J. Lance Lichtor, M.D., and Joseph F. Antognini, M.D., Editors
The reality of drug shortages—the case of the injectable agent propofol. N Engl J Med 2010; 363:806–7
Propofol—a sterile injectable, fast-onset, short-acting sedative-hypnotic—is often used for the induction and maintenance of anesthesia or sedation. However, because of manufacturing concerns, two of the three producers of propofol recalled multiple batches of the drug. At the time the current article was published, these companies have not provided additional supplies and one of them has permanently stopped making the drug.
Nearly half (46%) of all drug shortages in 2009 involved sterile injectables, which tend to be particularly susceptible to shortages. These types of drugs are difficult to make, require a long manufacturing lead-time, and generally are not produced in excess of current market demands. Furthermore, reduced profit margins from generic competition may lead to lower production of these drugs.
Clinicians and the entire health care community should rapidly communicate any potential drug shortages to the Food and Drug Administration as soon as possible. Furthermore, the Food and Drug Administration strongly encourages all health care professionals to take extra care in reading labels during periods of drug shortages because substitutes may have different administration requirements.
In the past year, drug shortages have become commonplace. The 2010 propofol shortage has been particularly frustrating for anesthesiologists. The authors noted that no immediate remedy for this problem is expected. It is suggested that anesthesiologists be aware of shortages before they become critical so that proper measures can be undertaken to minimize their impact.
Primary payer status affects mortality for major surgical operations. Ann Surg 2010; 252:544–51
Patient socioeconomic status is an independent predictor of operative mortality. Ann Surg 2010; 252:552–7
Numerous studies have demonstrated that poor socioeconomic status is associated with increased postsurgical mortality. However, the specific factors that contribute to this disparity remain unclear.
Two retrospective studies analyzed records from the Healthcare Cost and Utilization Project Nationwide Inpatient Sample database to determine the impact of primary payer status and patient socioeconomic status, respectively, on surgical outcomes. The first study examined 893,658 records, whereas the second analyzed 1,059,003 records.
In the first study, patients with private insurance had the lowest incidence of all in-hospital complications, except systemic complications, and the lowest mortality rates when compared with Medicare, Medicaid, and uninsured patients. Patients with Medicaid had the highest incidence of postoperative wound complications, infections, gastrointestinal complications, and systemic complications. They also had the longest mean hospital length of stay and highest total costs.
The second study identified age, patient socioeconomic status, hospital procedure-specific volume, and hospital socioeconomic status as core predictors of mortality.
Many patient comorbidities and characteristics increase risk of complications and mortality after surgery. These two studies examined a large sample of patient records. One study found that, compared with patients who had private insurance, risk-adjusted mortality was increased in patients who had Medicaid, Medicare, or no health care insurance. The second study found that lower socioeconomic status increased operative mortality, independent of comorbidities. Insurance status appears to be a marker for broader socioeconomic factors that affect postoperative patient outcomes.
Outcome for middle cerebral artery aneurysm surgery. Neurosurgery 2010; 67:755–61
Interpretations from multicenter trials are often confounded by the inherent variability in procedure specifics and patient selection across centers. To better understand adverse events associated with surgical intervention for repair of unruptured middle cerebral artery (MCA) aneurysms, a retrospective study was conducted using data from one surgeon. This data was collected prospectively for 10 yr.
Patients who had at least one surgically treated unruptured MCA aneurysm were included. Patients underwent one of the following interventions: simple clipping, suture repair of the aneurysm neck, suture repair supplemented with clipping, or trap with bypass surgery.
Of 1,422 unruptured aneurysms in 1,039 consecutive patients, data from 263 patients with 339 unruptured MCA aneurysms were included. Average patient age was 53 yr, most were female, and most aneurysms were 5–6 mm in diameter. Increased patient age and aneurysm size were significantly correlated with increased surgical risk (P = 0.001 and P = 0.02, respectively). Aneurysms larger than 12 mm in size contributed to 25% of surgical downgrades. Patients older than 60 yr with aneurysms larger than 12 mm had a combined morbidity and mortality rate of 22.2% compared with 9.3% for patients in the low-risk group.
