• Curing HIV?
• Inhaled Corticosteroids Might Provoke Diabetes
• Erythema Ab Igne from Laptop Computers
• Does Red Meat Consumption Raise Stroke Risk?
• Methamphetamine Abuse Continues to Flourish Despite Tracking Laws
• Morphine Overdoses with Approved Product
• Long-Term Opioids for Nonmalignant Pain
• FDA Limits Acetaminophen in Prescription Combination Products;
Requires Liver Toxicity Warnings
• Beet-Borne Nitrates Seen to Aid Exercise
• Hormone Therapy and Dementia Risk: A Critical Window?
• Otitis Media Is Better Treated with Antibiotics Than Observation
• Lost in Spacer: How Many Breaths?
Blood 2010 Dec 8
Curing HIV?
An HIV-infected man who underwent stem-cell transplantation for leukemia continues to be free of the virus more than 3.5 years later, without any antiretroviral therapy.
In early 2009, researchers described an HIV-infected man with acute myeloid leukemia (AML) who had achieved drug-free virologic control for at least 20 months after receiving a stem-cell transplant from a donor who was homozygous for the CCR5 32 allele (JW AIDS Clin Care Feb 13 2009). Now, investigators describe this patient at 45 months of follow-up.
Prior to transplantation, the patient underwent conditioning with cytotoxic agents, total body irradiation, and antithymocyte globulin. On the day of transplantation, he stopped all antiretroviral therapy (ART) and has not reinitiated it. At 13 months, his AML relapsed, he received another intensive conditioning regimen (this time including gemtuzumab, an anti-CD33 monoclonal antibody that was withdrawn from the U.S. market in 2010), and he received a second transplant from the same donor.
In the 45 months since the initial transplantation, the patient's blood and gut mucosa have been reconstituted with CCR5-negative donor CD4 cells. Notably, his new immune system is not impervious to HIV infection: His CD4 cells express the CXCR4 coreceptor and can be infected ex vivo with an exogenous X4-tropic HIV clone. Nevertheless, the investigators have not found any signs of HIV in his blood, gut mucosa, bone marrow, cerebrospinal fluid, or brain (the last of which was biopsied because of leukoencephalopathy, which turned out to be transplant-related). Moreover, his titer of HIV-specific antibodies has continued to decline over time, providing further evidence that viral antigen is no longer present.
Comment: This singular case raises important questions. Did the intensive conditioning regimens that the patient received prior to transplantation somehow eradicate HIV? Was transplantation of CCR5-negative stem cells the key to this success? Can stem cells from infected patients be manipulated to resist HIV infection and thereby abrogate viral replication after ART is stopped? The answers to these difficult questions could inform the development of safer and more practical strategies for curing HIV infection.
— Rajesh T. Gandhi, MD Published in Journal Watch HIV/AIDS Clinical Care January 10, 2011
Citation(s): Allers K et al. Evidence for the cure of HIV infection by CCR5 32/ 32 stem cell transplantation. Blood 2010 Dec 8; [e-pub ahead of print].
(http://dx.doi.org/10.1182/blood-2010-09-309591)
Top of Page
Am J Med 2010 Nov; 123:1001.
Inhaled Corticosteroids Might Provoke Diabetes
In an observational study, diabetes was more likely to develop in current steroid users.
Randomized trials have not shown a connection between use of inhaled corticosteroids (ICS) and risk for diabetes; however, these studies might have been underpowered to detect this association. In a new study, researchers evaluated risk for developing diabetes among nearly 400,000 residents of Quebec, Canada, who did not have diabetes when they began ICS treatment between 1990 and 2005.
During a mean follow-up of 5.5 years, 30,000 people initiated treatment for diabetes. After adjustment for multiple factors, including respiratory disease severity and comorbid conditions, incidence of diabetes was 34% higher among people who were currently receiving ICS than among those who were not. A dose–response association also was detected (rate ratios, 1.18 for low-dose ICS; 1.64 for high-dose ICS). Use of ICS also was associated with 34% higher risk for insulin therapy.
