Rules, Guidelines, Protocols

FigureDo you know how many zillions of dollars are spent on clinical research each year that result in no clinical benefit to our patients? It is simply unfathomable that governments, foundations, and for-profit organizations are willing to spend so much money on so many studies that are so poorly designed that their conclusions are meaningless. These are studies so underpowered that they are unable to provide definitive conclusions. These are studies that involve selected patient populations that do not allow extrapolation of results to the broader community at large. Zillions. And what about the good studies? Do they result in changes in physician behavior? Do patients or the health system benefit? Difficult to say, but it certainly is uncommon. Take the Ottawa ankle rules, for example. There are at least 15 studies of these rules in the clinical literature. All show that the rules are sound, reproducible, accurate, and easy to use. They are as good a set of rules as is likely ever to be developed. How much money was spent on the studies? It had to be plenty, perhaps hundreds of thousands of dollars if not more. Is anybody using the rules to reduce unnecessary x-rays? Has the promise of the 20 percent or better reduction in x-rays been achieved? After all, the whole point of constructing good rules is to get them in place so they can improve patient care. If no one uses the rules, all the clinical studies behind the rules are a waste of time and money. You didn't get the ball across the goal line. You get no credit. Nice try, but sorry. It's as if the rules were never developed. Despite repeated validation of the Ottawa Ankle Rules, use of x-rays in blunt ankle and midfoot trauma has not decreased The following paper looks at whether the Ottawa ankle rules are being used in their home country of Canada, where there is much less fear of litigation than in the U.S. and where the obsession with patient satisfaction is likely to be tempered with some constraints regarding access to care. What did the study show? There was no reduction in the use of ankle x-rays, even when every effort seemed to be made to familiarize clinicians with the rule. Uptake of Validated Clinical Practice Guidelines: Experience with Implementing the Ottawa Ankle Rules, Holroyd BR, et al, Am J Emerg Med, 2004;22(3):149 BACKGROUND: The Ottawa Ankle Rule (OAR) is a widely accepted and validated clinical guideline that has been consistently reported to reduce unnecessary use of ankle and midfoot x-rays for blunt trauma by 20 percent to 30 percent in clinical trials. METHODS: This study from the Universities of Alberta and Calgary in Canada examined the impact of an intensive OAR implementation strategy on use of ankle/midfoot x-rays in appropriate adults. X-ray utilization in four hospitals was examined for a baseline period during 1995 and during a second period in 1996 after province-wide dissemination of OAR information and pocket cards to all physicians. During a period in 1996–1997, an uptake intervention was implemented in two of the four hospitals consisting of educational sessions by local opinion leaders to promote use of the OAR. During a period in 1997, emergency physicians in these same two hospitals received additional personalized education and feedback on x-ray utilization. RESULTS: During the entire study period, clinically significant fractures were diagnosed in 23 percent of the 5,920 patients having x-rays. Rates of x-ray utilization were 92 percent in the two intervention hospitals and 92 percent in the two control hospitals during the pre-OAR period, 92 percent and 92 percent respectively, after province-wide dissemination of OAR information, 94 percent vs. 92 percent, respectively, with directed education in the intervention hospitals, and 93 percent vs. 90 percent, respectively with the addition of personalized feedback. CONCLUSIONS: In this study, an intensive OAR-based intervention did not significantly reduce the use of x-rays in adults with blunt ankle/midfoot trauma. Ottawa Knee Rules What about the Ottawa knee rules? Have they been more successful in reducing utilization of knee x-rays? Not according to the next study. X-ray Requesting Patterns Before and After Introduction of the Ottawa Knee Rules in a UK Emergency Department, Atkinson P, et al, Eur J Emerg Med, 2004;11(4):204 BACKGROUND: The Ottawa Knee Rules (OKR) have been shown in North American studies to predict reliably the presence or absence of bony injury in adults with acute blunt knee trauma, and to have the potential to reduce substantially unnecessary x-rays. The Ottawa Knee Rules did not reduce x-rays in patients with blunt knee trauma METHODS: This prospective study evaluated the effect of introduction of the OKR in a British ED. Findings were evaluated in 58 adults with isolated blunt knee trauma who presented during a one-month period before introduction of the OKR, and in 72 similar