Lit Bits: November 2, 2005

From the recent medical literature...

1. The Case of the Febrile Injection Drug User.

IV drug users who report having a fever with no localizing findings pose a particular diagnostic problem. Infective endocarditis will be diagnosed in what percent of IV drug users hospitalized with fever?

The Answer is:

Infective endocarditis will be diagnosed in up to 20% of injection-drug users hospitalized with fever (NEJM, November 3, 2005, pg. 1949).

2. Elderly with acute coronary syndrome less likely to get recommended care

October 26, 2005

NEW YORK (Reuters Health) - Elderly patients with non-ST-segment elevation acute coronary syndromes (NSTE ACS) are less likely to receive recommended care, a new observational multicenter study shows.

However, Dr. Karen P. Alexander of the Duke Clinical Research Institute in Durham, North Carolina and colleagues report, patients over 75 who did receive treatment that followed American College of Cardiology/American Heart Association (ACC/AHA) guidelines had the same reduced risk of death as their younger counterparts.

A number of past studies have found frequent underuse of cardiac drugs in elderly NSTE ACS patients, Dr. Alexander and her team report in the October 18th issue of the Journal of the American College of Cardiology.

To determine if the new ACC/AHA guidelines may have changed this pattern, the researchers compared care in 56,963 patients with NSTE ACS treated at 443 US hospitals. The patients were stratified into four groups: younger than 65, 65 to 74, 75 to 84, and 85 or older. Thirty-five percent of the patients were 75 or older, while 11% were 85 or older.

The new guidelines, Dr. Alexander and her team note, do not offer different recommendations based on age, "apart from encouraging attention to comorbidities, preferences, and appropriate dosing of medications in elderly patients."

Even after controlling for contraindications and comorbidities, the researchers found, the use of several therapies remained less frequent in older patients. Elderly patients were less likely to receive antiplatelet and antithrombin therapy in the first 24 hours of care, less likely to receive invasive care, and less likely to be discharged on lipid-lowering agents and clopidogrel.

Specifically, while 92% of patients older than 85 had positive cardiac markers, only 29.9% were given clopidogrel and 12.8% received platelet glycoprotein IIb/IIIa inhibitors. After age 65, less than 50% received invasive care, and that number fell to 11.2% after age 85.

"A formal test for interaction between treatment and age was significant because of the larger difference in mortality between younger and older patients given fewer therapies compared with those given all therapies," the researchers write. "This indicates a similar or greater impact of treatment on outcomes in elderly patients."

Dr. Alexander and her team conclude by noting that clinicians may be "understandably cautious" about using newer treatments and more invasive care in their older patients at high risk.

In an editorial accompanying the study, Drs. H. Vernon Anderson of the University of Texas Health Science Center in Houston and Richard G. Bach of Washington University Medical Center in St. Louis argue that physicians should begin to look at elderly ACS patients in terms of opportunities rather than risks.

"The opportunity score would give an estimate of the benefit to be gained by a treatment or group of treatments," they explain. "Patients with higher baseline risks from the underlying disease, such as the elderly, would have higher opportunity scores for benefit, even allowing for some of the greater risks from the treatment."

J Am Coll Cardiol 2005;46:1479-1489

3. New coma measurement system validated

October 26, 2005. NEW YORK (Reuters Health) - A group of neurologists from the Mayo Clinic College of Medicine in Rochester, Minnesota has developed and validated a new coma scale, which they say is a reliable and easy-to-use tool that has the potential to replace the Glasgow Coma Scale (GCS).

The so-called FOUR (Full Outline of UnResponsive) Score consists of four components -- eye, motor, brainstem, and respiration -- to determine coma depth. A score of four represents normal functioning in each category, while a score of zero indicates nonfunctioning.

Dr. Eelco F. M. Wijdicks, a Mayo Clinic neurologist and inventor of the FOUR score, said the widely adopted GCS has "many drawbacks" and misses key elements of a neurological exam of comatose patients. The inability of the GCS to test brainstem reflexes is an "important shortcoming," he added.

Dr. Wijdicks and colleagues prospectively studied the FOUR score in 120 ICU patients with diverse acute neurological conditions (not just trauma) and compared it with the GCS.

They report in the October issue of the Annals of Neurology that the agreement among raters -- neuroscience nurses, neurology residents, and neurointensivists -- was "good to excellent." The interrater reliability of the FOUR score and the GCS was of equal magnitude, according to the researchers.

