A review of staphylococcal Endocarditis
VG Fowler Jr, Miro JM, Hoen B, et al for the ICE Investigators. Staphylococcus aureus endocarditis: a consequence of medical progress. JAMA. 2005;June 22;293(24):3061-3062.
Agroup of infectious diseases experts from centers throughout the world formed the International Collaboration on Endocarditis (ICE) in 1999 to gain a global understanding of infective endocarditis. Using the Duke Criteria patients with definite infective endocarditis in a prospective manner, 275 variables were reported with these cases to a central database maintained at Duke University. The ICE-Prospective Cohort Study (ICE-PCS) enrolled 1,779 patients with infective endocarditis in 39 centers in 16 countries between June 15, 2000, and December 31, 2003, and has been described in a recent report. (Cabell CH, Abrutyn E. Infect Dis Clin North Am. 2002;16:255-72). Staphylococcus aureus was the most common cause of infective endocarditis in this group of patients (n=558; 31.6%); the authors characterized risk factors and clinical issues associated with these cases in this report.
By univariate analysis, compared with non-Staphylococcus aureus infective endocarditis, patients with infective endocarditis due to Staphylococcus aureus were more likely than patients with infective endocarditis due to other pathogens to be female (P<0.001), hemodialysis dependent (P<0.001), have diabetes mellitus (P=0.009), have other chronic illnesses (P<0.001), have a healthcare association (P<0.001), have higher rates of stroke (P<0.001), have systemic embolization (P<0.001), have persistent bacteremia (P<0.001), or die (P<0.001).
Although healthcare associated Staphylococcus aureus infective endocarditis was the most common form of Staphylococcus aureus infective endocarditis, more than 60% of healthcare-associated patients acquired the infection outside the hospital. This reflects the global trend in healthcare delivery patterns favoring ambulatory treatment (e.g., outpatient medication infusion via long-term IV access, hemodialysis)
Multivariate analysis, clinical features independently associated with Staphylococcus aureus infective endocarditis (versus non-Staphylococcus aureus infective endocarditis) were: IV drug use (OR, 9.3; 95% CI, 6.3-13.7); first clinical presentation less than one month after first symptoms (OR, 5.1; 95% CI, 3.2-8.2); healthcare-associated infection (OR, 2.9; 95% CI, 2.1-3.8), persistent bacteremia (OR, 2.3; 95% CI, 1.5-3.8), presence of a presumed intravascular device source (OR, 1.7; 95% CI, 1.2-2.6), stroke (OR, 1.6; 95% CI, 1.2-2.3), or diabetes mellitus (OR, 1.3; 95% CI, 1.1-1.8).
Patients from the United States with Staphylococcus aureus infective endocarditis were more likely to be hemodialysis-dependent, to be diabetic, to have a hemodialysis fistula or a chronic indwelling central catheter as a presumed source of infection, and to have a non-nosocomial healthcare association. Patients from the United States and Brazil were more likely to have Methicillin-resistant Staphyloccocus aureus (MRSA) than were patients from Europe, the Middle East, Australia, or New Zealand. In-hospital mortality rates were similar across regions, although American patients were significantly more likely to develop persistent bacteremia (25.6%, P<0.001).
Characteristics independently associated with mortality among patients with nonintravenous drug-use-associated Staphylococcus aureus infective endocarditis by multivariate analysis included stroke (OR, 3.67; 95% CI, 1.94-6.94), persistent bacteremia (OR, 3.06; 95% CI, 1.75-5.35), diagnosis in Southern Europe or the Middle East (OR, 3.21; 95% CI, 1.17-10.56).
This study establishes Staphylococcus aureus infective endocarditis as the leading cause of infective endocarditis in many regions of the world and spotlights the global emergence of healthcare contact as a risk factor for Staphylococcus aureus infective endocarditis. In a significant portion of these patients, an IV device was the presumed source of bacteremia; prosthetic cardiac devices (pacemakers, defibrillators, or prosthetic cardiac valves) were present in almost one-quarter of the patients.
MRSA was a significant cause of Staphylococcus aureus infective endocarditis and displayed regional variation, accounting for almost 40% of the infective endocarditis caused by Staphylococcus aureus in some regions. Patients with infective endocarditis caused by MRSA were significantly more likely to have pre-existing chronic conditions and healthcare associated infective endocarditis by both univariate and multivariate analysis. They also were often associated with persistent bacteremia. On the other hand, 20% of patients with MRSA infective endocarditis developed their infection in the absence of identifiable healthcare contact.
