Eric W. Mueller

An analgesia-delirium-sedation protocol for critically ill trauma patients reduces ventilator days and hospital length of stay.

BACKGROUND Analgesics and sedatives are required to maintain a calm and comfortable mechanically ventilated injured patient. Continuous sedative infusions have been shown to lengthen mechanical ventilation and hospital length of stay. Daily interruption of sedative infusions may reduce both of these variables. Implementation of an Analgesia-Delirium-Sedation (ADS) Protocol using objective assessments with a goal of maintaining an awake and comfortable patient may obviate the need for daily interruption of infusions in critically ill trauma patients. We examined the effects of such a protocol on ventilator duration, intensive care unit (ICU) length of stay, hospital slength of stay, and medication requirements. METHODS A multidisciplinary team designed the protocol. Objective measures of pain (visual/objective pain assessment scale-VAS/OPAS), agitation (Richmond Agitation-Sedation Scale-RASS), and delirium [Confusion Assessment Method {CAM-ICU}] were used. Medications were titrated to a RASS of -1 to +1 and VAS/OPAS <4. Haloperidol was used to treat delirium in CAM-ICU positive patients. Retrospective review of the local Project IMPACT database for a 6-month period in 2004 was compared with the same seasonal period in 2006 in which the ADS protocol was used. All mechanically ventilated trauma patients receiving infusions of narcotic, propofol, or benzodiazepine were included. Age, APACHE II score, Injury Severity Score, ventilator days, ventilator-free days at day 28, ICU length of stay, and hospital length of stay are reported as median values (interquartile range). Medication usage is reported as mean values (+/-SD). Differences in data were analyzed using Wilcoxons rank-sum test or t test, as appropriate. Gender, mortality, and mechanism of injury were analyzed using chi analysis. RESULTS A total of 143 patients were included. Patients who died during their hospitalization were excluded except in the analysis of ventilator-free days at day 28. After exclusions, 61 patients were in the control group and 58 in the protocol group. The median duration of mechanical ventilation in the protocol group was 1.2 days (0.5-3.0) which was significantly reduced compared with 3.2 days (1.0-12.9) in the control group (p = 0.027). Analysis of ventilator-free days at day 28 found that the protocol group had 26.4 ventilator-free days (13.9-27.4) compared with 22.8 days (10.5-26.9) in the control group (p = 0.007). The median ICU length of stay was 5.9 days (2.3-18.2) in the control group and 4.1 days (2.5-8.3) in the protocol group (p = 0.21). Hospital length of stay was 12 days (7-17) in the protocol group in contrast to 18 days (10-27) in the control group (p = 0.036). Opiate equivalents and propofol use per patient was significantly reduced in the protocol group from 2,465 mg (+/-1,242 mg) to 1,641 mg (+/-1,250 mg) and 19,232 mg (+/-22,477 mg) to 10,057 (+/-14,616 mg), respectively (p < 0.001, p = 0.01). CONCLUSION An objective assessment- based ADS protocol without daily interruption of medication infusion decreases ventilator days and hospital length of stay in critically ill trauma patients.

The impact of preinjury antiplatelet and anticoagulant pharmacotherapy on outcomes in elderly patients with hemorrhagic brain injury

BACKGROUND More elderly trauma patients are identified with preinjury use of clopidogrel, aspirin, or warfarin (CAW). The purpose of this study was to determine whether preinjury CAW use was an important predictor of mortality in patients aged >or=50 years with blunt, hemorrhagic brain injury (HBI). METHODS A retrospective review of patients with blunt, HBI aged >or=50 years with subgroup analysis for older (>70 years) and younger (50-70 years) patients was performed. CAW use was analyzed for differences in age, gender, hospital length of stay (LOS), Injury Severity Score (ISS), Glasgow Coma Score (GCS), mechanism of injury (MOI), platelet transfusion therapy (PLT), disposition at discharge, and in-hospital mortality. RESULTS From January 2003 to October 2005, 416 patients were identified. The mean age was 69+/-1 years. No differences were found for ISS (24 +/- 0.5), GCS (12 +/- 0.2), or LOS (8 +/- 0.4 days). CAW use was present in 40% of patients and significantly higher in older patients. Mortality was not different between older and younger CAW(+) patients, but it significantly increased for older CAW(-) patients. Significant predictors of death included age, ISS, and GCS (P<.02). CONCLUSIONS Preinjury CAW use in older blunt, HBI patients is not associated with increased mortality. Age was a significant predictor of mortality independent of CAW use.

