospital. Results: The mean onset of delirium was 2.6 days (S.D.±1.7), and the mean duration was 3.4±1.9 days. Of the 48 patients, 39 (81.3%) developed delirium, and of these 29 (60.4%) developed the complication while still in the ICU. The duration of delirium was associated with length of stay in the ICU (r=0.65, P=0.0001) and in the hospital (r=0.68, P<0.0001). Using multivariate analysis, delirium was the strongest predictor of length of stay in the hospital (P=0.006) even after adjusting for severity of illness, age, gender, race, and days of benzodiazepine and narcotic drug administration. Conclusions: In this patient cohort, the majority of patients developed delirium in the ICU, and delirium was the strongest independent determinant of length of stay in the hospital. Further study and monitoring of delirium in the ICU and the risk factors for its development are warranted.

Predictors of failure of noninvasive positive pressure ventilation in patients with acute hypoxemic respiratory failure: a multi-center study.

Abstract: Context: In patients with hypoxemic acute respiratory failure (ARF), randomized studies have shown noninvasive positive pressure ventilation (NPPV) to be associated with lower rates of endotracheal intubation. In these patients, predictors of NPPV failure are not well characterized. Objective: To investigate variables predictive of NPPV failure in patients with hypoxemic ARF. Design: Prospective, multicenter cohort study. Setting: Eight Intensive Care Units (ICU) in Europe and USA. Patients: Of 5,847 patients admitted between October 1996 and December 1998, 2,770 met criteria for hypoxemic ARF. Of these, 2,416 were already intubated and 354 were eligible for the study. Results: NPPV failed in 30% (108/354) of patients. The highest intubation rate was observed in patients with ARDS (51%) or community-acquired pneumonia (50%). The lowest intubation rate was observed in patients with cardiogenic pulmonary edema (10%) and pulmonary contusion (18%). Multivariate analysis identified age >40 years (OR 1.72, 95% CI 0.92–3.23), a simplified acute physiologic score (SAPS II) ≥35 (OR 1.81, 95% CI 1.07–3.06), the presence of ARDS or community-acquired pneumonia (OR 3.75, 95% CI 2.25–6.24), and a PaO2:FiO2 ≤146 after 1 h of NPPV (OR 2.51, 95% CI 1.45–4.35) as factors independently associated with failure of NPPV. Patients requiring intubation had a longer duration of ICU stay (P<0.001), higher rates of ventilator-associated pneumonia and septic complications (P<0.001), and a higher ICU mortality (P<0.001). Conclusions: In hypoxemic ARF, NPPV can be successful in selected populations. When patients have a higher severity score, an older age, ARDS or pneumonia, or fail to improve after 1 h of treatment, the risk of failure is higher.

Delirium in an intensive care unit: a study of risk factors

Abstract: Objectives: (1) To establish risk factors for the development of delirium in an intensive care unit (ICU) and (2) to determine the effect of delirium on morbidity, mortality and length of stay. Design: Prospective study. Setting: Sixteen-bed medical/surgical ICU in a university hospital. Patients: Two hundred and sixteen consecutive patients admitted to the ICU for more than 24 h during 5 months were included in the study. Interventions: Medical history, selected laboratory values, drugs received and factors that may influence patient psychological and emotional well-being were noted. All patients were screened with a delirium scale. A psychiatrist confirmed the diagnosis of delirium. Major complications such as self-extubation and removal of catheters, as well as mortality and length of stay were recorded. Results: Forty patients (19%) developed delirium; of these, one-third were not agitated. In the multivariate analysis hypertension, smoking history, abnormal bilirubin level, epidural use and morphine were statistically significantly associated with delirium. Traditional factors associated with the development of delirium on general ward patients were not significant in our study. Morbidity (self-extubation and removal of catheters), but not mortality, was clearly increased. Conclusion: Predictive risk factors for the development of delirium in studies outside the ICU may not be applicable to critically ill patients. Delirium is associated with increased morbidity. Awareness of patients at risk may lead to better recognition and earlier intervention.

Critical illness polyneuropathy: risk factors and clinical consequences. A cohort study in septic patients.

Abstract: Objective: To determine risk factors and clinical consequences of critical illness polyneuropathy (CIP) evaluated by the impact on duration of mechanical ventilation, length of stay and mortality. Design: Inception cohort study. Setting: Intensive care unit of a tertiary hospital. Patients: Septic patients with multiple organ dysfunction syndrome requiring mechanical ventilation and without previous history of polyneuropathy. Interventions: Patients underwent two scheduled electrophysiologic studies (EPS): on the 10th and 21st days after the onset of mechanical ventilation. Results: Eighty-two patients were enrolled, although nine of them were not analyzed. Forty-six of the 73 patients presented CIP on the first EPS and 4 other subjects were diagnosed with CIP on the second evaluation. The APACHE II scores of patients with and without CIP were similar on admission and on the day of the first EPS. However, days of mechanical ventilation [32.3 (21.1) versus 18.5 (5.8); p=0.002], length of ICU and hospital stay in patients discharged alive from the ICU as well as in-hospital mortality were greater in patients with CIP (42/50, 84% versus 13/23, 56.5%; p=0.01). After multivariate analysis, independent risk factors were hyperosmolality [odds ratio (OR) 4.8; 95% confidence intervals (95% CI) 1.05–24.38; p=0.046], parenteral nutrition (OR 5.11; 95% CI 1.14–22.88; p=0.02), use of neuromuscular blocking agents (OR 16.32; 95% CI 1.34–199; p=0.0008) and neurologic failure (GCS below 10) (OR 24.02; 95% CI 3.68–156.7; p<0.001), while patients with renal replacement therapy had a lower risk for CIP development (OR 0.02; 95% CI 0.05–0.15; p<0.001). By multivariate analysis, CIP (OR 7.11; 95% CI 1.54–32.75; p<0.007), age over 60 years (OR 9.07; 95% CI 2.02–40.68; p<0.002) and the worst renal SOFA (OR 2.18; 95% CI 1.27–3.74; p<0.002) were independent predictors of in-hospital mortality. Conclusions: CIP is associated with increased duration of mechanical ventilation and in-hospital mortality. Hyperosmolality, parenteral nutrition, non-depolarizing neuromuscular blockers and neurologic failure can favor CIP development.

