Position paper on critical care pharmacy services. Society of Critical Care Medicine and American College of Clinical Pharmacy Task Force on Critical Care Pharmacy Services.

OBJECTIVE: The goal of the Task Force on Critical Care Pharmacy Services was to identify and describe the scope of practice that characterizes the critical care pharmacist and critical care pharmacy services. Specifically, the aims were to define the level of clinical practice and specialized skills characterizing the critical care pharmacist as clinician, educator, researcher, and manager; and to recommend fundamental, desirable, and optimal pharmacy services and personnel requirements for the provision of pharmaceutical care to critically ill patients. Hospitals having comprehensive resources as well as those with more limited resources were considered. DATA SOURCES: Consensus opinion of critical care pharmacists from institutions of various sizes providing critical care services within several types of pharmacy practice models was obtained, including community-based and academic practice settings. Existing guidelines and literature describing pharmacy practice and medication use processes were reviewed and adapted for the critical care setting. CONCLUSIONS: By combining the strengths and expertise of critical care pharmacy specialists with existing supporting literature, these recommendations define the level of clinical practice and specialized skills that characterize the critical care pharmacist as clinician, educator, researcher, and manager. This Position Paper recommends fundamental, desirable, and optimal pharmacy services as well as personnel requirements for the provision of pharmaceutical care to critically ill patients.

A prospective, randomized, controlled evaluation of peripheral nerve stimulation versus standard clinical dosing of neuromuscular blocking agents in critically ill patients.

OBJECTIVES: To determine if vecuronium doses individualized by peripheral nerve stimulation are lower than those doses chosen by standard clinical techniques; and to determine whether patients monitored by peripheral nerve stimulation exhibit shorter recovery times and less prolonged neuromuscular blockade after discontinuation of vecuronium than control patients. DESIGN: A prospective, randomized, controlled, single-blind trial. SETTING: Two ten-bed medical intensive care units of a 937-bed tertiary care, not-for-profit, teaching hospital and health system. PATIENTS: Mechanically ventilated patients requiring continuous neuromuscular blockade as part of their therapy. INTERVENTIONS: After obtaining written, informed consent and baseline neurologic examinations, patients were randomized to treatment, where dosing was individualized by peripheral nerve stimulation or standard clinical assessment. Doses in the peripheral nerve stimulation group were adjusted to 90% blockade (Train-of-Four of 1/4). The standard clinical dosing group received doses individualized to clinical response by the medical team (blinded to Train-of-Four). Differences between groups were evaluated by Wilcoxon matched-pairs signed rank test. MEASUREMENTS AND MAIN RESULTS: A total of 77 patients (35 standard clinical patients vs. 42 peripheral nerve stimulation patients) were enrolled in the study. Despite no difference in initial doses and time to reach 90% blockade or clinical response between groups, the peripheral nerve stimulation group used less drug than the standard clinical group (0.040 +/- 0.028 vs. 0.070 +/- 0.030 mg/kg/hr, respectively, p = .001). The total cumulative amount of vecuronium for the episode of paralysis was greater in the control group (285.8 +/- 246.6 vs. 137.1 +/- 106.4 mg, p = .001). The peripheral nerve stimulation group recovered neuromuscular function (relative risk of 1.85, with 95% confidence interval [CI] of 1.02-3.35, p = .039) and spontaneous ventilation (relative risk of 1.86, 95% CI 1.00-3.45, p = .047) faster than the control group. In patients, adjusting for renal dysfunction, the likelihood of a faster recovery in the peripheral nerve stimulation group increased for neuromuscular function (relative risk of 1.89, 95% CI of 1.07-3.32, p = .018) and spontaneous ventilation (relative risk of 2.27, 95% CI of 1.23-4.21, p = .019). Patients with combined renal and liver failure similarly demonstrated a faster recovery in the peripheral nerve stimulation group. The recovery was affected to a lesser extent by adjusting for concurrent aminoglycoside and corticosteroid administration. CONCLUSIONS: Use of peripheral nerve stimulation for monitoring the degree of blockade and adjusting drug doses in continuously paralyzed critically ill medical patients results in lower doses of vecuronium to maintain a desired depth of paralysis, and allows a faster recovery of neuromuscular function and spontaneous ventilation.

