CPR DURING IN-HOSPITAL CARDIAC ARREST
Staff aren't performing it well. Should the guidelines be changed?
According to a recently published study, hospital staff members often perform cardiopulmonary resuscitation (CPR) inadequately.
Researchers conducted a small, prospective study based on data pertaining to 67 patients who underwent cardiac arrest while hospitalized. To collect the observational data during the 16-month period, the hospital code team used a specific investigational monitor that was able to defibrillate and to measure CPR parameters, such as the frequency of chest compressions and ventilations and their respective depths and volumes. From those data, the researchers also calculated the "no-flow fraction," the period during which there was no circulation because there were no chest compressions. The American Heart Association CPR guidelines and international CPR guidelines were considered as the main outcome measure of CPR performance standards.
The researchers analyzed the data every 30 seconds for the first five minutes of each cardiac arrest, finding that in 28% of the 30-second intervals the rate of chest compressions was less than 90 per minute, and that in 12.8% it was less than 80 per minute, and that ventilations of more than 20 per minute were delivered in 60.9% of the intervals. The depths of compressions also were not optimal, 37.4% of them being too superficial at less than 38 mm. Lastly, the mean no-flow fraction in the study was determined to be 0.24, compared with the recommended 0.17, the latter representing a 10-second pause in chest compression during each minute of arrest. The authors note also that, although the study was not designed specifically to measure outcome, there was a trend suggesting that patients with lower no-flow fractions had more successful resuscitation outcomes.
The deviation from the CPR guidelines was thought to be attributable to several factors, among them the anxiety induced by a chaotic cardiac arrest scene and the need for hospital staff members to have more frequent CPR review. The researchers offered one possible solution to the low quality of in-house CPR: the use of mechanical devices that deliver compressions automatically according to preset parameters.
How effective are CPR guidelines? An accompanying editorial, written by two physicians, questions the feasibility of the current CPR guidelines and the review process by which they are developed.
The argument is based on research study data that suggest that health care providers do not perform CPR adequately in both in-house and out-of-hospital settings. Specifically, it was found that a significant percentage of patients did not receive the appropriate amount of chest compressions, which are vitally important to postarrest survival.
The authors assert that the CPR guidelines have become too cumbersome. Changes made to them over the years have resulted in recommendation of more breaths than necessary and in too many interruptions in administration of chest compression. They argue that the basic life support-advanced cardiac life support algorithms are too complex, now applying to many circumstances besides cardiac arrest, such as shock resuscitation. They believe that the focus of CPR training and practice should be on first-responder cardiac arrest resuscitation.
Further, according to the authors, the process for revising the CPR guidelines, scheduled again for this year, is flawed. It begins with analysis of research studies that are evaluated according to specified criteria by which, for example, a randomized controlled trial (RCT) holds the greatest validity in terms of results. The problem, however, is that there are few RCTs to serve as references because it is difficult to conduct research in this patient population. Most studies that do so, they claim, are observational and are thus viewed as having less credibility than an RCT. Additionally, current guidelines generally have been "grandfathered in" during the review process, meaning that older practices are not subjected to the same scientific scrutiny that newer recommendations undergo.
The CPR guidelines must be reevaluated to make CPR easier to perform and the pertinent knowledge easier to retain. According to the authors, this could make a difference in the survival rates of cardiac arrest patients. -AT
Abella BS, et al. JAMA 2005;293(3):305-10; Sanders AB, Ewy GA. JAMA 2005; 293(3):363-5.
THE ICD VS. AMIODARONE
A study shows the implantable cardioverter-defibrillator more effective.
A recently published study reveals that a shock-only implantable cardioverter-defibrillator (ICD) reduces overall mortality, and that amiodarone does not, among patients with congestive heart failure (CHF).