Limitations of this report include the fact that this is an observational study of patients from self-selected centers. Each center most likely represents a portion of the local population, making it difficult to generalize findings to the entire population centers from which this report originates. Represented hospitals were typically referral centers that have cardiac surgery programs and may have widely differing populations with varied risk factors. Advantages include the large size of this prospectively evaluated cohort and the ability to analyze regional variations between continents with a contemporary nature of the patient sample (2000-2003).
Infectious Endocarditis in Olmsted County, Minn.
Tleyjeh IM, Steckelberg JM, Murad HS, et al. Temporal trends in infective endocarditis: a population-based study in Olmsted County, Minnesota. JAMA. 2005;293:3022-3028.
Tleyjeh and colleagues at the Mayo Clinic in Rochester, Minn., retrospectively studied 102 cases of infective endocarditis that occurred in 107 Olmsted County residents from 1970-2000. Main outcome measures were incidence of infective endocarditis, proportion of patients with underlying heart disease and causative micro-organisms and clinical characteristics. The records of all Olmsted County residents with infective endocarditis were identified and reviewed in detail. The definite and possible infective endocarditis cases as defined by modified Duke Criteria were used in the analysis.
The age- and gender-adjusted incidence of infective endocarditis ranged from 5.0 to 7.0 cases per 100,000 person-years during the study period and did not change significantly over time. There were 84 (79%) cases of native valve infective endocarditis and 23 (21%) cases of prosthetic valve infective endocarditis. Valves involved: aortic—36 (24%); mitral—49 (46%), aortic and mitral—12 (11%), right-sided or bilateral—8 (7%), or unknown—8 (7%). 16 (15%) had valve surgery within 42 days and the six-month mortality was 26% (n=28).
Infective endocarditis is a disease of the older individual in this population, with a mean age ranging from 54.1 years in 1980-1984 to 67.4 years in 1995-2000 (P=0.24 for trend). There was a male predominance (67%-83%), which did not significantly change over time.
Mitral valve prolapse was the most frequent underlying valvular heart disease. Viridans streptococci were the most common causative organisms (n=47; 44%) followed by Staphylococcus aureus (n=28 cases; 26%).
The overall average crude infective endocarditis incidence of the period 1970-2000 was 4.95 per 100,000 person-years. The age- and gender-adjusted annual incidence was 6.06 per 100,000 (95% CI, 4.89-7.22). There was no time trend for either streptococcus or Staphylococcus aureus infective endocarditis: the annual adjusted incidence of viridans group streptococcal infective endocarditis was 1.7 to 3.5 cases per 100,000 person years while Staphylococcus aureus infective endocarditis had an annual adjusted incidence of 1.0-2.2 cases per 100,000. The incidence rates of viridans group streptococcal and Staphylococcus aureus infective endocarditis have not changed significantly over time in this population.
Potential limitations of this study include possible incomplete case finding or recognition of the retrospective nature of the case reviews. The homogeneity of the patient population studied (primarily elderly white males with a low prevalence of intravenous drug use) limits the ability to generalize the results. Advantages include the fact that this is a population-based study at a medical center with detailed medical records of virtually all residents of a single county. This allows us to view the clinical features and etiologic factors of primarily left-sided infective endocarditis without the referral bias that tends to taint other studies typically published out of large medical centers with larger geographic referral bases.
Computers and Adverse Drug Events
Nebeker JR, Hoffman JM, Weir CR, Bennett CL, Hurdle JF. High rates of adverse drug events in a highly computerized hospital. Arch Intern Med. 2005;165:1111-1116.
Adverse drug events account for a significant number of hospital admissions and the ensuing costs associated with these hospitalizations. Electronic endeavors, such as computerized physician order entry (CPOE), bar code systems, and electronic medical records attempt to reduce the preventable adverse drug events.
Nebeker, et al. attempted to assess the effects of the implementation of CPOE and other computerized medication systems on adverse drug events in a tertiary care Veterans Administration Medical Center. They used an observational study design whereby 937 out of 2,306 newly admitted patients from several hospital wards were randomly chosen and assigned to a pharmacist reviewer during a 20-week period.
They reviewed complete medical records of hospital stays every other day to document adverse drug events. Not only were traditional adverse drug events identified, but harm from overdoses and/or inappropriate dose reductions or discontinuations, as well as intolerable harm from dose titration, were documented as adverse drug events. The harms caused by the drugs were considered only if the drugs were started in the hospital.