In the experience of Dr. Wijdicks and colleagues, "the FOUR score provides greater neurological detail than the GCS, recognizes a locked-in syndrome, and is superior to the GCS due to the availability of brainstem reflexes, breathing patterns, and the ability to recognize different stages of herniation."

The team also reports that in this study the likelihood of dying in the hospital was higher for the lowest total FOUR score when compared with the lowest total GCS score.

"With the FOUR Score, in contrast to the GCS, we can in a far more detailed way provide the family information on the patient's status and what the outcome will be," Dr. Wijdicks said.

Ann Neurol 2005;58:585-593.

4. Salt-water Cellulitis.

A patient presents to the ED with a cellulitis resulting from exposure of a laceration to salt water. What organism is of concern? What is the treatment?

Vibrio vulnificus should be considered. Standard antimicrobial therapy is with doxycycline (NEJM, Vol. 350, pg. 908).

5. When should we thrombolyse patients with pulmonary embolism?

A systematic review of the literature . T Harris and S Meek. Emergency Medicine Journal 2005;22: 766-771

ABSTRACT
The early mortality in pulmonary embolism (PE) is largely predicted by the associated cardiovascular response, with progressive right ventricular failure, hypotension, shock, and circulatory arrest being associated with increasing mortality. Thrombolysis may improve the prognosis of PE associated with these varying degrees of circulatory collapse, but has no place in the treatment of small emboli with no cardiovascular compromise, as it carries a significant risk of haemorrhage. This review sets out to guide the emergency physician in deciding which patients with PE may benefit from thrombolysis.

6. Cricoid pressure in emergency rapid sequence induction

John Butler, Consultant and Ayan Sen, Clinical Fellow. Emergency Medicine Journal 2005;22:815-816

ABSTRACT
A short cut review was carried out to establish cricoid pressure reduced aspiration during rapid sequence induction (RSI) of anaesthesia. A total of 241 papers were identified using the reported search, of which three represented the best evidence to answer the clinical question. The author, date and country of publication, patient group studied, study type, relevant outcomes, results, and study weaknesses of these best papers are tabulated. There is little evidence to support the widely held belief that the application of cricoid pressure reduces the incidence of aspiration during a rapid sequence intubation.

7. U.S. panel urges pertussis vaccine for adults

October 28, 2005. By Paul Simao

ATLANTA (Reuters) - U.S. adults should be vaccinated against pertussis to stop its spread among infants, who are especially vulnerable to the infection, a federal advisory panel recommended on Wednesday.

The Centers for Disease Control and Prevention's advisory committee on immunization unanimously approved the recommendation amid signs that pertussis is becoming more common, especially among teenagers.

Nearly 19,000 cases were reported across the United States in 2004, a sharp increase from previous years.

"This vote will help curtail that trend and help us stop the spread of pertussis," said Dr. Bill Schaffner, an infectious diseases expert and a member of the CDC advisory committee, which met in Atlanta.

Health experts advise that infants receive the DTP vaccine at 2-month intervals after birth, followed by a fourth dose about 1 year later and a booster between the ages of 4 and 6 years.

Immunity, however, only lasts between 5 and 10 years.

Prompted by rising numbers of cases among middle school and high school students, the CDC advisory committee recently recommended that adolescents between the ages of 11 and 18 receive the Tdap vaccine, which protects against tetanus, diphtheria and pertussis, in place of the previously recommended diphtheria tetanus booster.

It extended that recommendation on Wednesday to include adults to protect vulnerable newborns from exposure to an infected parent or other adult.

The recommendation, if followed, could also help reduce the number of adults with chronic cough caused by pertussis, Schaffner said. The Tdap vaccine, however, is not licensed for use in seniors.

8. Intensive statin therapy should begin soon after ACS, study shows

October 28, 2005. NEW YORK (Reuters Health) - Prompt, intensive statin therapy in patients with acute coronary syndromes (ACS) leads to a greater reduction in clinical events at 30 days than does standard statin therapy, researchers report in the October 18th issue of the Journal of the American College of Cardiology.

The rapidity of the response, investigator Dr. Christopher P. Cannon told Reuters Health, "surprised all of us, but the implication is that patients hospitalized for ACS need to leave the hospital on high-dose statin -- since there is such early benefit."