Elevation of troponin I in sepsis and septic shock.

Abstract: Objective: To detect myocardial damage in severe systemic inflammation by cTnI measurements in patients without acute coronary syndromes. Design: Prospective case control study. Setting: Tertiary referral center. Participants: Twenty patients with sepsis, septic shock, and systemic inflammatory response syndrome (SIRS) were examined and compared to controls without coronary artery disease or myocarditis. Measurements and results: cTnI levels were assessed in patients with SIRS, sepsis, and septic shock. Eight patients (two female/six male) suffered from septic shock, nine (three female/six male) from sepsis without shock, and three (three male) from SIRS. Seventeen patients (85%) showed elevated cTnI (median 0.57 µg/l; 0.17–15.4), whereas no patient in the control group showed elevated cTnI (P<0.0001). Six patients (30%), – three with septic shock and three with sepsis – died during hospitalization, five of them with elevated cTnI. Four out of five autopsies showed normal coronary arteries. Coronary angiography, autopsy, and stress echocardiography ruled out significant coronary artery disease in ten cTnI-positive patients (59%). In 41% of cTnI-positive patients, Streptococcus pneumoniae could be cultured, whereas no cTnI-negative or control patient showed signs of infection due to S. pneumoniae. Conclusion: Cardiac troponin I was elevated in 85% of patients with sepsis, septic shock or SIRS in our study. A high percentage showed infection caused by S. pneumoniae. In what way microorganisms cause cTnI elevations is not yet understood.

Base excess and lactate as prognostic indicators for patients admitted to intensive care

Abstract: Objective: To examine whether values of arterial base excess or lactate taken on admission to a general intensive care unit indicate prognosis, and whether this can be used as a screening tool for future intensive care admissions. Design: Observational study. Setting: University teaching hospital general adult intensive care unit. Patients: 148 consecutive patients admitted to the intensive care unit. Interventions: Arterial blood samples were obtained on admission to the intensive care unit and 24 h following admission. Measurements and results: Arterial base excess and lactate concentrations were measured from the blood samples. Both base excess and arterial lactate samples on admission have good prognostic abilities (area under the curve on receiver operator characteristic analysis of 0.73, 0.78, respectively). The value of base excess on admission with the best predictive ability was a base excess more negative than –4 mmol/l, and the corresponding value for lactate was greater than 1.5 mmol/l. The combination of these two markers on admission to the intensive care unit led to a sensitivity of 80.3 % and a specificity of 58.7 % for mortality. The achievement of this combination was associated with an increased mortality (50.6 % vs. 15 %, p < 0.0001), older age (70 vs. 61.5 years, p < 0.05), a greater requirement for inotropic support (30.9 % vs. 4.5 %, p < 0.0001) and higher organ failure scores both on admission and for the subsequent 24 h. Conclusions: Both base excess and lactate, or the combination of the two, can be used to predict outcome in patients admitted to the intensive care unit. These variables could be utilized to identify patients who have a high risk for mortality and thus who should be admitted to the intensive care unit.

The effects of vasopressin on hemodynamics and renal function in severe septic shock: a case series

Abstract: Objective: To review all cases of septic shock treated with vasopressin to determine the effects on hemodynamic and renal function and to document any adverse effects. Setting: A 14-bed mixed medical-surgical ICU of St. Paul's Hospital, a 450-bed tertiary referral hospital affiliated with the University of British Columbia. Patients: All ICU patients who received vasopressin for treatment of severe septic shock between August 5, 1997, and March 21, 1999. Results: We identified 50 patients: age 60 (±14); APACHE II score 27 (±7). Baseline data (T0) was compared to data at T4, T24 and T48 (4, 24 and 48 h) on infusion. Mean arterial pressure (MAP) increased by 18% from T0 to T4 and remained stable at T24 (p=0.006) and T48 (p=0.008). Systolic pulmonary artery pressure (PAP) was unchanged at 45±13 mmHg. Mean cardiac index (CI) decreased by 11% at T4 (p=0.03). Urine output increased 79% at T4 (p=0.005) and further increases were not significant at T24 and T48. Mean pressor dosage decreased by 33% at T4 (p=0.001), by 53% at T24 (p=0.002) and by 48% at T48 (p=0.01). Hospital mortality was 85%. There were six cardiac arrests; all but one occurred at a vasopressin dose of 0.05 U/min or more. Conclusions: In this group of patients with severe septic shock, vasopressin infusion increased MAP and urine output and decreased catecholamine requirements. Doses higher than 0.04 U/min were not associated with increased effectiveness and may have been associated with higher adverse effects.

High-volume haemofiltration in human septic shock.