Retrospective pharmacoeconomic evaluation of dosing vecuronium by peripheral nerve stimulation versus standard clinical assessment in critically ill patients.

Adjusting the dosage of vecuronium by peripheral nerve stimulation versus standard clinical dosing in critically ill patients reduces drug requirements to maintain a desired depth of paralysis and, on average, produces faster recovery of neuromuscular function. We retrospectively analyzed the health and economic outcomes of using train-of-four (TOF) end points by peripheral nerve stimulation in dosing neuromuscular blocking agents during continuous infusion in a medical intensive care unit (ICU). A decision-analytic model was used to calculate outcomes and costs of treatment using and not using TOF end points of dosing vecuronium. Data from our TOF trial provided the difference in neuromuscular and functional recovery time. Charges billed by the Patient Financial Services Department were used to determine hourly costs of ICU stay for recovery from neuromuscular blockade using costs:charges ratios estimated from a sample of 20 patients. The cost of vecuronium was determined using the hospital acquisition cost and the actual number of milligrams of drug given to each patient in the TOF trial. The cost of performing one TOF event was determined by timing six events performed by six pharmacists, and randomly selecting 60% of these to calculate a mean time/TOF event. The economic impact of dosing by TOF was determined by calculating the cost savings/patient dosed by TOF compared with those who had doses individualized by standard clinical assessment. One-way and multiway sensitivity analyses were performed to assess model uncertainty. The mean drug cost was $286 in the TOF group versus $580 in the standard dosing group. With a mean time/TOF assessment of 5.8 +/- 1.6 minutes, each episode cost $2.92 for a total TOF cost/patient of $23. At $54/hour of recovery time in the ICU, the estimated cost of ICU care for the TOF group was $34,214 versus $118,681 for the standard group. The estimated costs/patient were $459 and $1197, respectively, for a total cost savings/patient of $738. Sensitivity analyses showed the model to be robust. Estimated annual savings of $146,103 are projected by using TOF to individualize vecuronium doses in patients in the ICU. Individualizing vecuronium doses to TOF end points has both therapeutic and economic advantages. When considering costs of drug, TOF monitoring, and ICU, the total cost/patient was 40% of that in the control group.

Economic impact of prolonged motor weakness complicating neuromuscular blockade in the intensive care unit.

OBJECTIVE: We compared a case-series of ten patients who developed prolonged neuromuscular weakness after continuous, nondepolarizing, neuromuscular blockade with a group of controls without neuromuscular weakness to determine the economic impact of the neuromuscular weakness. DESIGN: Frequency-matched case control trial. SETTING: Medical and surgical intensive care units of a 937-bed tertiary care, university-affiliated teaching hospital. PATIENTS: Ten patients developed prolonged neuromuscular weakness after continuous administration of nondepolarizing neuromuscular blockers. Ten patients from a 1994 drug utilization database who did not develop motor weakness after paralysis were identified to serve as controls. MEASUREMENTS AND MAIN RESULTS: The medical and accounting records of the patients were retrospectively reviewed. Charge data were obtained from patient accounts. Institutional ratios to convert charges to full costs and marginal costs were obtained from the Hospital Finance Department of Henry Ford Hospital. The economic impact of the diagnosis and recovery of the motor weakness was estimated for the intensive care unit (ICU) and hospital stays and compared with those values for control patients. Median hospital charges (excluding rehabilitation), totaling $91,476, were attributed to the patients who developed neuromuscular weakness and included charges for neuromuscular blocking agents, continuous mechanical ventilation, ICU and hospital beds, neurologic studies, and physical therapy services. In the control patients, median charges were $22,191 (p = .001). The total median cost differential for a patient in the neuromuscular weakness group was in excess of $66,713 (95% confidence interval $23,485 to $189,214, p = .001). Significant differences were also found for patient charges, full costs, and marginal costs for mechanical ventilation (p = .002), neurologic studies (p = .014), as well as ICU (p = .002) and hospital (p = .001) stays. CONCLUSIONS: The development of motor weakness was associated with an increase in ICU and hospital stays, continued mechanical ventilation, and disproportionate healthcare expenditures in excess of $66,000 per patient. A prospective evaluation of the true prevalence of neuromuscular weakness after neuromuscular blockade and of the costs to the healthcare system is needed.