patients presenting after its introduction. Patients were generally managed by junior physicians, who recorded the presence or absence of specific clinical criteria (including OKR criteria) during both phases. RESULTS: A bony injury was diagnosed (on initial x-rays or at follow-up) in five of 58 Phase 1 patients (9%) and in seven of 72 Phase 2 patients (10%). Knee x-rays were ordered based on clinical judgment in 59 percent of the Phase 1 patients, including four of the five with bony injury (sensitivity and specificity of clinical judgment, 80 percent and 43 percent In Phase 2, knee x-rays were ordered in 56 percent of the patients, including all seven with bony injury (sensitivity and specificity 100% and 54%, but with very wide confidence intervals for both). Among patients not meeting criteria for x-rays according to the OKR, x-rays were ordered in 33 percent of Phase 1 patients and 11 percent of Phase 2 patients. CONCLUSIONS: The Ottawa Knee Rules may be an effective tool for selecting adult patients with blunt knee trauma who would benefit from x-rays, but hardly reduced x-ray utilization in this setting. Lispro for DKA It is clearly not adequate to develop good rules. This is only the first step in a complex process. Now the rules need to be embraced. How can this be accomplished? Without the answer to this question, we should stop generating treatment guidelines. If we cannot demonstrate that it is reasonable to expect that these rules will be adopted, we are wasting huge amounts of time and money developing rules and guidelines. What about the NEXUS cervical spine guidelines? Have they cut down on doctors ordering neck x-rays? No one knows. And is it even conceivable that the forthcoming NEXUS head CT rules will result in fewer CTs? From a brutally pragmatic point of view, unless there is some mechanism to induce physicians to use rules and guidelines, it can be expected that the entire guideline-creating endeavor may be an extraordinarily low yield and therefore extremely costly exercise. Here's a fascinating paper. It compared the use of subcutaneous lispro insulin with IV insulin infusions in 40 patients with diabetic ketoacidosis. Although the patients were “stable,” they definitely had DKA: sugars in the mid-600s, bicarb of 9-11, and pH levels of around 7.18. Of the patients getting lispro, half were treated on wards and half on monitored wards, while all of the patients treated with IV insulin infusions were treated by hospital protocol (because they were getting an IV insulin infusion) in the ICU. In stable adults with uncomplicated DKA, subcutaneous lispro was as safe and effective as infusion of regular insulin but did not require ICU admission All outcome measures were comparable. And, as expected, charges were much lower when care in the ICU was avoided. Will we be willing to make the switch? I won't hold my breath, yet with smaller and smaller insulin amounts being used to treat DKA you would think we could transition to this technique sooner rather than later. I remember the days when 50 and 100 units of insulin were given at a time to treat DKA, and now patients barely lick the stopper of the bottle for contemporary dosing. Efficacy of Subcutaneous Insulin Lispro Versus Continuous Intravenous Regular Insulin for the Treatment of Patients with Diabetic Ketoacidosis, Umpierrez GE, et al, Am J Med, 2004;117:291 BACKGROUND: Recently introduced analogues of human insulin such as insulin lispro (Humalog) and aspart (Novolog) have a rapid onset of action when given subcutaneously (10 to 20 minutes), and might be suitable for the treatment of DKA outside the ICU setting. METHODS: This prospective study from the University of Tennessee and Atlanta (GA) Medical Center compared outcomes in 40 adults with uncomplicated DKA who were randomized to subcutaneous insulin lispro (0.3 u/kg, followed by 0.1 u/kg per hour until the blood sugar decreased to 250 mg/dL, and then 0.05 u/kg per hour until resolution of DKA) or IV infusion of regular insulin (a 0.1 u/kg bolus followed by 0.1 u/kg/hour until blood sugar decreased to 250 mg/dl, and then 0.05 u/kg/hour until resolution of DKA). By hospital policy, the IV insulin group was managed in the ICU while the alternate group was managed in a step-down unit or general medicine floor. RESULTS: There were no significant differences between the groups in baseline characteristics or in the rate of decline of the plasma glucose level, correction of acid-base parameters, duration of insulin treatment or hospital stay, or amount of insulin given until DKA was resolved. One patient in each group developed mild hypoglycemia. The mean charges for DKA treatment were significantly reduced in the insulin lispro group ($8,801 vs. $14,429). CONCLUSIONS: In stable adults with uncomplicated DKA, administration of subcutaneous insulin lispro was as safe and effective as infusion of regular insulin, but did not require ICU admission, and was associated