Abstract: Objective: To evaluate whether high volume haemofiltration improves haemodynamics and affects serum cytokine and complement concentrations in human septic shock. Design and setting: Randomized cross-over clinical trial in a tertiary intensive care unit. Patients: Eleven patients with septic shock and multi-organ failure. Interventions: Patients were assigned to either 8 h of high-volume haemofiltration (HVHF; 6 l/h) or 8 h of standard continuous veno-venous haemofiltration (CVVH; 1 l/h) in random order. Measurements and main results: We measured changes in haemodynamic variables, dose of norepinephrine required to maintain a mean arterial pressure greater than 70 mmHg and plasma concentrations of complement anaphylatoxins and several cytokines. An 8-h period of HVHF was associated with a greater reduction in norepinephrine requirements than a similar period of CVVH (median reduction: 10.5 vs. 1.0 µg/min; p=0.01; median percentage reduction: 68 vs. 7%; p=0.02). Both therapies were associated with a temporary reduction (p<0.01) in the plasma concentration of C3a, C5a, and interleukin 10 within 2 h of initiation. HVHF was associated with a greater reduction in the area under the curve for C3a and C5a (p<0.01). The concentration of the measured soluble mediators in the ultrafiltrate was negligible. Conclusions: HVHF decreases vasopressor requirements in human septic shock and affects anaphylatoxin levels differently than standard CVVH

Plasma glutamine depletion and patient outcome in acute ICU admissions

Abstract: Objective: To evaluate whether low plasma glutamine (PG) is related to severity of illness, and actual and predicted hospital mortality. Design: Prospective cohort study. Setting: 18-bed closed format general intensive care unit (ICU) of a teaching hospital. Patients: Cohort of 80 seriously ill patients non-electively admitted to the ICU. Interventions: Blood sampling for the determination of PG at ICU admission. Measurements and results: Severity of illness and predicted mortality were calculated using the locally validated APACHE II, SAPS II, and MPM II 0 and 24 systems. Illness scores, and actual and predicted hospital mortality were compared between patients with total PG<0.420 mmol/l ("low PG") and patients with PG≥0.420 mmol/l. Mean total PG was 0.523 mmol/l, range 0.220–1.780 mmol/l. Low PG (n=25) was associated with higher age (P=0.03), shock as primary diagnosis, and higher actual hospital mortality (60% vs 29%, P=0.01). Normal to high PG was associated with high plasma creatine phosphokinase (P=0.007) There was a non-significant trend towards higher severity of illness scores and predicted mortality rates in the low PG group. The presence of low PG significantly improved mortality prediction when added as a factor to the APACHE II predicted mortality rate (P=0.02). Conclusions: Low PG at acute ICU admission is related to higher age, shock as primary diagnosis, and higher hospital mortality. Low PG represents a risk of poor outcome, not fully reflected in the presently used mortality prediction systems.

Impact of appropriateness of initial antibiotic therapy on the outcome of ventilator-associated pneumonia.

Abstract: Objectives: To evaluate the impact of appropriate initial antibiotic therapy (AB) on the outcome of ventilator-associated pneumonia (VAP). Design: Retrospective study (1992–97). Patients and methods: Episodes of VAP diagnosed on both clinical and microbiological criteria after ≥48 h of mechanical ventilation (MV). Initial AB was considered appropriate when all significant organisms were susceptible to at least one of the antibiotics started after distal bronchial sampling. Antibiotic treatment was modified within 48 h when susceptibility testing was available. Outcome was recorded at the ICU and hospital discharge. Results: One hundred and eleven patients were included (SAPS II=48±18, age=62±14 years, mean duration of MV before VAP=12±9 days). Initial AB was appropriate in 55 patients (49.5%). No difference between appropriate initial AB and inappropriate initial AB was found concerning severity indices at the time of VAP diagnosis. ICU length of stay was shorter with appropriate initial AB than with inappropriate initial AB for survivors (12±11 days vs 20±24 days, P=0.01). Crude hospital mortality tended to be lower with appropriate initial AB than with inappropriate initial AB (47.3% vs 60.7%, odds ratio=1.72, 95% CI=0.81–3.7). Relative crude mortality reduction with appropriate initial AB was 22%, 95% CI=–10% to 45%. Conclusion: Inappropriate initial AB of VAP during the first 48 h increased ICU length of stay after VAP diagnosis and tended to increase crude hospital mortality despite equal severity of illness at the time of VAP diagnosis, when compared to appropriate initial AB in a population of 111 ICU patients.

International Consensus Conferences in Intensive Care Medicine: non-invasive positive pressure ventilation in acute respiratory failure. Organised jointly by the American Thoracic Society, the European Respiratory Society, the European Society of Intensive Care Medicine, and the Société de Réanimation de Langue Française, and approved by the ATS Board of Directors, December 2000.

Abstract: Noninvasive positive pressure ventilation (NPPV) was applied first to patients with chronic pulmonary disease but is now being used to support those with acute respiratory failure (ARF) An International Consensus Conference in Intensive Care Medicine considering the role of NPPV in ARF was held in Paris, France, from April 13‐14, 2000; sponsored by the Critical Care Assembly of the American Thoracic Society (ATS), the European Respiratory Society (ERS), the European Society of Intensive Care Medicine (ESICM), and the Societe de Reanimation de Langue Fran aise (SRLF) The methods of the Consensus were established by the National Institutes of Health (1) and adapted subsequently for use in critical care medicine (2) Briefly, the process comprised four phases First, five key questions were formulated by the scientific advisors designed to address issues integral to the evaluation of noninvasive ventilatory support in its current and future roles Second, a comprehensive literature search was performed and key articles precirculated to a jury of 10 clinician scientists who were not experts in the field under discussion Third, authorities in NPPV selected by the Organizing Committee and scientific advisors delivered focused presentations during a two-day symposium attended by the jury and about 150 delegates Each presentation was followed by debate and discussion Finally, the jury summarized the available evidence in response to the questions over the two days immediately after the conference For the purposes of this report, NPPV was defined as any form of ventilatory support applied without the use of an endotracheal tube, and was considered to include continuous c positive airway pressure (CPAP), with or without inspiratory pressure support; volume- and pressure-cycled systems, proportional assist ventilation (PAV), and adjuncts such as the use of helium‐oxygen (heliox) gas mixtures The term acute respiratory failure (ARF) was considered to include patients with acute lung injury (ALI) and acute respiratory distress syndrome (ARDS), those with acute exacerbations of obstructive airflow limitation (ie, asthma and COPD); acutely decompensated patients with the obesity hypoventilation syndrome (OHS) and cardiogenic pulmonary edema (CPE); patients developing ARF in the perioperative period; and those with either difficulty weaning from invasive mechanical ventilatory support, or in whom endotracheal intubation (ETI) was considered inappropriate The information presented to the jury was designed to address the following five questions

Accuracy of pulse oximetry in the intensive care unit.