Cerebral aneurysms are often life-threatening and usually require surgical intervention. In this study of patients undergoing clipping of MCA aneurysms, two patient factors were associated with increased morbidity and mortality: age greater than 60 yr and aneurysm size larger than 12 mm. This information can be used to counsel patients regarding prognosis.
Critical Care Medicine
Jean Mantz, M.D., Ph.D., Editor
Glucose control in severely thermally injured pediatric patients: What glucose range should be the target? Ann Surg 2010; 252:521–8
Burn patients with hyperglycemia are at higher risk for bacteremia or fungemia and mortality. Studies have shown conflicting results regarding the benefit of euglycemic control, through insulin administration, in critically ill patients. In addition, the optimal target glucose range is unclear.
To determine if hypoglycemia and hyperglycemia are detrimental to clinical outcomes for severely injured burn victims, a randomized, prospective trial was conducted in thermally injured children. These children had burns covering 30% of their total body surface area and required at least one surgery. Patients received either tight euglycemic, conventional insulin, or no anabolic or anticatabolic experimental drugs (control). Daily blood glucose concentrations were collected to divide patients into good (less than 130 mg/dL for more than 75% of hospital stay) versus poor (more than 130 mg/dL for less than 75% of hospital stay) glucose control groups.
Most (148 of 208) patients had poor glucose control, were Hispanic (86%), and male (63%). Average patient age was 9 yr. Burn size, daily average glucose, and 6:00 am glucose concentrations were predictors for mortality (P ≤ 0.001). Patients with average glucose concentrations of 140 mg/dL for 70% of their hospitalization had improved outcomes. Patients with good glucose control had significantly decreased incidence of minor infections, sepsis, and multiorgan failure (P < 0.05). No patients in the good glucose control group died, compared with 12% of patients in the poor glucose control group. Patients in the poor glucose control group required more insulin (P < 0.05) and had a significant increase in resting energy expenditure (P < 0.05).
Although strict glucose control is an emerging target for patients undergoing surgery, it is unclear if such paradigms are appropriate for all patients. In pediatric patients with total body surface area burns of 30% or more, maintaining glucose concentrations at 130 mg/dl improved clinical outcomes, including survival.
Pentobarbital coma for refractory intra-cranial hypertension after severe traumatic brain injury: Mortality predictions and one-year outcome in 55 patients. J Trauma 2010; 69:275–83
Pentobarbital coma has been shown to lower intracranial pressure among patients with intracranial hypertension after traumatic brain injury. Although pentobarbital coma is not widely used because of complications and adverse effects, it may be beneficial as an add-on therapy in patients for whom traditional methods of reducing intracranial pressure have failed.
A retrospective cohort analysis was conducted to determine efficacy and long-term functional outcomes for patients with severe traumatic brain injury who were treated with pentobarbital coma after maximal medical management and hyperosmolar therapy failed. Review of electronic and paper medical records from a single institution was conducted along with direct patient and/or caregiver contact (for long-term outcomes) during 7 yr.
Of 4,934 patients screened, 55 patients were included in the study. Mean patient age was 16.8 yr. Most injuries were a result of motor vehicle collisions (64%). Overall mortality rate was 64%. For those who survived to discharge, the mean length of stay in the intensive care unit was 16 days; mean hospital length of stay was 23 days. Of the 19 patients available for long-term follow up, 13 (68%) had “good outcomes” (i.e. , Glasgow Outcome Scale score, 4 or 5).
The use of barbiturates in critically ill patients has been challenged. Although this study was a retrospective cohort analysis, these data indicate that satisfactory outcomes were obtained using high doses of pentobarbital as a continuous infusion for a significant proportion of young patients with refractory posttraumatic intracranial hypertension. These results suggest that barbiturates may yet have utility in the treatment of severe traumatic brain injury with refractory intracranial hypertension.
Impact of simulation-based learning on medication error rates in critically ill patients. Intensive Care Med 2010; 36:1526–31
Despite implementation of several new interventions (e.g. , advanced infusion pumps, unit-dose bar coding), up to 78% of all serious hospital errors are medication related. Communication errors, such as dissemination of drug information, may account for 29% of medication errors.