As he commented, "It is a big change."

Dr. Cannon, with the TIMI Study Group at Harvard Medical School, Boston, and colleagues examined data from the Pravastatin or Atorvastatin Evaluation and Infection Therapy -- Thrombolysis in Myocardial Infarction 22 trial, known as PROVE IT - TIMI 22. In that study, more than 4,100 patients with ACS were randomized to standard therapy with pravastatin 40 mg or intensive therapy with atorvastatin 80 mg.

The original study results showed that intensive therapy improved outcomes at 2 years. In the current analysis, the researchers "sought to evaluate the timing of this benefit by examining the very early clinical benefit as well as the late effects of intensive statin therapy."

At 30 days, the composite endpoint of death, MI or rehospitalization for recurrent ACS was reached by 3% of the intensive therapy patients and by 4.2% of those on standard therapy, for a hazard ratio of 0.72 (p = 0.046).

Among stable patients, the corresponding endpoint rates were 9.6% and 13.1% (HR = 0.72; p = 0.003).

In light of these findings, continued Dr. Cannon, "we start immediately with the highest dose of a strong statin, and then monitor for safety, but also then afford the patient the very rapid clinical benefit in avoiding a second heart attack."

He added, "We used to take a slower approach, check the fasting blood tests, and slowly up-titrate the statin dose. Now we do the reverse."

In summary, he and his colleagues conclude in their paper, "Treatment of patients with ACS should begin in-hospital with high-dose intensive statin therapy to achieve these early clinical benefits and should be continued long-term."

J Am Coll Cardiol 2005;46:1405-1410.

9. Esophageal Intubations.

Amer J Emerg Med 2005;23: 754-758 Aaron E. Bair, MDa, Erik G. Laurin, MDa, Brandi J. Schmitt, MSb

Abstract
Introduction
Early detection of an inadvertent esophageal intubation can be particularly challenging in cases when the current standard of care, carbon dioxide detection, is unreliable. We sought to determine the sensitivity and specificity of an inexpensive and portable device, the gum elastic bougie (Eschmann Tracheal Tube Introducer, SIMS Portex, Inc, Keene, NH), as an endotracheal tube placement confirmation device.

Methods
We conducted a prospective blinded trial in 20 human cadavers. Each cadaver was randomized to a mixed series of 5 esophageal and 5 tracheal intubations. Each intubation was assessed with the bougie twice, once by a novice to the technique, and once by an assessor who was constant through the trial. Assessors used the bougie to feel for clicks of the tracheal rings and to appreciate hang up of the bougie as it was advanced into the smaller airways. Absence of these findings was presumed to indicate an esophageal intubation. Actual placement was confirmed by bronchoscopy. Each assessor made an independent determination of tube location. Descriptive statistics were used to summarize the data.

Results
Overall, 93% (95% confidence interval [CI], 86%-97%) of tracheal placements were correctly identified. The constant assessor was able to correctly identify 98% (95% CI, 90%-100%). Tracheal rings were detected in 92% of tracheal placements. Ring clicks were 95% specific for tracheal intubation. Hang up was reported in 100% of tracheal placements with a specificity of 84%. Overall, 95% (95% CI, 88%-98%) of esophageal intubations were detected. The constant assessor detected 100% of esophageal intubations.

Conclusion
In the cadaver model used in this study, the gum elastic bougie (Eschmann Tracheal Tube Introducer) shows promise as an endotracheal tube confirmation device.

10. The San Francisco Syncope Rule vs physician judgment and decision making

Quinn, Stiell, et al.Amer J Emerg Med 2005; 23:782-786

Syncope is a transient loss of consciousness with a return to preexisting neurologic function. A common problem, 1 of 4 people will faint during their lifetime, and 1% to 2% of all emergency department (ED) visits and hospital admissions are related to a transient loss of consciousness [1-4].

Patients with syncope create a difficult dilemma for physicians. Most causes are benign, but occasionally, it is a symptom associated with significant morbidity and mortality. Some patients will require emergent hospitalization for workup and treatment of life-threatening or potentially life-threatening causes, others should get outpatient evaluation, whereas some patients need no further evaluation.