Abstract: Objective: Pulse oximetry (SpO2) is a standard monitoring device in intensive care units (ICUs), currently used to guide therapeutic interventions. Few studies have evaluated the accuracy of SpO2 in critically ill patients. Our objective was to compare pulse oximetry with arterial oxygen saturation (SaO2) in such patients, and to examine the effect of several factors on this relationship. Design: Observational prospective study. Setting: A 26-bed medical ICU in a university hospital. Patients: One hundred two consecutive patients admitted to the ICU in whom one or serial arterial blood gas analyses (ABGs) were performed and a reliable pulse oximeter signal was present. Interventions: For each ABG, we collected SaO2, SpO2, the type of pulse oximeter, the mode of ventilation and requirement for vasoactive drugs. Measurements and results: Three hundred twenty-three data points were collected. The mean difference between SpO2 and SaO2 was –0.02% and standard deviation of the differences was 2.1%. From one sample to another, the fluctuations in SpO2 to arterial saturation difference indicated that SaO2 could not be reliably predicted from SpO2 after a single ABG. Subgroup analysis showed that the accuracy of SpO2 appeared to be influenced by the type of oximeter, the presence of hypoxemia and the requirement for vasoactive drugs. Finally, high SpO2 thresholds were necessary to detect significant hypoxemia with good sensitivity. Conclusion: Large SpO2 to SaO2 differences may occur in critically ill patients with poor reproducibility of SpO2. A SpO2 above 94% appears necessary to ensure a SaO2 of 90%.

Patterns of corticosteroid-binding globulin and the free cortisol index during septic shock and multitrauma

Abstract: Objective: To study the time course of corticosteroid binding-globulin (CBG) level and the free cortisol index (FCI) in comparison with total cortisol and ACTH concentrations during acute and prolonged critical illness. Design: Prospective observational clinical study. Setting: Twenty-bed medical/surgical intensive care unit. Patients and participants: Thirty patients with septic shock, eight patients with multitrauma, and forty healthy control subjects. Measurements and results: During 14 days or until discharge/death, we serially measured serum concentrations of CBG, cortisol, TNF-α, IL-6, plasma ACTH immunoreactivity, and the FCI (=cortisol/CBG × 100). We also recorded haemodynamic parameters, APACHE II, ISS, SOFA scores, shock duration, inotrope use, and ICU mortality. In both groups we found markedly decreased CBG levels in the early phase (septic shock: 17.5±5.9, and trauma: 16.1±2.3 mg/l) in comparison with controls (37.3±5.3 mg/l). The FCI was high in this early phase (septic shock: 7.2±2.7; trauma: 6.5±1.3; controls: 1.25±0.76). During follow-up, CBG levels significantly increased, reaching normal levels from day 7 on. The FCI showed an opposite biphasic pattern, with near-normalising FCI values during the second phase. Regression analysis showed a negative correlation between CBG and IL-6 levels (rs=–0.63; P<0.05), but no relation between CBG concentrations and disease severity, shock duration or death was found. Conclusions: We found extremely low CBG levels in early stage septic shock and multitrauma. These dramatic changes are reflected in a concomitant higher FCI, indicating a higher free cortisol level. A second phase displays increasing and normalising CBG levels, independent from clinical parameters. We believe that CBG plays an active role in the glucocorticoid response to severe stress and in the regulation of cortisol availability to target tissues.

Pathogenesis and treatment of renal dysfunction in rhabdomyolysis

Abstract: Rhabdomyolysis is a major cause of acute renal failure, and recent experimental data have provided a better understanding of the pathophysiology of the renal dysfunction Renal failure is due to renal vasoconstriction, tubular damage caused by oxidant injury, and possibly tubular obstruction Recent studies have provided greater insight into the rationale behind current therapy and potential treatment strategies This review thus aims to summarise current understanding of the causes, pathogenesis and treatment of renal failure caused by rhabdomyolysis

Withdrawing and withholding life support in the intensive care unit: a Spanish prospective multi-centre observational study

Abstract: Objective: To determine how frequently life support is withheld or withdrawn from adult critically ill patients, and how physicians and patients families agree on the decision regarding the limitation of life support. Design: Prospective multi-centre cohort study. Setting: Six adult medical-surgical Spanish intensive care units (ICUs). Patients and participants: Three thousand four hundred ninety-eight consecutive patients admitted to six ICUs were enrolled. Measurements and results: Data collected included age, sex, SAPS II score on admission and within 24 h of the decision to limit treatment, length of ICU stay, outcome at ICU discharge, cause and mode of death, time to death after the decision to withhold or withdraw life support, consultation and agreement with patient's family regarding withholding or withdrawal, and the modalities of therapies withdrawn or withheld. Two hundred twenty-six (6.6%) of 3,498 patients had therapy withheld or withdrawn and 221 of them died in the ICU. Age, SAPS II and length of ICU stay were significantly higher in patients dying patients who had therapy withheld or withdrawn than in patients dying despite active treatment. The proposal to withhold or withdraw life support was initiated by physicians in 210 (92.9%) of 226 patients and by the family in the remaining cases. The patient's family was not involved in the decision to withhold or withdraw life support therapy in 64 (28.3%) of 226 cases. Only 21 (9%) patients had expressed their wish to decline life-prolonging therapy prior to ICU admission. Conclusions: The withholding and withdrawing of treatment was frequent in critically ill patients and was initiated primarily by physicians.