Effect from multiple episodes of inadequate empiric antibiotic therapy for ventilator-associated pneumonia on morbidity and mortality among critically ill trauma patients.

BACKGROUND Empiric antibiotic therapy is routinely initiated for patients with presumed ventilator-associated pneumonia (VAP). The impact of inadequate empiric antibiotic therapy (IEAT) may vary among critically ill populations. The purpose of this retrospective study was to determine the effect of IEAT on the outcome for adult trauma patients with VAP. METHODS This study enrolled 82 patients with multiple VAP episodes (200 VAP episodes; mean, 2.4 +/- 0.65 per patient; range, 2-5 episodes). An episode of IEAT was a VAP episode with empiric therapy having no in vitro activity against causative bacteria. Overall, there were 78 (39%) IEAT episodes involving 54 patients. Most often, IEAT was attributable to the presence of Acinetobacter spp, Stenotrophomonas maltophilia, or Alcaligenes xylosoxidans. All the patients received appropriate definitive therapy according to the final culture. The patients were classified by number of IEAT episodes: 0 (n = 28), 1 (n = 34), and more than 1 (n = 20). RESULTS Demographics and injury severity were similar among the groups. The mortality rate was 3.6% for no episodes, 8.8% for one episode, and 45% for more than one episode (p < 0.001). On the basis of multiple logistic regression, experiencing multiple IEAT episodes was independently associated with the risk of death (odds ratio, 4.28; 95% confidence interval, 1.44-12.71). Additionally, experiencing multiple IEAT episodes was associated with prolonged intensive care unit stay (p = 0.007) and prolonged mechanical ventilation (p = 0.005). CONCLUSIONS Critically ill trauma patients experiencing multiple episodes of IEAT for VAP have increased morbidity and mortality. These findings reinforce the importance of developing and refining a unit-specific pathway for the empiric management of VAP.

The appropriate diagnostic threshold for ventilator-associated pneumonia using quatititative cultures

BACKGROUND The use of quantitative cultures of the bronchoalveolar lavage (BAL) effluent to distinguish between posttraumatic inflammatory response and ventilator-associated pneumonia (VAP) is becoming more common. However, the diagnostic threshold of either 10 or 10 colonies/mL remains debatable. Because mortality from VAP is related to treatment delay, some have chosen a lower diagnostic threshold (>10 colonies/mL). This may result in unnecessary antibiotic use with its sequelae: increased resistant organisms, antibiotic-related complications, and increased costs. The purpose of this study is to determine the optimal diagnostic threshold for VAP diagnosis using quantitative cultures of the BAL effluent. METHODS Data on patients with fiberoptic bronchoscopy with BAL are maintained in a prospectively collected database at our Level I trauma center. This database was reviewed for timing and frequency of BAL and the colony counts of each organism identified. Indication for bronchoscopy was clinical evidence of VAP. VAP was defined as >10 colonies/mL in the BAL effluent. A false-negative BAL was defined as any patient who had <10 colonies/mL and developed VAP with the same organism up to 7 days after the previous culture. RESULTS Over a 46-month period, 526 patients underwent 1,372 fiberoptic bronchoscopy procedures with BAL. Of these, 72% were male patients, 91% followed blunt injury, and mean age and Injury Severity Score were 43 years and 30, respectively. Overall mortality was 14%. There were 1,898 organisms identified (42% were gram-positive and 58% were gram-negative). VAP was diagnosed in 38% of BAL. Overall, there were 43 episodes in 38 patients defined as false-negative (3%). The false-negative rate was 9% in patients with 10 organisms. The most common false-negative organisms were Pseudomonas and Acinetobacter species. CONCLUSION The VAP diagnostic threshold for quantitative BAL in trauma patients should be >10 colonies/mL. One may consider a threshold of >10 colonies/mL in severely injured patients with Pseudomonas or Acinetobacter species.