Comment: This population-based observational study showed a significantly higher incidence of diabetes and more-rapid diabetes progression (defined as initiating insulin therapy) among individuals who received ICS. Current guidelines recommend close monitoring of patients who receive high-dose ICS for development of osteoporosis or cataracts; this study suggests that diabetes could be added to this list. The findings also remind us to follow asthma and chronic obstructive pulmonary disease guidelines on when to initiate inhaled corticosteroid therapy and on using the lowest effective dose.
— Jamaluddin Moloo, MD, MPH Published in Journal Watch General Medicine January 11, 2011
Citation(s): Suissa S et al. Inhaled corticosteroids and the risks of diabetes onset and progression. Am J Med 2010 Nov; 123:1001.
(http://dx.doi.org/10.1016/j.amjmed.2010.06.019) PMID:20870201
Top of Page
Pediatrics 2010 Nov; 126:e1227
Erythema Ab Igne from Laptop Computers
An easily prevented condition reminds us not to get too close to our computers.
Erythema ab igne (EAI) is a reticulated, maroon-to-brownish dermatosis that develops after repeated exposure to a heat source. Historically, it was observed on the lower legs of individuals exposed to chronic heat from open fires or coal stoves. It became less common after the advent of central heating. In modern times, the elderly or those with chronic pain may develop EAI after repeated use of heating pads. EAI from a laptop computer was first reported in 2004. Since then, it has been reported in 10 laptop users, including, recently, the youngest, a 12-year-old boy.
The authors of this new report reviewed all 10 cases. The typical location of the dermatosis was the anterior thigh, appearing after several months of laptop use at an average of 6 to 8 hours a day, although use by the 12-year-old was only 2 to 3 hours a day. The asymmetric distribution is nearly pathognomonic and correlates with placement of the heat-generating areas of the laptop, usually the optical drive, the battery, or the ventilation fan.
EAI is diagnosed clinically, and no effective treatment exists, although the discoloration may gradually fade over time with avoidance of the heat source.
Comment: With the increasing use of laptops for work and recreation, the prevalence of laptop erythema ab igne is likely to rise. Placement of a laptop case or other material between the skin and the computer to avoid direct contact with the heated area should easily prevent this problem.
— Mary Wu Chang, MD Published in Journal Watch Dermatology January 14, 2011
Citation(s): Arnold AW and Itin PH. Laptop computer induced erythema ab igne in a child and review of the literature. Pediatrics 2010 Nov; 126:e1227 PMID: 20921068
Top of Page
Stroke 2010 Dec 16
Does Red Meat Consumption Raise Stroke Risk?
Swedish women in the highest stratum of red meat consumption were more likely to experience cerebral infarctions than were those in the lowest stratum.
Findings about red meat consumption and stroke risk are sparse and inconsistent. Investigators assessed self-reported red meat consumption in 34,670 participants in the Swedish Mammography Cohort in relation to incidence of stroke (cerebral infarction, intracerebral or subarachnoid hemorrhage, or unspecified stroke) as reported in the Swedish Hospital Discharge Registry during a mean follow-up of 10.4 years.
After multivariate adjustment, women in the highest decile of red meat consumption (>102 g or >3.6 ounces daily) were at 42% higher risk for cerebral infarction than were women in the lowest decile (<25 g or <0.9 ounces daily). Subgroup analysis showed that never-smokers without diabetes in the highest decile of red meat consumption were at 68% higher risk for cerebral infarction than were similar women in the lowest decile. For processed meat consumption, women in the highest quintile were at 24% higher risk for cerebral infarction than were those in the lowest quintile. No significant association was found between red or processed meat consumption and risk for intracerebral or subarachnoid hemorrhage.