Prospective evaluation of patients refused admission to an intensive care unit: triage, futility and outcome

Abstract: Objectives: To evaluate factors associated with decisions to refuse ICU admission and to assess the outcome of refused patients. Design and setting: Prospective, descriptive evaluation in a multi-disciplinary intensive care unit, university referral hospital. Patients and participants: All adult emergency referrals over a 7-month period. Interventions: The number of beds available at the time of referral, the patient's age, gender, diagnosis, mortality probability model score and hospital survival were documented. The outcome of the referral and the reason for refusal were recorded. Measurements and results: Of 624 patients 388 were admitted and 236 (38%) refused. Reasons for refusal were triage (n=104), futility (n=82) and inappropriate referral (too well; n=50). The standardised mortality ratio (SMR) for refused and admitted groups was 1.24 (95% CI 1.05–1.46) and 0.93 (0.78–1.09) respectively. The SMR ratio (refused SMR/admitted SMR) was highest in the middle range of illness (1.95, 1.19–3.20). Inappropriate referrals had a better than expected outcome despite refusal, with a SMR ratio of 0.39 (0.11–0.99). Excluding inappropriate referrals, multivariate analysis demonstrated that refusal was associated with older age, diagnostic group and severity of illness. Triage decisions were associated with a diagnosis of sepsis, and futility decisions with greater severity of illness and recent cardiac arrest. Conclusions: Refusal of admission to our ICU is common. Excess mortality of patients refused is most marked in the middle range of severity of illness. Age, diagnostic group, and severity of illness are important in decision making. Strategies should be developed to create admission criteria that would identify patients in the middle range of severity of illness who should benefit most from ICU care.

Validating the Sedation-Agitation Scale with the Bispectral Index and Visual Analog Scale in adult ICU patients after cardiac surgery.

Abstract: Objective: To validate the Sedation-Agitation Scale (SAS) with the Visual Analog Scale (VAS) and Bispectral Index (BIS) in adult ICU patients after cardiac surgery. Design: Prospective study comparing blinded evaluations of the SAS, VAS and BIS. Setting: Forty-two-bed multidisciplinary ICU. Patients and participants: Convenience sample of 39 adults after cardiac surgery. Measurements and results: Bispectral Index 3.2 was continuously recorded using the Aspect A-1000 and evaluators were blinded to this value. The bedside nurse and a trained researcher independently rated wakefulness using a 100 mm VAS upon patient arrival on the ICU, at first awakening, when ventilator weaning was started and after extubation; the researcher also evaluated patients using SAS. Upon arrival on the ICU, the median SAS score was 2 (interquartile range =1–3), the mean VAS was 26±30 and the mean BIS was 70±16. Twenty-four patients underwent a trial of weaning from mechanical ventilation with a SAS of 4 (IQR =4), VAS of 86±12 and BIS of 87±10. SAS correlated well with VAS performed by one researcher (r=0.91, p<0.001) or by 19 different bedside nurses (r=0.43, p<0.001) and with BIS 3.2 (r=0.60, p<0.001). The correlation between SAS and BIS was reduced in patients with above average electromyogram (EMG) power. As a measure of construct validity, significant differences were noted for the BIS, SAS, VAS and EMG between ICU arrival and extubation (all p<0.001). Conclusions: Sedation-Agitation Scale and BIS are valid measures of wakefulness after cardiac surgery, but EMG interference may affect the accuracy of BIS for a small percentage of patients not receiving neuromuscular blockade.

Acute renal failure: time for consensus.

Abstract: The syndrome of so-called acute renal failure (ARF) is common in the ICU and may affect from 1% to 25% of patients [1, 2] depending on the population and the criteria used to define its presence. In many ways, its nature and epidemiology resemble those of other ICU syndromes, such as severe sepsis, septic shock or ARDS. In this editorial, however, we wish to focus on one important aspect that still distinguishes ARF and which is holding research back in this field of critical care medicine: the lack of consensus criteria to define its presence and severity. We intend to argue that the time has come to address this problem, and we wish to propose some ways in which this could be done. What is ARF in the ICU?

Comparison of procalcitonin with C-reactive protein and serum amyloid for the early diagnosis of bacterial sepsis in critically ill neonates and children.

Abstract: Objectives: To evaluate procalcitonin (PCT) as a diagnostic marker of bacterial sepsis in critically ill neonates and children and to compare the results of PCT with those of C-reactive protein (CRP) and serum amyloid (SAA). Design and setting: Prospective, observational study in neonatal and pediatric intensive care units. Patients: A total of 116 divided into four groups according to age and diagnosis: neonates (aged 3–30 days) with sepsis (n=20), neonates without sepsis (n=26), children (aged 2–12 years) with sepsis (n=32), and children without sepsis (n=38). Interventions: Serum PCT, CRP, and SAA were measured on admission or when a bacterial sepsis was suspected. Area under the receiver operating characteristic (ROC) curve, optimum predictive values, and optimum diagnostic cut off values were evaluated. Results: Admission PCT was significantly higher in neonates and children with sepsis than in the other groups. In the neonates the area under the ROC curve was 0.99 for PCT, 0.95 for CRP, and 0.98 for SAA; in the children it was 1 for PCT, 0.93 for CRP, and 0.96 for SAA. Cutoff concentrations for optimum prediction of sepsis in the neonates were PCT >6.1 ng/ml (diagnostic efficiency: 93.8%), CRP >23.0 mg/l (89.7%), and SAA >41.3 mg/l (95.3%); in the children they were PCT>8.1 ng/ml (100%), CRP>22.1 mg/l (89.8%), and SAA>67.2 mg/l (94.4%). Conclusion: In critically ill children PCT concentration is a better diagnostic marker of sepsis than CRP and SAA. In critically ill neonates, however, PCT, CRP, and SAA are similar diagnostic markers of sepsis. A PCT concentration higher than 8.1 ng/ml identified all children with bacterial sepsis.