Is it time to reposition vasopressors and inotropes in sepsis?

OBJECTIVES: To review the literature on the current use of vasopressors and inotropes in patients with sepsis and sepsis syndrome with respect to the choice of agent, therapeutic end points, and safe and effective doses to be used. To examine the available evidence that supports or refutes goal-directed therapy toward supranormal oxygen transport in optimizing the outcome of critically ill sepsis syndrome patients. DATA SOURCES: All pertinent English and French articles dealing with hemodynamic support with selected vasopressors and inotropic agents in human sepsis and sepsis syndrome retrieved from a computerized MEDLINE search from 1985 to 1994. STUDY SELECTION: Clinical studies with norepinephrine, epinephrine, phenylephrine, dopamine, and dobutamine in sepsis syndrome were considered if goal-directed therapy with oxygen transport variables was utilized. Emphasis was placed on prospective, randomized, controlled comparative trials. However, open-label, observational, and comparative studies, or case series, were also evaluated when limited data were available. DATA EXTRACTION: From the selected studies, information was obtained regarding patient population, dosing regimen, type of therapeutic goals or end points (hemodynamic, or normal vs. supranormal oxygen transport variables) and outcome data (e.g., achievement of goals, resolution of the episode, mortality rate, and development of end-organ dysfunction). DATA SYNTHESIS: When used in larger than usual doses, epinephrine, norepinephrine, and phenylephrine uniformly increased hemodynamic values. Epinephrine may increase oxygen transport values more reliably than norepinephrine. Dobutamine doses in the range of 2.5 to 6 microgram/kg/min increase oxygen transport variables and hemodynamics to predetermined goals in only 30% to 70% of patients. Larger infusion rates offer no further benefits. CONCLUSIONS: Insufficient evidence exists to support goal-directed therapy with vasopressors and inotropes in the treatment of sepsis syndrome. No definitive recommendations can be made about the superiority of a vasopressor or inotropic agent due to the lack of data. However, it may be that evaluation of vasopressors earlier in sepsis syndrome will yield more promising results. Large, comparative, controlled trials assessing mortality rate and development of multiple organ system dysfunction are needed.

Bar code documentation of pharmacotherapy services in intensive care units.

Bar code technology has been used for 5 years to improve the efficiency of identifying and documenting clinical pharmacy services at our institution. Data for an entire year (1993) were analyzed to quantify the nature and magnitude of pharmacy services provided for critically ill patients in intensive care units (ICU). Patients in the medical (MICU), respiratory (RICU), intermediate (IMU), and surgical (SICU) units (3234/3743 patients, 86%) were reviewed. Clinical interventions and expected outcomes were documented by pharmacists using an automated bar code system. There were 11,628 pharmacotherapy interventions, 3.6/patient; 12/pharmacist work day. Of patients whose drug therapy was reviewed at least once, 50% (1610/3234) received at least one intervention. The mean number of interventions/patient was 7.2 in the MICU, 6.1 in RICU, 3.4 in IMU, and 2.4 in the SICU, corresponding to APACHE III scores of 71.2, 66.2, 42.8, and 43.3, respectively. The majority of interventions were to modify dosages of antimicrobial agents, and were performed to achieve optimum efficacy (42%) and to minimize toxicity (46.2%). These data support the necessity for pharmacists to provide individualized care to critically ill patients.

Technical and interpretive problems of peripheral nerve stimulation in monitoring neuromuscular blockade in the intensive care unit.