A single-center, parallel, controlled, prospective study was conducted to determine the effects of simulation-based learning on the rate of medication administration errors in a medical intensive care unit versus coronary care unit. Nurses with adult patients were observed administering medication before and after attending a 20–30-min educational session. Content for educational sessions was determined based on observations obtained during baseline monitoring. Sessions were didactic (intensive care unit nurses, n = 12) or simulation-based (coronary care unit nurses, n = 12).
Observations included 880 doses administered to 76 patients. Compared with baseline, errors were reduced among simulation-educated nurses only. Error reduction rate was sustained in the simulation group, whereas error rate continued to increase among didactic-educated nurses.
Medication errors contribute to significant mortality and morbidity in intensive care units. Reducing the rate of these errors through education is, therefore, an important part of improving care for critically ill patients. This single-center study suggests that simulation-based learning can decrease medication error rates better than didactic lectures.
Cognitive improvement during continuous sedation in critically ill, awake and responsive patients: The Acute Neurological ICU Sedation Trial (ANIST). Intensive Care Med 2010; 36:1505–13
Although continuous anxiolytic sedation is often required for intensive care unit (ICU) patients, it may result in long-term adverse effects, including impaired cognitive function. Dexmedetomidine (DEX), an α2-agonist, may provide ICU sedation without the adverse effects associated with anxiolytics.
A prospective, randomized, double-blinded, crossover study was conducted to compare the efficacy of propofol and DEX in awake, intubated ICU patients regardless of brain injury status. As patients were observed, cognitive evaluations were made at baseline, 30 min after sedation (propofol vs . DEX), and 30 min after achieving sedation with the alternate drug after a 3-h washout period. Patients received fentanyl, as necessary, for analgesia.
Mean patient age was 54 yr (N = 33). Most patients received insulin (85%) or anticoagulants (82%). Although adapted cognitive exam scores were diminished by 12.38% after sedation with propofol, DEX-treated patients had scores that were improved by 6.81%—for a between-group difference of 19.19%. This effect was maintained regardless of demographic covariates, concomitant fentanyl treatment, or treatment sequence. Baseline adapted cognitive exam scores did not change after propofol, but were positively correlated with those after treatment with Dexmedetomidine. The incidence of delirium was low in both ICU groups (n = 1).
The Acute Neurologic ICU Sedation Trial is a small prospective, randomized, double-blind, controlled trial. It showed that targeted sedation with DEX can improve cognitive function in non–brain-injured, critically ill patients. This is the first report supporting a beneficial role for different sedation strategies on cognitive function in the ICU. Larger studies are needed to validate these preliminary findings.
Prediction of critical illness during out-of-hospital emergency care. JAMA 2010; 304:747–54
To determine the need for appropriate acute care, assessment tools are necessary for triage in general populations at risk for critical illness. Objective, out-of-hospital factors may help determine the patients who are most likely to develop critical illness during hospitalization.
This retrospective study was conducted to develop a predictive tool for critical illness by analyzing a population-based cohort of emergency medical services records and hospital discharge data. Using data from a two-tier system that transports patients to 1 of 16 hospitals, 5 yr of emergency medical services records were assessed.
In the development (n = 87,266) and validation (n = 57,647) phases of the study, 5% of patients developed critical illness during hospitalization. The development phase identified 8 predictors including: aged 65 yr or older, respiratory rate of 35 breaths/min or higher, systolic blood pressure of 90 mmHg or lower, heart rate of 120 beats/min or higher, pulse oximetry less than 80%, Glasgow Coma Scale score lower than 8, and nursing home residency. During validation, the model was shown to have a sensitivity of 0.22, a specificity of 0.98, and a positive likelihood ratio of 9.8.
Triage decisions and patient orientation in the emergency setting may be improved by a better understanding of the preadmission factors that lead to critical illness during hospitalization. In this study, the predictors of critical illness were determined. A score was generated for each nontrauma patient admitted to emergent settings. When externally validated, such a score may improve care quality and regional organization of critical care patients.
Timothy J. Brennan, Ph.D., M.D., Editor
Clinical characteristics of veterans prescribed high doses of opioid medications for chronic non-cancer pain. Pain 2010 151:625–32
There are limited data regarding the characteristics and outcomes of patients receiving high-dose opioids for nonmalignant pain. Because of the numerous risks associated with opioid use—all of which are heightened for patients receiving high doses—it is crucial to further understand this patient population.