Studied were 2,521 patients with New York Heart Association (NYHA) class II or III CHF (70% and 30% of patients, respectively) and a left ventricular ejection fraction (LVEF) of 35% or less (median, 25%); they were assigned randomly in equal number to receive conventional therapy for CHF and placebo, conventional therapy and amiodarone, or conventional therapy and a conservatively programmed, shock-only, single-lead ICD. The median duration of follow-up was 45.5 months. The placebo and amiodarone were administered in a double-blind fashion and the primary end point was death by any cause.
The results revealed 244 deaths in the placebo group (29%), 240 in the amiodarone group (28%), and 182 in the ICD group (22%), indicating a comparable risk of death in the former two groups and a 23% comparatively lesser risk of death in the latter. Moreover, ICD therapy was found to be associated with a 7.2% absolute decrease in mortality after five years. Results did not vary according to whether there were ischemic or nonischemic causes of CHF, but they did vary according to the NYHA CHF class.
When investigating a patient's treatment options, say the researchers, providers might choose an ICD over medication, depending on the NYHA CHF class and LVEF. -PWF
Bardy GH, et al. N Engl J Med 2005; 352(3):225-37.
ANGINA AND NORMAL CORONARY ARTERIES
The risks of myocardial infarction and death have been identified.
In women with angina who have normal angiograms or ones showing nonobstructive coronary disease, the prognosis may not be as benign as previously believed.
According to recently published research among patients with chest pain, a normal angiogram or nonobstructive coronary disease found at angiography is not uncommon and is seen in 10% of women and 6% of men presenting with ST-segment elevation myocardial infarction. The symptoms of such nonobstructive atherosclerotic disease of the coronary arteries often are identical to those of obstructive coronary artery disease.
Nonwhite women constitute the largest group among patients in whom diagnostic findings were consistent with myocardial ischemia or myocardial infarction and nonobstructive atherosclerotic disease of the coronary arteries. Among patients with unstable angina and nonobstructive atherosclerotic coronary artery disease there was a 2% risk of either death or myocardial infarction at 30-day follow-up. Additionally, recent studies have shown that at least 20% of women with normal angiograms or ones showing nonobstructive coronary disease have myocardial ischemia.
It was concluded that patients with chest pain and either normal angiograms or ones showing nonobstructive coronary disease are female predominantly, and that the customary reassurance in lieu of a specific approach to treatment may no longer be appropriate. Symptom-relieving pharmacotherapy (including the use of tricyclic agents and [beta]-blockers) and aggressive antiatherosclerotic therapy with statins, angiotensin-converting enzyme inhibitors, or both, should be undertaken in the presence of risk factors. -PWF
Bugiardini R, Bairey Merz CN. JAMA 2005; 293(4):477-84.
ED USE BY PATIENTS LATER INVOLVED IN HOMICIDES
Identifying those at risk.
Patients at risk of involvement in homicides can be identified according to the reasons for and frequency of ED visits, according to a new study.
Researchers at a large, level I trauma center determined the study population by conducting a retrospective search of six years of police records. They identified homicide victims and perpetrators for whom there were also physician-billing records generated by hospital visits. Health care system data from the three years preceding the homicide were analyzed. To produce controls, the researchers compared the findings to data pertaining to demographically comparable patients who had not been involved in homicides.
The study revealed that use of the ED by the victims and perpetrators of homicides followed certain trends (with the exception of visits associated with drug abuse and those made for firearms injuries, in which there was a greater incidence among perpetrators). Men constituted 80% of the victims and perpetrators in the homicide group; they were more likely to receive treatment at an ED visit, in comparison with the control group (85% and 59%, respectively), and to be hospitalized. In addition, the researchers found that there were numerous ED visits among subjects in the homicide group during the three-year study period, with one or more per subject attributable to injury-most often involving firearms and assault-and substance abuse, specifically of alcohol.
Another significant finding is that as the date of the homicide drew near, the frequency of ED visits among those in the homicide group increased. The authors conclude that the findings could help ED personnel identify those at risk of being involved in homicides. -AT
Crandall CS, et al. Ann Emerg Med 2004; 44(6):646-55.
RETINOPATHY AND CHF
Microvascular retinal disease may signal future heart failure.