Harms were classified based on prior literature and included standards for pharmacological typology, causality assessment, error type, event terminology, drug class, seriousness index, and medication error category indexing. Additional uncommon classifications were also used, including additional resource utilization. Consensus meetings were held weekly to confirm classification of adverse drug events. Of the admissions reviewed, 483 adverse drug events were identified of which 93% were drug reactions while 7% were due to over- or underdosing. Of the drug reactions, 90% were considered dose-dependent while 10% were considered to be idiosyncratic.
Two different indexing scales were used in classifying the harms. Using the LDS Hospital Scale, it was suggested that 91% of the adverse drug events caused moderate harm while 9% caused serious harm. Using the National Coordinating Council for Medication Error Reporting and Prevention indexing, it was suggested that 87% of the adverse drug events fell into category E (requiring treatment) and 4% into category F (requiring prolonged hospitalization). Twenty-seven percent of the total adverse drug events were thought to be due to errors, including execution and planning steps. Sixty-one percent of errors occurred with the ordering mechanism while 25% of the errors occurred in the monitoring process.
This study highlighted rates of adverse drug events five to 19 times higher than baseline. The authors explained this higher-than-expected rate in part by study elements, such as the use of clinical pharmacists as reviewers, iterative case reviews, and accessible electronic data that make adverse drug events more noticeable.
Weaknesses of this study included issues of comparability of CPOEs because there were significant feature differences among commercial software programs. In addition, this was an observational study lacking a control group. The authors felt that their study did not support the idea that the computerized patient record of the study institution had caused adverse drug events. Rather, the study supported the idea that the system increased the visibility of adverse drug events compared with a paper system. In addition, the authors recommended that the choice of CPOEs be carefully considered, with a focus on decision support features when integrated into a healthcare organization.
The Questionable Benefit of Medical Emergency Teams
Hillman K, Chen J, Cretikos M, et al. Introduction of the medical emergency team (MET) system: a cluster-randomised controlled trial. MERIT study investigators. Lancet. 2005;365:2091-2097.
Previous studies have reported that the MET system reduces the incidence of cardiac arrests, deaths, and unplanned ICU admissions. A MET is a preplanned group of healthcare practitioners who respond to acute patient deteriorations in hospitalized patients.
METs are usually identical to hospital code teams, with the exception that they respond prior to a patient’s developing cardiac arrest. This early response has been shown to significantly decrease unexpected hospital mortality in hospitals in the United States, Australia, and Great Britain. Even though the system has been reported since 1995, few hospitals have knowledge of or experience with METs.
Unexpected hospital deaths and cardiac arrests are often preceded by clinical warning signs. In addition, unplanned ICU admissions may be foreshadowed by abnormalities in the patient’s vital signs that may progress if appropriate interventions are not undertaken. METs assess patients with abnormal physical findings or when there is a concern about the patient’s condition. These patients have findings that may precede a serious event or cardiac arrest, but otherwise don’t meet existing criteria to call a code.
The theory is that if a MET responds to see a patient who is becoming unstable (see “Table 1: MET Calling Criteria,” at left), early interventions may reduce the likelihood of arrest. Published studies have shown a reduction in cardiac arrests and ICU length of stay in virtually all systems in which MET has been introduced (although most studies are hampered by the use of historical controls).
The MERIT study randomized 23 hospitals in Australia to continue functioning as usual (n=11) or to introduce a MET system (n=12). The study sites included a wide range of tertiary, metropolitan, and non-metropolitan hospitals in different states across Australia. The primary outcome was the composite of cardiac arrest, unexpected death, or unplanned ICU admission during the six-month study period after MET activation, using intention to treat analysis.
A four-month educational period was undertaken with the MET centers prior to initiation of the trial. Control hospitals did not receive any education about the MET concept. This was followed by a six-month trial period. Cardiac arrest teams were maintained at all hospitals. The MET consisted of at least one doctor and a nurse from the ED or ICU.
The eligible patients included those residing on a medical ward (including critical care units); the ICUs, OR, postoperative recovery areas, and ED areas were not regarded as general wards.
The primary outcome for the study was the composite outcome of the incidence (events divided by number of eligible patients admitted to the hospital and residing on a medical ward during the study period) of:
- Cardiac arrests without a pre-existing “not-for-resuscitation” (NFR) order;
- Unplanned ICU admissions; and
- Unexpected deaths (those without a pre-existing NFR order).
The results of the study:
- During the study period, the overall rate of calls for the cardiac arrest team or MET was significantly higher in intervention hospitals than in control hospitals. Calls not associated with events were more common in MET hospitals than in controls. Half of the total calls were not associated with a cardiac arrest or unexpected death, whereas in MET hospitals more than 80% of calls were not associated with a cardiac arrest or death (P<0.0001).