Data were extracted from electronic medical records for patients seen at US Department of Veterans Affairs facilities during 1 yr. Patients with chronic (at least 3–6 months' duration) noncancer pain were classified into three groups based on opioid prescriptions for 90 consecutive days: high-dose opioids (180 mg or more morphine equivalent per day), traditional dose opioids (5–179 mg morphine equivalent per day), or no opioids.
The majority (90.7%) of patients were men aged approximately 55 yr. The high-dose opioid group had the highest rate of comorbidities, including congestive heart failure (12.3%), chronic pulmonary disease (36.8%), paraplegia/hemiplegia (5.4%), and psychiatric diseases (up to 66.9%). The high-dose group had an average daily opioid dose 10 times greater than the traditional dose group (324.9 vs. 32.4 mg). Patients in the high-dose group received short-acting opioids, only 40.8% had urine drug screens, and 32% received concurrent benzodiazepines.
Recently, guidelines for the use of opioids in chronic noncancer pain have been developed. These guidelines are intended to optimize treatment and safety in this patient population. In this review of U.S. Department of Veterans Affairs pain clinic patients with chronic nonmalignant pain, variability in the application of these guidelines was demonstrated.
Vertebroplasty versus conservative treatment in acute osteoporotic vertebral compression fractures (Vertos II): An open-label randomised trial. Lancet 2010; 376:1085–92
Two recent randomized studies failed to show a benefit of vertebroplasty—percutaneous injection of bone cement into the fractured vertebral body—compared with sham controls. However, because these studies included a mixed patient population and varying methodologies, clinical interpretation may be affected.
The current open-label randomized trial compared percutaneous vertebroplasty with conservative pain treatment in patients with acute vertebral compression fractures. Inclusion criteria were as follows: aged 50 yr or older, spine radiograph–identified vertebral compression fracture at T5 or lower, and back pain for no more than 6 weeks.
Baseline pain scores, as assessed through a visual analog scale, were similar between study groups. Both groups experienced pain reductions after treatment. However, pain score decreases were significantly greater in the vertebroplasty group at all time points up to 1 yr. Significant pain relief was achieved earlier in patients who underwent vertebroplasty compared with conservative treatment (29.7 vs. 115.6 days). Total mean medical costs were lower at 1 month and 1 yr in the conservative group, with the cost difference being roughly equivalent to the cost of the procedure.
In previous studies, patients with acute and vertebral fractures were randomized with sham control procedures. Results showed minimal benefits of vertebroplasty for pain relief. In this open label study of acute compression fractures, no sham control was utilized. However, it has been argued that patients with acute vertebral fractures are most likely to benefit from vertebroplasty. Randomized and blinded studies with sham control procedures are warranted.
Electrical impedance of acupuncture meridians: The relevance of subcutaneous collagenous bands. PLoS One 2010; 5:e11907
Despite its long history, little is known regarding the mechanism of action for acupuncture. Studies have suggested that reduced electrical impedance and correlating intermuscular connective tissue planes may play a role.
A small study of healthy volunteers (N = 28) was conducted to determine whether electrical impedances are lower at acupuncture meridians than in adjacent control tissue. In addition, investigators assessed whether echogenic collagen is associated with electrical impedance obtained at test sites. Measurements were taken at three segments on the skin surface along the acupuncture meridians including the large intestine (upper arm), liver (thigh), and bladder (lower leg).
Mean impedances of skin at 10 and 100 kHz were 353 and 446ω, respectively, although there was significant within- and between-subject variability. Only the large intestine meridian demonstrated a significant decrease in impedance compared with control sites. Differences in echogenic density were positively associated with differences in electrical impedance.
Acupuncture has been used for thousands of years in pain management. However, its efficacy has been variable and its mechanisms of action are not well understood. This study demonstrates that meridians used for electrical stimulation in acupuncture may represent areas of low impedance/high conductance of electricity. This observation may explain why the specified body regions were chosen as sites for acupuncture. Learning more about acupuncture may improve its utility and efficacy in the future.