It has been suggested that the use of hospitalization for patients with syncope is inefficient and highly variable [5-10]. Many things can cause syncope and the potential diseases that cause it span multiple specialties, making it difficult to develop an optimal disposition for these patients. Accordingly, a survey of physicians revealed that the disposition of patients with syncope was the second most common decision problem for North American physicians [11]. A highly sensitive and specific decision rule that would aid and improve physician decision making could have the potential to significantly reduce health care costs and improve efficiency and patient care.

The San Francisco Syncope Study is a prospective multiphase study. Phase 1 involved derivation of a decision rule using 684 patients to help predict patients at risk for acute outcomes. Variables were assessed for their interobserver agreement and univariate association with acute outcomes. The final San Francisco Syncope Rule (SFSR), derived from recursive partitioning of the most important variables, was found to be highly sensitive and specific (Fig. 1) [12]. To justify the time and effort involved in validating and disseminating a decision rule, it is important to know if the rule can improve upon the diagnostic accuracy and reliability of unstructured physician judgment and eventual decision making. We sought to determine whether the SFSR would have performed better than physician decision making during phase 1 of the study.

Abstract
Objective
To compare a clinical decision rule (San Francisco Syncope Rule [SFSR]) and physician decision making when predicting serious outcomes in patients with syncope.

Methods
In a prospective cohort study, physicians evaluated patients presenting with syncope and predicted the chance (0%-100%) of the patient developing a predefined serious outcome. They were then observed to determine their decision to admit the patient. All patients were followed up to determine whether they had a serious outcome within 7 days of their emergency department visit. Analyses included sensitivity and specificity to predict serious outcomes for low-risk patients and comparison of areas under the receiver operating characteristic curve for the decision rule, physician judgment, and admission decisions.

Results
During the study period, there were 684 visits for syncope with 79 visits resulting in serious outcomes. The area under the receiver operating characteristic curve was 0.92 (95% confidence interval [CI], 0.88-0.95) for the SFSR compared with physician judgment 0.89 (95% CI, 0.85-0.93) and physician decision making 0.83 (95% CI, 0.81-0.87). Physicians admitted 28% of patients in a low-risk group, with a median length of stay of 1 day (interquartile range, 1-2.5 days). The SFSR had the potential to absolutely decrease admissions by 10% in this low-risk group and still predict all serious outcomes.

Conclusions
Physician judgment is good when predicting which patients with syncope will develop serious outcomes, but contrary to their judgment, physicians still admit a large number of low-risk patients. The SFSR performs better than current physician performance and has great potential to aid physician decision making.

11. Use of spiral CT contrast angiography and US to exclude the diagnosis of PE in the ED

Anderson, Stiell, Wells, et al. J Emerg Med 2005;29:399-404.

The diagnosis of pulmonary embolism remains problematic. The clinical signs and symptoms of this diagnosis are mimicked by other disorders and the diagnostic test, ventilation-perfusion lung scanning, is both insensitive and non-specific (1,2). In recent years much attention has been directed towards the use of spiral computed tomography (CT) contrast angiography for the investigation of patients with suspected pulmonary embolism. However, the accuracy of spiral CT contrast angiography is uncertain and its sensitivity may be as low as 70%.

Despite its relatively poor sensitivity, spiral CT contrast angiography is used in many centers as the sole imaging test to exclude the diagnosis of pulmonary embolism. The purpose of this study was to determine the safety of a management strategy to exclude the diagnosis of pulmonary embolism in high-risk patients presenting to the Emergency Department on the basis of a negative spiral CT contrast angiography scan and negative lower extremity venous ultrasound imaging.

Abstract
Spiral computed tomography (CT) contrast angiography is a promising imaging modality for the diagnosis of pulmonary embolism but the negative predictive value of this test remains controversial. We performed a multi-center prospective cohort study to determine the safety of relying on a negative spiral CT contrast angiography scan to exclude pulmonary embolism. Patients presenting to the Emergency Departments of three tertiary care institutions with clinically suspected pulmonary embolism were potentially eligible for the study. Patients underwent a clinical evaluation to categorize pretest probability into low, moderate, and high categories, and had D-dimer testing performed. Patients at low pretest probability with normal D-dimer were considered to have pulmonary embolism excluded.