- In patients with documented MET calling criteria in association with cardiac arrest or unexpected death, the call rate was similar in MET and control hospitals.
- There were no significant differences between the MET and control hospitals for any outcome.
- The response to changes in vital signs was not adequate—even in MET centers.
These findings are surprising in view of previously reported findings using the MET system. Potential reasons for lack of difference between MET centers and controls include:
- Number of study sites or the duration of the study may not have been adequate for implementation or education;
- Hospitals may already be efficient in detecting and managing unstable patients;
- Patient selection criteria may have been overly restricted. For example, other studies have used 30 respirations per minute for tachypnea as a calling criterion compared with 36 breaths per minute used in this trial;
- Knowledge of the study may have leaked to control hospitals;
- Cardiac arrest teams function as METs at times: Nearly half of the calls to cardiac arrest teams in control hospitals were made without a cardiac arrest or unexpected death; and
- The selected outcomes may not be sensitive enough.
Even though this large, multicenter controlled trial was unable to show a significant benefit of METs, we should not be discouraged from performing further controlled trials in different settings. The use of METs is clearly an exciting and evolving area of medicine.
Barriers to Patient Safety
Amalberti R, Auroy Y, Berwick D, Barach P. Five system barriers to achieving ultrasafe health care. Ann Intern Med. 2005;142:756-764
Patient safety in our healthcare system is a growing concern. One area of dialogue concerning preventable healthcare-associated harms involves the comparability of the healthcare industry with non-medical industries, such as aviation and nuclear power, that have adapted successful strategies shown to provide ultrasafe environments. Amalberti, et al. discuss risk assessment in a variety of industries and explain the need for a benchmarking approach in order to optimize or achieve safety in the healthcare field.
The authors identify five systemic barriers from literature that are fundamentally connected to the ability of the healthcare field to achieve an extremely safe environment.
Barrier 1—acceptance of limitations on maximum performance: The first barrier is the type of expected performance in the field. This is illustrated by the tradeoffs associated with ultrasafety versus productivity. The amount of risk involved was directly related to the limits placed on maximum performance. The first barrier is the acceptance that every system has limits. When a producer exceeds their limit, then safety suffers. An example used is that of blood donation: The limits of collection speed are weighed against the needed screening process.
Barrier 2—abandonment of professional autonomy: The second barrier concerns the concept of professional autonomy. While more teamwork and regulations reduce individual autonomy, this appears to improve safety in the healthcare environment. The bottom line is the importance of teamwork. The example used is that of traffic on a highway: Autonomous units work together to function safely.
Barrier 3—transition from the mindset of craftsman to that of an equivalent actor: The third barrier to achieving high levels of safety includes an equivalent actor mindset. This entails establishing a reliable standard of excellent care in lieu of focusing on individuality, similar to the notion that passengers on an airline usually do not know their pilots, but have established confidence in the airline itself.
Barrier 4—the need for system-level arbitration to optimize safety strategies: The fourth barrier identified is a need for system-level arbitration to optimize safety strategies. This need results from the pressure for justice (usually through litigation) once an accident occurs. Top-down arbitration of safety will be less successful than system level design.
Barrier 5—the need to simplify professional rules and regulations: The final barrier results from the many of layers of guidelines as they serve to create an environment of excellence. This barrier necessitates the removal of these layers to simplify the environment. Existing guidelines should be distilled down to those shown to promote quality and safety. Byzantine rules can obscure the goal of safety and glorify rules, for rules sake.
Certain structural limitations within the field, such as worker shortages in the face of increasing public demands and the reliance of the field on trainees such as students, interns, and residents, create other hurdles. The authors conclude by suggesting a two-tiered system of healthcare whereby ultrasafety could be more easily accomplished in areas of medicine considered more stable (first tier), and a second tier of care that would include the more unstable conditions, and thus inherently, represent the higher risk situations where errors are more likely to occur.
Another provocative point of this article is the need to move toward educating and training teams—not individuals.
The Importance of Implementing COPD Guidelines
Harvey PA, Murphy MC, Dornom E, et al. Implementing evidence-based guidelines: inpatient management of chronic obstructive pulmonary disease. Intern Med J. 2005;35:151-155.
COPD is a common diagnosis that sometimes requires hospitalization. Evidence-based guidelines for disease management, including that of hospitalized patients, exist, but there is a paucity of knowledge about the actual quality of care delivered in the hospital as it aligns with published guidelines. This study by Harvey, et al. explores the quality of care delivered in the hospital for patients with COPD, while at the same time investigating an intervention for the medical staff in an effort to improve adherence to evidenced-based guidelines of the disease.