with a significant reduction in the costs of care. Steroids for Septic Shock Steroids are controversial in a variety of settings, and septic shock is one of those. In the past, patients have been given large doses (30 mg/kg methylprednisolone, in essence a fluid challenge), and the results were negative. What about smaller doses for documented adrenal insufficiency? That seems to be the answer. Two papers currently in our database have researched this question, and both have concluded that physiologic doses of steroids are helpful when cortisol levels are low. In the ED, I would draw a cortisol level in septic patients and empirically give 100-300 mg of hydrocortisone because the cortisol level is not likely to be available on an urgent basis. Subsequent dosing (if any) can be determined when the cortisol level is reported. Corticosteroids for Severe Sepsis and Septic Shock: A Systematic Review and Meta-Analysis, Annane D, et al, Brit Med J, 2004;329:480 BACKGROUND: Severe sepsis has a 30 percent mortality rate (50% to 60% for patients in septic shock). Despite a theoretical benefit of corticosteroids, early studies that generally utilized short courses of high-dose steroids failed to show a benefit. More recent trials of longer courses of low-dose steroids (300 mg or less of hydrocortisone or equivalent daily) designed to treat adrenal insufficiency have reported more favorable results. METHODS: These French authors analyzed the results of 16 randomized, controlled trials (including 2,063 pediatric and adult patients) of steroids in severe sepsis and septic shock published between 1955 and 2003. Underlying infections were common in neonates febrile at home but afebrile upon ED arrival RESULTS: The relative risk (RR) of mortality at 28 days with steroid therapy was 0.97 (p=NS) in studies of short-course, high-dose steroids, but 0.80 (p=0.01) in studies of longer-course, low-dose steroids. Corresponding RRs for in-hospital mortality were 0.89 (p=NS) vs. 0.83 (p=0.02), respectively. The rate of reversal of shock by seven days was 56 percent in treated patients vs. 45 percent in controls in the six studies that evaluated this parameter, but 50 percent vs. 31 percent in the four trials that studied the effects of longer courses of low-dose steroids (RR 1.6, p<0.0001). There was no evidence that steroid therapy was associated with a significantly increased risk of GI bleeding (RR 1.16), superinfection (RR 0.93), or hyperglycemia (RR 1.22). One trial reported a significant increase in serum sodium in 30 percent of patients treated with steroids (6/20) vs. five percent of controls (1/20). CONCLUSIONS: Although the authors suggest that this review does not permit formulation of definitive recommendations, they feel that the results support a five- to 11-day course of hydrocortisone (200-300 mg daily) for patients with septic shock and evidence of relative or absolute adrenal insufficiency on adrenocorticotropin hormone testing. Septic Evaluations for Neonates A parent presents to the ED with a neonate who they say had a fever at home. The child is afebrile in the ED. Does a septic evaluation still need to be done? Apparently, the answer is yes, according to the following study by Lance Brown and colleagues at Loma Linda University in Southern California. Of 27 such neonates, 10 ultimately were diagnosed with infections (one with meningitis). Investigation of Afebrile Neonates with a History of Fever, Brown L, et al, Can J Emerg Med, 2004;6(5):343 BACKGROUND: It is commonly held that neonates under one month of age who are febrile when brought to the emergency department warrant testing for bacterial infection, empiric antibiotics, and hospital admission, but the optimal approach to those with a history of fever at home who are afebrile upon presentation is uncertain. One study reported a 0 percent rate of serious bacterial infection in such children, but the upper limit of the 95% confidence interval was 5.5 percent. METHODS: This study from Children's Hospital in Loma Linda, CA, represents a retrospective review of the records of 27 neonates aged 28 days or younger with recorded fever (at least 38°C) at home who were afebrile upon ED arrival, but who underwent lumbar punctures in the ED (considered to be a marker of performance of a sepsis work-up). RESULTS: Recorded temperatures at home ranged between 38.2°C and 39.4°C. Infections were ultimately diagnosed in 10 of the neonates (37%), including urinary tract infections in three, pneumonia in two, viral meningitis in two, and one case each of bacterial meningitis, bronchiolitis, and enteritis. Acetaminophen was given at home to four of the patients, three of whom were ultimately diagnosed with infections. CONCLUSIONS: Underlying infections were common in these neonates who were reportedly febrile at home but afebrile upon ED arrival, but who were felt to warrant a sepsis work-up.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.