The remaining patients underwent spiral CT contrast angiography scan of the pulmonary arterial circulation and bilateral venous ultrasound of the proximal leg veins. Patients who were confirmed to have pulmonary embolism or deep vein thrombosis were treated with anticoagulant therapy. Patients in whom the diagnosis of pulmonary embolism was excluded did not receive anticoagulant therapy and were followed for a 3-month period for the development of venous thromboembolic complications.

Eight hundred fifty-eight (858) patients were enrolled in this study. Three-hundred sixty-nine (369) patients had low pretest probability and negative D-dimer results and no further diagnostic tests were performed. None of these patients subsequently developed venous thromboembolic complications (0%, 95% confidence interval [CI] 0% to 1.0%). The remaining 489 were referred for spiral CT contrast angiography scan and ultrasound.

Sixty-seven patients were confirmed to have pulmonary embolism and an additional 15 patients with negative CT scans had proximal deep vein thrombosis (DVT) on ultrasound for a total prevalence of venous thromboembolism of 82/489 (16.8%). Two of 409 patients who had pulmonary embolism excluded in the initial evaluation phase developed proximal venous thromboembolism (0.5%; 95% CI 0% to 1.8%) in the 3-month follow-up period. These findings suggest that the combination of a negative spiral CT contrast angiography scan and normal venous ultrasound imaging safely excludes the diagnosis of pulmonary embolism in the Emergency Department setting.

12. Clinical decision rules and cervical spine injury in an elderly patient: A word of caution

Thomas B. Barry, MD, Robert M. McNamara, MD. J Emerg Med 2005;29:433-436

Abstract
We report a case of a clinically significant cervical spine fracture in an elderly patient without midline cervical tenderness. Application of the NEXUS rule by the treating physicians ruled out the need for radiography. However, knowledge of the Canadian C-spine rule and clinical judgment prompted obtaining a three-view trauma series of the cervical spine and, when the patient's pain increased, a computed tomography scan of the cervical spine. A type III fracture of the dens was found. In review of the case it was recognized that application of the NEXUS rule for this patient was problematic regarding the assessment of mental status. Specifically, the treating physicians did not strictly adhere to the detailed explanations attached to the NEXUS criteria regarding mental status. Clinicians may wish to preferentially apply the Canadian rule for patients over the age of 64 years.

The treating physicians for this patient concluded that the NEXUS criteria indicated no need for CSR in this patient. In this case, the physicians did not go beyond the Glasgow Coma Scale, orientation, and general impression to determine 'normal alertness' in the application of NEXUS. Of note, the specific explanation attached to NEXUS for this item is as follows: "An altered level of alertness can include any of the following: 1) Glasgow Coma Scale of 14 or less; b) disorientation to person, place, time, or events; c) inability to remember three objects at 5 minutes; d) delayed or inappropriate response to external stimuli; or e) other."

The treating physicians for this patient did not address three-item recall at 5 min or complete a detailed assessment of the mental status. In an elderly patient, the NEXUS broad category of 'other' could presumably mean a complete mental status examination given the difficulty in recognizing cognitive defects without such testing. This report suggests that elderly patients should routinely undergo more formal mental status testing in the application of the NEXUS criteria. On the other hand, the Canadian rule with its age criteria eliminates the need for such an approach for patients over the age of 64 years, as they will all require CSR. Given the imprecise definition of altered alertness under NEXUS, the Canadian rule may be preferable for the elderly.

For the original canadian study, seeStiell IG, Wells GA, Vandemheen K, et al.. The Canadian C-spine Radiography Rule for alert and stable trauma patients. JAMA. 2001;286:1841-1848.

13. Oral anticoagulation can safely be self managed

Fitzmaurice et al BMJ 2005;331:1057

Appropriately trained, patients taking oral anticoagulants long term can safely and effectively use a point of care device to manage their own anticoagulation. In a multicentre open randomised trial of more than 600 patients taking warfarin, Fitzmaurice and colleagues found that the percentage of time spent within the therapeutic range and the incidence of adverse events were similar in patients who self managed and those who received routine care provided by anticoagulation clinics. The intervention also improved management of patients with initially poor control.

Abstract: http://bmj.bmjjournals.com/cgi/content/abstract/331/7524/1057

14. Vaccinating children against pneumococcal disease protects older adults too

In 2000, the United States authorities licensed a 7 valent pneumococcal vaccine (PCV 7) for children younger than 5 years. The incidence of invasive pneumoccocal disease in this age group fell quickly, but the benefits also extended to older US citizens. Active surveillance across eight distinct geographical areas shows that the incidence of invasive diseases, such as pneumonia, meningitis, and bacteraemia of unknown origin, has continued to fall among elderly US citizens, and has now reached the government target of 42 cases per 100 000 for the over 65s.