OBJECTIVE: To review the literature and provide an overview of the technical and interpretive problems associated with peripheral nerve stimulation in monitoring neuromuscular blockade in the intensive care unit. DATA SOURCES: A computerized search on MEDLINE from 1985 through 1994 was performed to identify English-language comparative studies, abstracts, and review articles pertaining to peripheral nerve stimulation, train-of-four monitoring, and neuromuscular blockade in the critical care setting. STUDY SELECTION AND DATA EXTRACTION: Relevant studies in humans were selected and information was extracted on the use of peripheral nerve monitoring in the critically ill. DATA SYNTHESIS: Use of peripheral nerve stimulation is complicated in the intensive care unit. Problems may occur with the patient, the device, as well as operator technique, all of which may lead to errors in interpretation of the depth of paralysis. The critically ill patient has changing comorbid disease states and total body water composition, which may impair the accuracy or reproducibility of measurements. Technical problems relate to the operation of the device, electrode placement, and suboptimal delivery of the desired current. Difficulties in performing peripheral nerve stimulation and interassessor variability contribute to errors of interpretation. CONCLUSIONS: These difficulties compromise the precision, accuracy, and reliability of the peripheral nerve stimulator as a tool for monitoring neuromuscular blockade in the critically ill. Peripheral nerve stimulation should be used in conjunction with clinical parameters to make decisions regarding dose adjustments. Doses should be reduced as much as possible to provide the minimum depth of paralysis that is clinically appropriate. Technical directions and training programs for peripheral nerve stimulation should be developed, and designated individuals should be trained in its application. Large, prospective, controlled studies are necessary to evaluate the incidence of prolonged paralysis or motor neuropathy with administration of neuromuscular blocking agents in patients whose dose is adjusted on the basis of peripheral nerve stimulation.

Disposition of procainamide in patients with chronic congestive heart failure receiving medical therapy.

Dosage reduction of procainamide has been recommended in patients with congestive heart failure (CHF). However, these recommendations are based primarily on studies with unmatched control groups, suboptimal blood sampling, and in patients not receiving angiotensin-converting enzyme (ACE) inhibitors. These agents increase renal blood flow, which theoretically may offset alterations in drug disposition in patients with CHF. The pharmacokinetics of procainamide in patients with chronic CHF and in matched controls were compared. A single intravenous dose of 750 mg of procainamide was administered to 9 patients with chronic New York Heart Association (NYHA) class II or III CHF (mean +/- SD left ventricular ejection fraction 22 +/- 9%) receiving medical therapy and 7 control subjects matched for age and gender. Blood and urine samples were collected at intervals over a period of 48 and 72 hours, respectively. Patients with CHF and control subjects were demographically similar, with the exception of concomitant medications, including ACE inhibitors (8/9 versus 1/7, respectively). There were no significant differences between patients with CHF and control subjects in mean +/- SD peak serum concentrations (Cmax), area under the serum concentration-time curve (AUC0-infinity), total clearance, renal clearance, half-life (t1/2), or volume of distribution (Vd) of procainamide. Similarly, there were no significant differences between patients with CHF and control subjects in the mean +/- SD Cmax, AUC0-infinity, renal clearance, or t1/2 of N-acetylprocainamide (NAPA). Procainamide dosage reduction may not be necessary in patients with chronic stable CHF who are receiving medical therapy.

Inhibition of N-acetylation of procainamide and renal clearance of N-acetylprocainamide by para-aminobenzoic acid in humans.

Procainamide administration often results in excessively high serum N-acetylprocainamide (NAPA) concentrations and subtherapeutic serum procainamide concentrations. Inhibition of N-acetylation of procainamide may prevent accumulation of excessive NAPA while maintaining therapeutic serum procainamide concentrations. The purpose of this randomized, two-way crossover study was to determine if para-aminobenzoic acid (PABA) inhibits N-acetylation of procainamide in healthy volunteers. Eleven (7 female, 4 male) fast acetylators of caffeine received, in random order, PABA 1.5 g orally every 6 hours for 5 days, with a single intravenous dose of procainamide 750 mg administered over 30 minutes on the third day, or intravenous procainamide alone. Blood samples were collected during a 48-hour period after initiation of the infusion. Urine was collected over a 72-hour period. Serum procainamide and NAPA concentrations were analyzed using fluorescence polarization immunoassay. Urine procainamide and NAPA concentrations were measured with high performance liquid chromatography. PABA did not significantly influence total or renal procainamide clearance, elimination rate constant, AUC0-00, amount of procainamide excreted unchanged in the urine, or volume of distribution. However, concomitant PABA administration with procainamide resulted in increases in NAPA AUC0-00 and t1/2 and reductions in NAPA Ke, procainamide acetylation (NAPA formation) clearance, and NAPA renal clearance. Although PABA inhibits metabolic conversion of procainamide to NAPA, it also impairs the renal clearance of NAPA (but not procainamide) in healthy subjects. Therefore, PABA may not be useful for optimizing the safety of efficacy of procainamide in patients.