Sunitinib-Related Fulminant Hepatic Failure: Case Report and Review of the Literature

Drug‐induced hepatotoxicity is an infrequent but life‐threatening complication. Sunitinib is a multitargeted receptor tyrosine kinase inhibitor approved for treatment of renal cell carcinoma and gastrointestinal stromal tumor. However, results from preapproval clinical trials suggest an equivocal hepatic risk profile for sunitinib. We describe a 75‐year‐old woman with renal cell carcinoma who was admitted to the intensive care unit after experiencing fulminant hepatic failure during sunitinib therapy. The patients hepatic and renal chemistries had been within normal limits throughout four previous cycles of sunitinib therapy spanning 9 months. After the fifth cycle, she complained of a 3‐day history of severe diarrhea and dehydration. Her abnormal laboratory test results included the following: total bilirubin level 5.9 mg/dl, aspartate aminotransferase level 3872 U/L, alanine aminotransferase level 3332 U/L, ammonia level 897 μg/dl, and an international normalized ratio of 4.8. Use of the Naranjo adverse drug reaction probability scale indicated a possible relationship between sunitinib and hepatotoxicity. Supportive care including aggressive intravenous hydration and reversal of coagulopathy was successful. The patient was discharged home on hospital day 7 without apparent longstanding sequelae. Clinicians should be aware of this possible adverse effect of sunitinib, and continued pharmacovigilance is imperative to accurately quantify the possible risk of sunitinib‐related hepatotoxicity.

Repeat bronchoalveolar lavage to guide antibiotic duration for ventilator-associated pneumonia.

BACKGROUND Using an arbitrary day cutoff or clinical signs to decide the duration of antibiotic therapy for ventilator-associated pneumonia (VAP) may be suboptimal for some patients. We sought to determine whether antibiotic duration for VAP can be safely abbreviated in trauma patients using repeat bronchoalveolar lavage (BAL). METHODS This was an observational case-controlled pilot study. Fifty-two patients were treated for VAP using a repeat BAL clinical pathway. Definitive antibiotic therapy for VAP was discontinued if pathogen growth was <10,000 colony forming units/mL on repeat BAL performed on day 4 of antibiotic therapy (responder), otherwise therapy was continued per managing team. A matched control group of 52 VAP patients treated before (immediately consecutive) the pathway was used for comparison. RESULTS Antibiotic duration in pathway patients was shorter than control patients (9.8 days +/- 3.8 days vs. 16.7 days +/- 7.4 days; p < 0.001), including nonfermenting gram-negative bacilli VAP (10.7 days +/- 4.1 days vs. 14.4 days +/- 4.2 days; p < 0.001). There were no differences in pneumonia relapse, mechanical ventilator-free intensive care unit (ICU) days, ICU-free hospital days, or mortality. Of study group isolates, 86 (82.7%) responded on repeat BAL and were treated for 8.8 days +/- 3.3 days. Of these without concomitant infections (n = 65), antibiotic duration was 7.3 days +/- 1.2 days compared with 14.4 days +/- 2.6 days for nonresponding isolates (n = 18) (p < 0.001). CONCLUSIONS Repeat BAL decreased the duration of antibiotic therapy for VAP in trauma patients. More adequately powered investigations are needed to appropriately determine the effects of this strategy on patient outcome.

Effects of Critical Illness and Organ Failure on Therapeutic Argatroban Dosage Requirements in Patients with Suspected or Confirmed Heparin-Induced Thrombocytopenia