Medication administration errors in adult patients in the ICU.

Abstract: Objective: To quantify the incidence and specify the types of medication administration errors from a list of error-prone medications and to determine if patient harm resulted from these errors. Design: An observational evaluation. Setting: Five intensive care units (ICUs) in the United States. Patients and participants: Eight hundred fifty-one patients who were at least 18 years of age and admitted to surgical, medical or mixed ICUs during a 3 month period were included. Interventions: None. Measurements and results: A list of error-prone medications was adapted from the literature and evaluated for medication errors and patient harm. Of 5,744 observations in 851 patients, 187 (3.3%) medication administration errors were detected. the therapeutic classes most commonly associated with errors were vasoactive drugs 61 (32.6%) and sedative/analgesics 48 (25.7%). The most common type of error was wrong infusion rate with 71 (40.1%) errors. Twenty-one errors did not reach the patient and 159 reached the patient but did not result in harm, increased monitoring or intervention. Five errors required increased patient monitoring and two required intervention. None of the errors resulted in patient death. Conclusions: This multicenter evaluation found fewer medication administration errors than the published literature, possibly due to the varying observational techniques and pharmacist involvement. Lorazepam and wrong infusion rates are associated with errors that occurred frequently, resulted in the greatest potential for harm and were common oversights in the system. These errors should be considered potential areas for betterment in the medication use process to improve patient safety.

Non-invasive pressure support ventilation in severe community-acquired pneumonia

Abstract: Objective: To explore three aspects of non-invasive pressure support ventilation (NIPSV) applied by face mask to patients with acute respiratory failure (ARF) due to severe community-acquired pneumonia (CAP): (1) the initial acute effects on respiratory rate, gas exchange and hemodynamics, (2) the clinical course and outcome during ICU and hospital stay, (3) the nursing workload as measured by the daily PRN 87 (Project Research in Nursing) score. Setting: Medical ICU, University Hospital. Design: Prospective, observational study. Patients: Patients without any prior history of chronic lung disease, consecutively admitted to the ICU to receive NIPSV for ARF due to severe CAP. Measurements and results (means ± SD): Twenty-four patients aged 49±17 years, admission APACHE II 13±5, were included. Admission PaO2/FIO2, alveolar-arterial oxygen difference (DA-aO2) and PaCO2 were 104±48, 447±120 and 40±10 mmHg, respectively. All patients were normotensive. During the initial NIPSV trial respiratory rate decreased from 34±8 to 28±10 breaths/min (p<0.001) and arterial oxygenation improved (PaO2/FIO2 104±48 vs 153±49, DA-aO2 447±120 vs 370±180 mmHg, p<0.001) while PaCO2 remained unchanged. There were no hemodynamic effects. Subsequently, a total of 133 NIPSV trials were performed (median duration 55 min, range 30–540 min) over 1–7 days. No complication occurred during NIPSV. Sixteen patients were intubated (66%) 1.3±1 days after inclusion. Upon inclusion, the patients who were subsequently intubated were older (55±15 vs 37±12 years) and more severely hypoxemic (63±11 vs 80±15 mmHg, p<0.05) than those not requiring intubation. Eight patients died (33%), all in the intubated group. Median lengths of stay in the ICU and hospital were longer in intubated patients (ICU 16 days, range 3–64 vs 6 days, range 3–7, p<0.05; hospital 23 days, range 9–77 vs 9.5 days, range 4–42, p<0.05). Mean daily total PRN points were stable throughout the NIPSV period and were not different between the groups. Only 14% of PRN points resulted from respiratory therapy interventions. PRN score was higher during the first 24 h following intubation than during the first 24 h of NIPSV (278±55 vs 228±24 points, p<0.05). Conclusion: Despite initial improvement in arterial oxygenation with NIPSV in patients with ARF due to severe CAP, the intubation rate is high. However, the more favorable outcome and shorter ICU and hospital stays when intubation is avoided, as well as the short delay required to assess the success or failure of NIPSV warrants a trial of NIPSV in this setting. The nursing workload remains stable during NIPSV and does not result predominantly from respiratory therapy interventions.

Using the EuroQol-5D to measure changes in quality of life 12 months after discharge from an intensive care unit.

Abstract: Objective: To compare changes in the health-related quality of life (HRQOL) of critical care patients by diagnostic category. Design: Prospective, cohort study. HRQOL assessed 3 months before admission and 1 year after discharge from the intensive care unit (ICU). Patients were classified as: trauma injury (TI), scheduled surgery (SS), unscheduled surgery (US), and other medical conditions (MC). Setting: Department of Intensive Medicine, University Hospital of Bellvitge, Barcelona, Spain. Patients: Three hundred and thirty-four patients admitted to ICU from October 1994 to June 1995 (62 TI patients, 181 SS patients, 19 US patients, and 72 MC patients). Interventions: Surgical and medical procedures. Measurements and results: Changes in HRQOL varied considerably between diagnostic categories, with TI patients having significantly worse HRQOL one year after discharge from the ICU compared to 3 months prior to admission [change in median EQ Visual Analogue Scale (EQ-VAS) score from 100 to 65, P<0.001], and SS patients reporting improved HRQOL (change in median EQ-VAS scores from 60 to 75, P<0.001). Slight deterioration was observed in the other two diagnostic categories. Twelve months after discharge, the EQ dimension in which the largest proportion of patients in all groups reported problems was usual activities (47% of SS and US patients; 69% of TI patients). Using proxy scores at baseline or follow-up had little effect on results. Conclusions: The degree and direction of change in ICU patients' HRQOL 1 year after discharge depends considerably on diagnostic category. Proxy responses can be reliably used with the EQ-5D when measuring change in HRQOL.