Using ICD-10 codes for a COPD diagnosis, the study incorporated a retrospective chart review of 49 hospital admissions prior to the intervention and 35 admissions after the intervention in a hospital in Melbourne, Australia. Data were collected pertaining to the hospital management of COPD as it compared with what the authors considered to be Level A—or the highest level of evidence summarized from several professional organizations. The intervention delivered to the medical staff included a summarized presentation of the results from the initial audit of the 49 charts, as well as an educational package that was given to them following the presentation.
Except for inappropriate use of intravenous aminophylline, of which there was a 100% concordance to Level A guidelines, the initiation of systemic steroids (intravenous and/or oral) had the highest concordance rate of 80% and 83%, pre- and postintervention respectively. Appropriate steroid duration (seven to 14 days) had the lowest concordance rates at 10% and 29%, pre- and postintervention respectively.
In addition, preintervention concordance (10%) involving steroid duration was the only rate considered significantly different in the postintervention group (29%). While concordance rates were high for the use of any type of nebulized bronchodilator (96% preintervention and 80% postintervention), the Level A guidelines the authors used suggested that beta-agonist bronchodilators should be used alone prior to the initiation of ipratropium bromide. The concordance rates for this guideline were 27% preintervention and 34% postintervention.
Largely, the authors felt their intervention failed to improve concordance rates to the Level A guidelines investigated and also that their findings of variable and lower concordance rates across the board corroborated other similar studies. The major weaknesses of this study included the small sample size and the nonrandomness of the sampling.
In addition, the authors report that the particular hospital studied included junior doctors who rotated on and off service, which likely prevented the effects of the intervention from being measured on a provider level. In spite of the weaknesses in the study, the article brings to light the need for a more effective translation of evidence-based guidelines to actual practice, especially in the practice of COPD management in the hospital. Further methods of guideline implementation in the clinic setting must be elucidated to improve the care patients with COPD receive in the hospital.
Not all Troponin Elevations Are Myocardial Infarctions
Jeremais A, Gibson CM. Narrative review: alternative causes for elevated cardiac troponin levels when acute coronary syndromes are excluded. Ann Intern Med. 2005;142:786-791.
Troponins are regulatory proteins that control the calcium-mediated interaction of actin and myosin during muscle contraction. All muscle tissue contains troponins, but cardiac troponin T and I have amino acid sequences that are different from skeletal and smooth muscle troponins, allowing them to be detectable by monoclonal antibody-based assays.
In the event of reversible or irreversible cell damage—or possibly even from transiently increased cell membrane permeability—cardiac troponins are released from myocytes into circulation. This characteristic provides a sensitive test for detecting myocardial injury and damage; however, this test is not specific for acute coronary syndromes. And any disorder that causes myocyte damage may cause an elevated troponin.
The 2002 American College of Cardiology/American Heart Association practice guidelines for unstable angina and non-ST-segment elevation myocardial infarction acknowledge that the myocardial necrosis signified by troponin elevation may not necessarily be caused by atherosclerotic coronary artery disease. Such nonthrombotic troponin elevation can be caused by four basic mechanisms, as discussed by Dr. Jeremias and Dr. Gibson.
- Demand ischemia refers to a mismatch between myocardial oxygen demand and supply in the absence of flow-limiting epicardial stenosis. Conditions such as sepsis or septic shock and the systemic inflammatory response syndrome, hypotension or hypovolemia, tachyarrhythmias, and left ventricular hypertrophy can all cause release of cardiac troponin.
- Myocardial ischemia in the absence of fixed obstructive coronary disease can be caused by coronary vasospasm, acute stroke or intracranial hemorrhage, and ingestion of sympathomimetics.
- Direct myocardial damage can be seen in cardiac contusion, direct current cardioversion, cardiac infiltrative disorders such as amyloidosis, certain chemotherapy agents, myocarditis, pericarditis, and cardiac transplantation.
- Myocardial strain occurs when volume and pressure overload of the left and/or right ventricle cause excessive wall tension. Congestive heat failure, acute pulmonary embolism, and chronic pulmonary hypertension can lead to myocardial strain and troponin elevation.
Another condition that can lead to persistently elevated cardiac troponins is end-stage renal disease. This elevation may be due to small areas of clinically silent myocardial necrosis, an increased left ventricular mass, or possibly from impaired renal troponin excretion. Although troponins are believed to be cleared by the reticuloendothelial system, recent evidence shows that troponin T is fragmented into molecules that are small enough to be renally excreted.
In summary, elevated troponin can be found in many clinical settings and is associated with impaired short- and long-term survival. TH