Researchers estimate that in 2002 and 2003, there were 12 500 fewer cases of invasive disease and 1100 fewer deaths among elderly US citizens compared with the two years before the vaccine was introduced. Adults aged between 75 and 84 years benefited most. In this age group the incidence of invasive pneumoccocal disease in 2002-3 was 35% lower than the corresponding incidence in 1998-9.

The authors think it likely that the new children's vaccine is responsible for these trends. An analysis that stratified invasive disease by serotype showed that most of the overall decline among elderly people was accounted for by the serotypes in the childhood vaccine. Disease caused by other serotypes remained stable.

JAMA 2005;294: 2043-51.

15. Drug-Resistant Avian Flu Virus Reported

The H5N1 avian flu virus has now spread from Asia to Europe. The virus is already a great threat to the world’s poultry industry, but the greater threat is that it will mutate, enabling easy infection and spread among humans. The only defenses against a new pandemic are immunization and antiviral treatment.

In vitro testing has shown that most H5N1 isolates are sensitive to the antiviral drug oseltamivir (Tamiflu). This has led many countries to stockpile the drug (for example, the U.K. has allocated funds to obtain enough to treat over 20% of its population). Flu viruses mutate readily, however, leading to concern that drug resistance could rapidly develop in an epidemic.

An international team reports that a 14-year-old Vietnamese girl infected with the H5N1 virus, and treated for 3 days with oseltamivir, developed resistance to the drug. Fortunately, the girl recovered. When ferrets were infected with the resistant virus, oseltamivir was ineffective. The oseltamivir-resistant strains of the virus were, however, easily killed by zanamivir (Relenza), in vitro and in the ferrets.

Comment: This report is discouraging, though not surprising. It calls into question the public-health value of investing in huge stockpiles of oseltamivir. Although the authors argue that their data suggest it might be useful to stockpile zanamivir, the data provide no reason to believe that resistance to that drug would not develop just as rapidly.

— Anthony L. Komaroff, MD; Published in Journal Watch November 4, 2005
Source: Le QM et al. Isolation of drug-resistant H5N1 virus. Nature 2005 Oct 20; 437:1108.

16. Reliability of Urine Collection in Young Children

The gold standard for diagnosis of urinary tract infection (UTI) is urine culture. Because bag-collected urine culture specimens are unreliable in young children, catheter specimens are most often used (or at least recommended) for children who are not toilet trained. However, catheter specimens can be difficult to obtain, can traumatize the child and family, and can introduce infection. These researchers sought to determine whether urinalysis of bag specimens can be used to screen for UTIs.

Paired bag and catheter specimens were collected from 303 non–toilet-trained children (age, <3 years) who presented to an emergency department with signs or symptoms suggestive of UTI (e.g., fever of unknown source, dysuria, foul-smelling urine). A positive urinalysis was defined as more than five white blood cells per high-power field on microscopic examination, or greater than trace leukocyte esterase or positive nitrite on dipstick testing. Culture of urine obtained by catheterization served as the gold standard. According to dipstick testing, urinalysis of the bag specimen was significantly more sensitive than urinalysis of the catheter specimen (0.85 vs. 0.71).

Comment: The authors conclude that selective catheterization (i.e., only if urinalysis of a bag specimen is positive) is appropriate in non–toilet-trained children with low likelihood of having a UTI (e.g., febrile children without UTI symptoms, recurrent UTIs, reflux, or impaired renal function). This strategy is consistent with American Academy of Pediatrics guidelines, which recommend that, when urinalysis is negative, clinicians may follow the clinical course without initiating antimicrobial therapy. An editorialist argues that the key is deciding which children do, in fact, have low likelihood of UTI.

— Robert A. Dershewitz, MD, MSc. Published in Journal Watch November 4, 2005

Source: McGillivray D et al. A head-to-head comparison: "Clean-void" bag versus catheter urinalysis in the diagnosis of urinary tract infection in young children. J Pediatr 2005 Oct; 147:451-6; Wald ER. To bag or not to bag. J Pediatr 2005 Oct; 147:418-20.