Comment: These results agree with those from the Nurses' Health Study, in which women who consumed 1 serving of red or processed meat daily were at excess risk for ischemic stroke (as well as hormone receptor–positive breast cancer [JW Womens Health Jan 11 2007]). One limitation of this study is that constant consumption of red and processed meat for 10 years was assumed (diet was only ascertained once). Furthermore, applicability to U.S. women might be limited: Many might consider 3.6 ounces of red meat daily to be low consumption! Nonetheless, these results show that even healthy, nonsmoking women can lower their stroke risk by cutting back on red meat intake; clinicians should consider inquiring about the dietary habits of all their patients
— Wendy S. Biggs, MD Published in Journal Watch Women's Health January 13, 2011
Citation(s): Larsson SC et al. Red meat consumption and risk of stroke in Swedish women. Stroke 2010 Dec 16; [e-pub ahead of print].
(http://dx.doi.org/10.1161/STROKEAHA.110.596510)
Top of Page
Methamphetamine Abuse Continues to Flourish Despite Tracking Laws
Tracking the sales of cold medicines used to make methamphetamine have failed to slow the drug trade. Instead, they have created a large, highly lucrative market for profiteers to buy over-the-counter products and sell them to meth producers at a huge markup.
Thousands of new people have entered into the methamphetamine underworld over the last few years, according to an AP review, stating that it's almost like a subcriminal culture. DEA agents explain that the lawbreakers use GPS units set up in a van with a list of every single pharmacy or retail outlet. They'll spend the entire week going store to store and buy to the limit. Next, inside their vehicles they punch out blister packs into a bucket and even clip coupons.
http://www.google.com/hostednews/ap/article/
Top of Page
Morphine Overdoses with Approved Product
Roxane Laboratories and the FDA have notified healthcare professionals that serious adverse events and deaths have resulted from accidental overdoses of morphine sulfate oral solutions (especially with the high potency 100-mg/5-mL product). The morphine sulfate oral solutions were ordered in milligrams (mg) but were mistakenly interchanged for milliliters (mL) of the product in most of these cases.
Morphine sulfate oral solution is a pre-1938 unapproved drug that has been on the market for well over 50 years. The approval of this specific product is part of FDA's unapproved drugs initiative. Up to this recent approval, Roxane marketed a morphine sulfate oral solution with the strength expressed as 20 mg/mL, using a container label and carton labeling that had brown lettering on a white background. The newly FDA approved product labeling and packaging feature revisions were intended to reduce the risk of medication errors.
http://www.fda.gov/Safety/MedWatch/SafetyInformation/
Top of Page
N Engl J Med 2010 Nov 18; 363:1981
Long-Term Opioids for Nonmalignant Pain
The dilemma: Undertreating patients with pain or overprescribing opioids?
Scrutiny of long-term use of narcotic analgesics for nonmalignant pain has sharpened because of increased frequency of use, higher awareness of adverse events, and more reports of deaths from unintentional overdose. According to the CDC, unintentional drug overdose is the second-leading cause of accidental death in the U.S., with about 40% of such deaths caused by prescription opioids.1 A focus on perceived undertreatment of pain, coupled with aggressive marketing of opioids such as extended-release oxycodone (OxyContin), might have produced unintended negative consequences, particularly the long-term and potentially harmful use of opioids for nonspecific musculoskeletal pain. In several new studies, researchers addressed these issues.