BACKGROUND Critically ill patients often require therapeutic argatroban dosages lower than those recommended in package labeling. The magnitude of dosage alteration in relation to severity of organ failure is unknown. OBJECTIVE To compare therapeutic argatroban dosages between critically ill and noncritically ill patients with confirmed or suspected heparin-induced thrombocytopenia and investigate the relationship between total Sequential Organ Failure Assessment (SOFA) score and therapeutic argatroban dosage. METHODS This retrospective cohort study was conducted at an urban academic medical center. Adults without Child-Pugh class C hepatic dysfunction who received argatroban for more than 24 hours over a 3-year period were included. Therapeutic argatroban dosage was that resulting in 2 consecutive activated partial thromboplastin time (aPTT) values 1.5–3 times the patient-specific baseline obtained at least 4 hours apart. Initial argatroban dosages were at the discretion of the managing service. RESULTS Fifty-three patients (critically ill, n = 34; noncritically ill, n = 19) were included. Critically ill patients had higher median [interquartile range] Acute Physiology and Chronic Health Evaluation (APACHE II) (17 [12–21] vs 10 [3.25–17.75]; p = 0.007) and SOFA (11 [7–13] vs 2 [0–2.75]; p < 0.001) scores. Critically ill patients required lower mean ± SD therapeutic argatroban dosage (0.6 ± 0.5 vs 1.4 ± 0.9 µg/kg/min; p < 0.001). There was no significant difference in time to therapeutic aPTT or proportion of aPTTs within therapeutic range. Argatroban dosage was inversely related to SOFA score tertiles (<6: 1.34 ± 0.82 µg/kg/min; 6–9: 0.93 ± 0.54; ≥10: 0.40 ± 0.27; p < 0.001). Total SOFA score at the time of argatroban initiation was independently associated with an argatroban dosage less than 0.75 µg/kg/min (OR 1.5, 95% CI 1.2 to 1.8; p < 0.001). Adverse events were similar between groups. CONCLUSIONS Critically ill patients with single or multiple organ failure require lower therapeutic argatroban dosages compared with noncritically ill patients. Because of an inverse relationship with SOFA score, initial argatroban dosage in critically ill patients should be based on the presence and magnitude of organ failure.

High-Dose Ciprofloxacin for Serious Gram-Negative Infection in an Obese, Critically Ill Patient Receiving Continuous Venovenous Hemodiafiltration

Objective: To describe the pharmacokinetic profile and clinical outcome associated with high-dose ciprofloxacin therapy in a patient with the triad of extreme obesity, multiple organ failure, and deep-seated infection. Case Summary: A 45-year-old, class 3 obese (185 kg; body mass index 53.7), critically ill trauma patient receiving continuous venovenous hemodiafiltration (CVVHOF) was treated with ciprofloxacin 800 mg intravenously every 12 hours for presumed Enterobacter aerogenes (ciprofloxacin minimum inhibitory concentration [MIC] ≤1 μg/mL) lumbar spine osteomyelitis. Four sequential plasma ciprofloxacin samples were obtained and analyzed to determine the steady-state pharmacokinetic profile. The observed steady-state maximum (Cmax) and calculated minimum (Cmin) ciprofloxacin plasma concentrations measured on treatment day 8 were 13 μg/mL and 4.8 μg/mL, respectively, corresponding to an estimated half-life, area under the curve (AUC0-24), total systemic clearance, and clearance by CVVHDF of 7.6 hours, 132 μg·h/mL, 139 mL/min, and 26 mL/min, respectively. These concentrations produced AUC0-24/MIC ratios >125 and plasma Cmax/MIC ratios >10 for MICs ≤1 μg/mL. Intravenous colistin and polymyxin B lumbar wound irrigation were initiated on ciprofloxacin days 12 and 15, respectively, for concomitant multidrug-resistant Acinetobacter baumannii infection. Lumbar tissue cultures on day 24 of ciprofloxacin therapy demonstrated no growth, coinciding with overall improvement of the invasive wound, A week later, the patient developed worsening septic shock and died secondary to an occult subdiaphragmatic abscess. Discussion: Pharmacodynamic outcome studies suggest that AUC0-24/MIC ratios >125 and plasma Cmax/MIC ratios >10 are good predictors of clinical and microbiologic success of ciprofloxacin against gram-negative pathogens. These pharmacodynamic goals were achieved in the plasma with high-dose ciprofloxacin for MICs ≤1 μg/mL. Conclusions: Critically ill obese patients with deep-seated infection involving organisms with MICs >0.5 μg/mL likely require ciprofloxacin dosages greater than traditional daily doses of 400–800 mg during CVVHDF to achieve optimal pharmacodynamic targets.