In patients with the signs of retinopathy, there is a twofold greater risk of developing congestive heart failure (CHF), compared with those without them, according to a recently published prospective cohort study. This finding supports the hypothesis that microvascular disease plays a role in CHF.
A large sample consisting of 11,612 subjects ranging from 49 to 73 years of age was recruited from a study database. Each participant had retinal images taken and classified according to the signs of microvascular disease, as indicated by retinopathy and focal or retinal arteriolar narrowing and arteriovenous nicking. The researchers followed each one annually for seven years to ascertain whether the participant had been hospitalized; further information was obtained from hospital discharge and death records.
After controlling for known CHF risk factors, such as the presence of hypertension, coronary heart disease, and others, the researchers found that retinopathy had an independent association with "CHF incidents," defined as any hospitalization for or death from CHF that occurred since retinal examination. An additional noteworthy finding is that, in the presence of retinopathy, the risk of developing CHF was three times greater in subjects who had no history of diabetes, hypertension, or coronary heart disease. Lastly, there was a significantly greater association between CHF and retinopathy in the presence of diabetes alone, compared with other isolated risk factors, leading researchers to believe that the microvascular damage diabetes induces might have broad implications in CHF treatment.
The researchers suggest that retinal examination might serve as a beneficial assessment tool in predicting the development of CHF. In particular, they stress that the presence of retinopathy in patients with diabetes might warrant additional cardiac diagnostic tests. -AT
Wong TY, et al. JAMA 2005;293(1):63-9.
EFFECT OF CORTICOSTEROIDS ON PATIENTS WITH SEPSIS
In general, mortality rates unaffected.
A metaanalysis of trials comparing the effects of corticosteroids with placebo or supportive treatment alone (control) on mortality rates among patients with severe sepsis and septic shock revealed that although overall mortality rates weren't affected, they were reduced in a subgroup that received low doses for long periods.
A search of the Cochrane Infectious Diseases Group Trials Register, the Cochrane Central Register of Controlled Trials, MEDLINE, EMBASE, and LILACS (Literatura Latino-Americana e do Caribe em Ciencias da Saude) through August 2003 produced 23 randomized and quasirandomized trials in any language or holding any publication status. Sixteen eligible trials were selected (n = 2,063), in which treatment with corticosteroids could include any preparation as long as it was administered intravenously. The control group, on the other hand, received standard medical treatment (with or without a placebo) such as the administration of antibiotics, inotropics, or vasopressors; renal replacement therapy; fluid replacement; and mechanical ventilation. Researchers categorized treatment duration at full doses of corticosteroids as either long (five days or more) or short (less than five days), and daily doses as either low (300 mg or less of a corticosteroid) or high (more than 300 mg). They were interested primarily in all-cause mortality at 28 days.
The results showed that the all-cause, 28-day mortality rates in 15 trials (n = 2,022) were comparable in the corticosteroid and control groups, 34% and 33%, respectively, and that the relative risk of death in the corticosteroid groups was 92% of that in the control group. The results showed great heterogeneity. When the researchers analyzed only the trials in which low doses of corticosteroids were taken for long periods of time (five trials, n = 465), the relative risks of death at 28 days and at discharge were significantly less (80% and 83%, respectively), and the results were not heterogeneous. This reduction, as well as the absence of heterogeneity, wasn't seen in the analysis of trials in which high doses of corticosteroids were taken for short periods of time (eight trials, n = 1,115). Corticosteroid use was associated with lesser mortality in the ICU (when patients were administered low doses of corticosteroids for long durations) and with increased shock reversal at the seventh and 28th days. Additionally, the incidence of adverse effects, such as gastro-duodenal bleeding, superinfections, and hyperglycemia, did not increase.
Although corticosteroid use, in general, wasn't found to reduce the mortality rate in patients with severe sepsis and septic shock, the administration of low doses of it over a long period of time may be a worthwhile practice, given its safety profile and favorable effects on short-term mortality rates.--JC
Annane D, et al. BMJ 2004;329(7464): 480-8. t