In a study from two large western U.S. health plans with about 4 million enrollees, the prevalence of long-term opioid use for noncancer pain increased strikingly from 1997 to 2005 in all age groups. In older women (age, 65), the prevalence of chronic opioid use rose from 5% to nearly 9% during that 8-year period; in older men, the prevalence rose from 3% to 5%. Among middle-aged people (age range, 45–64), the prevalence was also about 5% in 2005.2
In two studies, investigators evaluated a Medicare database from Pennsylvania and New Jersey that included beneficiaries who initiated analgesic therapy for nonmalignant pain from 1999 through 2005. In one study, researchers compared three cohorts, each with 4280 older arthritis patients whose demographic and clinical characteristics were matched except for the prescribed analgesic — nonselective nonsteroidal anti-inflammatory drug (nsNSAID), coxib, or opioid. Compared with risk for adverse cardiovascular events in the nsNSAID group, relative risk was higher with coxibs and opioids by 28% and 77%, respectively. Relative risk for gastrointestinal bleeding was 40% lower with coxibs than with opioids or nsNSAIDs. Opioids were associated with a fourfold higher fracture risk, 70% higher risk for hospitalization, and doubling of all-cause mortality, compared with nsNSAIDs.3
In the other study, the researchers compared five cohorts, each with 6275 patients who used one of five different classes of opioids — codeine, hydrocodone, oxycodone, propoxyphene, or tramadol; otherwise the cohorts were similar demographically and clinically. Compared with hydrocodone, codeine was associated with 60% higher risk for adverse cardiovascular events; tramadol and propoxyphene were associated with lower relative risks for fracture (by 80% and 46%, respectively); and oxycodone and codeine were associated with doubled mortality.4
Editorialists commented on both of these studies and urged some caution with regard to their clinical implications, both because of methodologic weaknesses in cohort studies that involve administrative databases and because of the lack of biological plausibility for some of the results (such as the cardiovascular risks of codeine). However, they emphasize that risk for falls and fractures and the associated excess all-cause mortality of long-term opioid use, particularly oxycodone and codeine, mandate greater caution in their use, greater focus on nonpharmacologic approaches to treating musculoskeletal pain, and additional controlled clinical trials.5,6
Legislative and regulatory responses to opioid prescribing are on the horizon. The FDA has proposed a new Risk Evaluation and Management Strategy (REMS) for opioids, but the REMS has not been approved yet, because an FDA committee wants stronger requirements, including mandatory physician education programs. And, in Washington State, a new law that takes effect in mid-2011 will require documented evaluations (diagnosis, treatment plan, and objectives) before physicians initiate long-acting opioid therapy.1 The law also will require written agreements signed by patients, periodic review of patients' progress, continuing education for physicians who prescribe long-acting opioids, and mandatory consultation with pain management specialists when physicians prescribe morphine equivalent doses exceeding 120 mg daily. Notably, the law does not apply to patients who have chronic pain caused by cancer, or acute pain caused by injury or surgery, or who are receiving hospice or end-of-life medical care.
Finally, in a provocative essay that captures the dilemma often experienced in primary care practice, a physician describes his struggle to simultaneously avoid undertreatment of pain and overprescribing of opioids. He concludes that an entirely new approach is needed, including more focus on mental health care, referral access to pain specialists, and guidelines that set clear limits on opioid dosage.7
— Thomas L. Schwenk, MD, and Allan S. Brett, MD Published in Journal Watch General Medicine January 13, 2011
Citation(s):1. Okie S. A flood of opioids, a rising tide of deaths. N Engl J Med 2010 Nov 18; 363:1981.
Top of Page
FDA Limits Acetaminophen in Prescription Combination Products;
Requires Liver Toxicity Warnings
A maximum of 325 milligrams of acetaminophen per tablet/capsule has been requested by the FDA to manufacturers to reduce risk of liver toxicity. The FDA is also requiring manufacturers to update labels of all prescription combination acetaminophen products to warn of the potential risk for severe liver injury.
The higher-dose prescription combination acetaminophen products will be phased in over three years and should not create a shortage of pain medication. The FDA believes that prescription combination products containing no more than 325 mg of acetaminophen per tablet are effective for treating pain.
http://pharmalive.com/news/index.cfm?articleID=755057&categoryid=9&newsletter=1
Top of Page
Beet-Borne Nitrates Seen to Aid Exercise
Beet juice yields performance benefits during walking and running; the findings likely extend to whole beets, borscht, and all nitrate-rich vegetables
by Craig Weatherby
Because of concerns about processed beef and pork products, it’s become common to associate dietary nitrates with cancer.