The Use of Extended-Interval Aminoglycoside Dosing Strategies for the Treatment of Moderate-to-Severe Infections Encountered in Critically Ill Surgical Patients

BACKGROUND Extended-interval dosing strategies have been developed to exploit the concentration-dependent bactericidal activity and time-dependent host toxicity associated with aminoglycoside the therapy. The ability of published extended-interval dosing nomograms to achieve optimal pharmacodynamic endpoints may be limited in certain critically ill surgical patients. METHODS Review of pertinent English language literature. Presentation of descriptive, graded recommendations for extended-interval aminoglycoside dosing in critically ill surgical patients. RESULTS Aminoglycoside dosing considerations in critically ill surgical patients should attempt to maximize the bacterial and host pharmacodynamic attributes of these agents. Empirically, extended-interval aminoglycoside doses proposed by published nomograms are reasonable for most patients; however, because of clinically meaningful variations in aminoglycoside pharmacokinetics, routine use of published extended-interval aminoglycoside dosing nomograms to determine an appropriate dosage interval is discouraged in many critically ill surgical patients. Critically ill surgical patients receiving extended-interval aminoglycoside dosages should undergo individualized pharmacokinetic analysis to characterize efficiently and more effectively plasma concentration-to-bacterial minimum inhibitory concentration (MIC) relationships and determine an appropriate dosing interval, considering site and severity of infection, plasma clearance, and the apparent post-antibiotic effect. CONCLUSIONS The use of extended-interval aminoglycoside dosage regimens in critically ill surgical patients should be based on pharmacodynamic endpoints and patient-specific pharmacokinetic assessment.

Effect of a dalteparin prophylaxis protocol using anti-factor Xa concentrations on venous thromboembolism in high-risk trauma patients

BACKGROUND Low anti-factor Xa (anti-Xa) concentrations with twice-daily enoxaparin are associated with venous thromboembolism (VTE) in high-risk trauma patients. Concerns have been raised with once-daily dalteparin regarding effectiveness and achievable anti-Xa concentrations. The purpose of this before-and-after study was to evaluate the effectiveness of a VTE prophylaxis protocol using anti-Xa concentrations and associated dalteparin dose adjustment in high-risk trauma patients. METHODS Adult trauma patients receiving VTE chemoprophylaxis and hospitalized for at least 3 days were prospectively followed during two 6-month epochs before (PRE) and after (POST) implementation of anti-Xa monitoring. In both groups, high-risk patients received dalteparin 5,000 U subcutaneously once daily; low-risk patients received subcutaneous unfractionated heparin. High-risk POST patients with anti-Xa less than 0.1 IU/mL 12 hours after initial dalteparin dose received dalteparin every 12 hours. All patients underwent routine VTE ultrasound surveillance of the lower extremities. The primary outcome was incidence of VTE. RESULTS A total of 785 patients (PRE, n = 428; POST, n = 357) were included. Demographics, injury patterns, Injury Severity Score (ISS), red blood cell transfusions, intensive care unit and hospital stays, and mortality did not differ between groups. Overall, POST patients had lower VTE (7.0% vs. 13%, p = 0.009) including acute VTE (6.4% vs. 12%, p = 0.01) and proximal deep vein thromboembolism (2.2% vs. 5.7%, p = 0.019). Between high-risk patients, VTE occurred in 53 (16.3%) PRE compared with 24 (9.0%) POST patients (p = 0.01); there was no difference in VTE between low-risk patients (PRE, 2.0% vs. POST, 1.1%; p = 0.86). Among 190 high-risk POST patients with anti-Xa, 97 (51%) were less than 0.1 IU/mL. Patients with low anti-Xa had higher rates of VTE (14.0% vs. 5.4%, p = 0.05) and deep vein thromboembolism (14.4% vs. 3.2%, p = 0.01). Younger age (odds ratio, 0.97; 95% confidence interval, 0.95–0.99) and greater weight (odds ratio, 1.02; 95% confidence interval, 1.00–1.03) predicted low anti-Xa on multivariate regression. CONCLUSION A VTE prophylaxis protocol using anti-Xa–based dalteparin dosage adjustment in high-risk trauma patients was associated with decreased VTE. Once-daily dalteparin 12-hour anti-Xa concentrations are suboptimal in a majority of patients and associated with VTE. LEVEL OF EVIDENCE Therapeutic study, level IV.