Cerebro-splanchnic oxygenation ratio (CSOR) using near infrared spectroscopy may be able to predict splanchnic ischaemia in neonates

Abstract: Objective: To investigate whether near infrared spectroscopy (NIRS) can detect differences in oxyhaemoglobin signal when applied to the abdomens of neonates with surgically proven splanchnic ischaemia. Design: Prospective, observational cohort study. Setting: Tertiary neonatal referral centre. Patients: Medical and surgical neonates were studied. Two groups were identified, neonates with acute abdomens referred for surgery and those with normal abdomens admitted for medical reasons. Interventions: Tissue oxygenation indexes (TOI) of cerebral and splanchnic regions were measured using near infrared spectroscopy (NIRS) and their relative values expressed as a cerebro-splanchnic oxygenation ratio (CSOR). Measurements were made on admission or immediately prior to surgery and subsequently repeated on a daily basis, whenever possible, until discharge from our unit. The area under the receiver operating characteristic (ROC) curve was evaluated and optimum diagnostic cut-off values determined. Results: Forty neonates were studied: 10 with acute abdomens, including four with necrotising enterocolitis (NEC), 29 controls with normal abdomens and one with cerebral hypoxic ischaemic injury. Median CSOR for the control group was 0.96 (interquartile range 0.83–1.02) whereas the acute abdomen group had a significantly lower median CSOR value of 0.66 (0.45–0.69) (p<0.001). The area under the ROC was 0.91 (95% confidence limits 0.78–1.00) for CSOR. Taking a boundary value of CSOR for the prediction of splanchnic ischaemia as less than 0.75, intestinal ischaemia was identified with a positive predictive value of 0.75 (0.43–0.95) and excluded with a negative predictive value of 0.96 (0.81–1.0). This was a better performance than using abdominal TOI alone. Conclusions: By comparing the TOI of cerebral and splanchnic regions it may be possible to establish the presence of normal splanchnic perfusion and detect when splanchnic ischaemia develops. CSOR had a 90% (56–100%) sensitivity to detect splanchnic ischaemia in neonates. Further work is necessary to confirm these early findings and establish whether abdominal NIRS has a clinical role in detecting splanchnic ischaemia.

Acute cor pulmonale in massive pulmonary embolism : incidence, echocardiographic pattern, clinical implications and recovery rate

Abstract: Objective: The indications for the use of thrombolytic agents in massive pulmonary embolism (MPE) remain controversial and it has been suggested that transthoracic echocardiographic (TTE) examination, which is able to detect an associated right ventricular dysfunction, may cast light on this question. The goal of this study was to examine the incidence of acute cor pulmonale (ACP) in MPE, diagnosed on the basis of TTE criteria, its clinical implications and its resolution rate. Design: Ten-year retrospective clinical study. Setting: A medical and a coronary intensive care unit, university hospital. Patients: One hundred sixty-one patients with proven MPE. Interventions: Acute cor pulmonale was defined as right ventricular end-diastolic area / left ventricular end-diastolic area (RVEDA/LVEDA) ratio in the long axis greater than 0.6 associated with septal dyskinesia in the short axis. ACP patients were divided into three groups according to circulatory status: 32 patients without circulatory failure constituted group 1, 32 patients with circulatory failure requiring inotropic support, but free of metabolic acidosis, constituted group 2 and 34 patients in whom circulatory failure was associated with metabolic acidosis (defined by a base deficit >5 mEq/l) constituted group 3. Results: Acute cor pulmonale was present in 61% of patients with MPE and carried a 23% mortality, but this mortality was very different in stable patients (groups 1 and 2, 64 patients, 3% mortality) and in unstable patients (group 3, 34 patients, 59% mortality). A multivariate logistic regression analysis showed that the TTE results were not predictive of the risk of death. Conversely, the same analysis showed that the presence of metabolic acidosis was a powerful predictor of death. Conclusion: Because none of the TTE measurements in ACP could be used to stratify the severity of MPE, TTE was of no help in deciding on medical thrombolysis. However, depending on its severity, metabolic acidosis could justify a large cooperative study to assess the impact of thrombolytic therapy on mortality rate in this specific group.

Prevalence, Risk Factors and Prevention of Pressure Ulcers in Dutch Intensive Care Units. Results of a Cross-Sectional Survey

Abstract: Objective: Evaluating the prevalence, risk factors and prevention of pressure ulcers in Dutch intensive care units (ICUs). Design: Cross-sectional design. Setting: ICUs of acute care hospitals that participated in the 1998 and 1999 national prevalence surveys. Data were collected on 1 day in each year. Patients: Eight hundred fifty patients admitted to Dutch ICUs. Interventions: None. Measurements and main results: Six categories of data were collected: (1) characteristics of the institution, (2) characteristics of the ward, (3) characteristics of the patients (age, sex, date of admission, reason for admission), (4) risk assessment using the Braden scale and two additional risk factors (malnutrition and incontinence), (5) severity of the pressure ulcers and (6) supportive surface used. The prevalence of pressure ulcers was 28.7%. In a forward logistic regression analysis, four risk factors were significantly associated with the presence of pressure ulcers: infection, age, length of stay and total Braden score. Of the patients at high risk of developing pressure ulcers but without actual pressure ulcers, 60.5% were positioned on a support system. Only 36.8% of the patients who were determined to need repositioning were actually being turned. Conclusions: The prevalence of pressure ulcers in Dutch ICUs is high and their prevention is flawed, especially as regards the use of support systems. Patients for whom turning is indicated are not being turned. Predicting pressure ulcers in ICU patients is difficult and needs further investigation.

Endemicity, molecular diversity and colonisation routes of Pseudomonas aeruginosa in intensive care units.

Abstract: Objective: We carried out a prospective study to evaluate the endemicity of Pseudomonas aeruginosa in intensive care units (ICUs). Pulsed-field gel electrophoresis (PFGE) was used to determine the genotypes of P. aeruginosa isolates. This allowed us to determine the importance of cross-colonisation and the colonisation routes of P. aeruginosa. Design: We screened epidemiological specimens (rectal swab, nose swab and tracheal aspiration) and routine clinical cultures from patients admitted to ICUs during a 2-year period, from 1st January, 1998, to 31st December, 1999. Setting: The study was carried out in four separate adult ICUs located in the Franche-Comte region of France. These four units admitted a total of 1,500 patients per year. Results: A total of 1686 specimens were collected from 473 patients; 122 of these patients were positive on admission, 351 became positive during hospitalisation. The overall incidence of P. aeruginosa was 15.7 cases per 100 patients and 15.1 cases per 1000 days of hospitalisation. Of 184 patients with at least one ICU-acquired positive clinical culture, 104 had been previously identified as carriers by a similar genotype. Typing of 208 non-replicate isolates revealed 101 major DNA patterns. Approximately 50% of P. aeruginosa carriage or colonisation/infection was acquired via cross-transmission; the other cases probably originated from endogenous sources. Conclusion: Cross-colonisation seems to play an important role in the general spread of P. aeruginosa in ICUs.

Human errors in a multidisciplinary intensive care unit: a 1-year prospective study

Abstract: Objectives: To determine the incidence and identify risk factors of critical incidents in an ICU. Design: Prospective observational study of consecutive patients admitted over 1 year to an ICU. Critical incidents were recorded using predefined criteria. Their causes and consequences were analysed. The causes were classified as technical failure, patient's underlying disease, or human errors (subclassified as planning, execution, or surveillance). The consequences were classified as lethal, leading to sequelae, prolonging the ICU stay, minor, or without consequences. The correlation between critical incidents and specific factors including patient's diagnosis and severity score, use of monitoring and therapeutic modalities was analysed by uni- and multivariate analysis. Setting: An 11-bed multidisciplinary ICU in a non-university teaching hospital. Patients: 1024 consecutive patients admitted to the ICU. Intervention: None. Measurementsandmainresults: The median length of ICU stay by the 1024 patients was 1.9 days. Of the 777 critical incidents reported 2% were due to technical failure and 67% to secondary to underlying disease. There were 241 human errors (31%) in 161 patients, evenly distributed among planning (n=75), execution (n=88), and surveillance (n=78). One error was lethal, two led to sequelae, 26% prolonged ICU stay, and 57% were minor and 16% without consequence. Errors with significant consequences were related mainly to planning. Human errors prolonged ICU stay by 425 patient-days, amounting to 15% of ICU time. Readmitted patients had more frequent and more severe critical incidents than primarily admitted patients. Conclusions: Critical incidents add morbidity, workload, and financial burden. A substantial proportion of them are related to human factors with dire consequences. Efforts must focus on timely, appropriate care to avoid planning and execution mishaps at the beginning of the ICU stay; surveillance intensity must be maintained, specially after the fourth day.

Source control in the management of sepsis

Abstract: The process of surgical decision making is based on both general principles that are amenable to evaluation using rigorous techniques of clinical research and the intangible element of surgical judgment that seeks to apply those principles to the care of an individual patient. The role of surgical judgment is inescapable, even though it is intrinsically subjective and recalcitrant to objective evaluation, for a host of factors modify the application of principle in each patient, and render the circumstances of a given problem sufficiently distinctive, that evidence must be tempered with common sense. We have tried to provide, through an evidence-based approach to a series of questions, the rationale for the basic principles that should guide the clinician in initiating or modifying source control, recognizing that sound clinical judgement demands, at times, that these be set aside. In the individual patient, evidence of clinical improvement is the most important marker of the approach selected. Evaluation of the adequacy of source control in the critically ill patient can be difficult. As with other modes of anti-infective therapy, effective source control measures are expected to result in clinical improvement, reflected in: Resolution of clinical signs of sepsis or systemic inflammation. Bacteriological resolution. Evidence of reversal of the metabolic sequelae of infection, with normal progression of wound healing, reflected in the formation of granulation tissue, and epithelialization. Radiographic evidence of control of an infectious focus. Prevention of further organ dysfunction, and resolution of existing organ dysfunction. Survival. Evaluation of the adequacy of source control may necessitate planned reoperation. The adequacy of debridement of necrotizing soft-tissue infections can be assessed by repeat exploration under general anesthesia, continuing the process until there is evidence of healthy granulation tissue throughout the wound. Planned reexploration is also indicated for patients with diffuse intestinal ischemia to ensure bowel viability. The appropriate interventions to determine the adequacy of source control are dictated by the clinical circumstances. A residual or recurrent abscess can usually be demonstrated by CT or ultrasound examination, while resolution of an abscess cavity can be monitored using sinograms. The diagnosis of persistent or evolving tissue necrosis is guided by the clinical setting. Retroperitoneal necrosis can be detected by CT, while sigmoid ischemia following aortic aneurysmectomy can be evaluated by sigmoidoscopy. Occasionally diagnostic peritoneal lavage assists in establishing a diagnosis of gut ischemia; the lavage fluid appears bloody with established ischemia. The diagnosis of an infected foreign body requires an appropriate history and is supported by recurrent bacteremia or by positive cultures drawn retrograde through an indwelling vascular or peritoneal dialysis catheter. Finally, ongoing contamination from a breach of the gastrointestinal tract can be documented by appropriate contrast studies. The general principles that guide the use of source control techniques in the management of the patient with severe sepsis or septic shock are readily articulated. Their implementation in practice, however, is more complex, and does not, as a rule, lend itself to simple algorithms that are applicable in all cases. Moreover evidence-based support for these principles is weak. In the final analysis, the elusive process of experienced surgical judgement is invaluable for all but the most straightforward problems.