Actually, the cancer risks associated with nitrate-preserved processed meats probably have more to do with the meat than the nitrates.
And to remove any risk from the nitrates, lunch-meat makers have long been required to add vitamin C (ascorbic acid), which blocks creation of nitrate-based carcinogens.
Nitrates aren’t inherently carcinogenic, and they’re pretty abundant in common vegetables (see our sidebar, “Vegetables by nitrate content”, below).
In fact, it’s looking more and more as though some of vegetables’ apparent health benefits stem from their nitrate content.
For example, last November we reported on a study in older adults, which showed that drinking juice from a nitrate-rich vegetable increased blood flow to key areas of the brain.
Now, research suggests that beet juice may also aid exercise capacity and efficiency.
And the findings likely extend to whole beets, borscht, and all nitrate-rich veggies.
Nitrates singled out as source of beet juice’s exercise benefits
Prior research from Britain’s University of Exeter showed that beet juice enabled people to exercise up to 16 percent longer and produced other substantial performance and health benefits (Bailey SJ et al. 2009; Bailey SJ et al. 2010; Vanhatalo A et al. 2010).
Amazingly, the humble vegetable’s deep red juice reduced oxygen needs to an extent never achieved in any other way.
The University of Exeter team wanted to confirm that the benefits of beet juice derived from its nitrates, as they supposed.
They enrolled nine healthy, physically active males for a randomized, double-blind, cross-over trial.
Some were assigned to drink regular beet juice, while others drank a nitrate-free beet juice.
The scientists compared the two groups on a range of measures such as blood pressure and mitochondrial oxidative capacity (Qmax) … a measure of how efficiently the muscles use oxygen.
The study began by recording all the participants’ scores on treadmill tests, maximal oxygen intake, blood pressure, heart rate, lung function and blood nitrite concentration. (Dietary nitrates largely convert to nitrites in the body.)
The participants were then randomized to receive 0.5 liters (16 ounces) per day of either regular beet juice (containing 6.2 mmol of nitrate) or deliberately nitrate-depleted beet juice (containing 0.003 mmol of nitrate) for six days.
The volunteers took part in treadmill exercise tests on days four and five. On day six, the men were asked to perform knee extension tests.
The knee extensions were designed to lower participants’ muscle phosphocreatinine levels, and also to allow the researchers to estimate the participants’ Qmax.
(Phosphocreatinine acts as a rapidly available reserve of high-energy phosphates in skeletal muscle and the brain.)
All of the tests were repeated during a second, “cross-over” period when the men in the two groups switched drinks for another six days.
Compared with drinking nitrate-depleted beet juice, drinking regular beet juice raised the men’s blood nitrite levels and reduced their systolic blood pressure.
Drinking regular beet juice also reduced the participants’ oxygen requirements during treadmill walking, moderate- or high-intensity running.
And drinking regular beet juice lengthened so-called “time-to-exhaustion” during severe-intensity running by 15 percent.
There was no difference in Qmax between drinking nitrate-free and regular beet juice.
Drawing the obvious conclusion, the Exeter team wrote, “These results indicate that the positive effects of six days of BR supplementation on the physiological responses to exercise can be ascribed to the high nitrate content.” (Lansley KE et al. 2010)
Sources
- Bailey SJ, Winyard P, Vanhatalo A, Blackwell JR, Dimenna FJ, Wilkerson DP, Tarr J, Benjamin N, Jones AM. Dietary nitrate supplementation reduces the O2 cost of low-intensity exercise and enhances tolerance to high-intensity exercise in humans. J Appl Physiol. 2009 Oct;107(4):1144-55. Epub 2009 Aug 6.
- Bailey SJ, Fulford J, Vanhatalo A, Winyard PG, Blackwell JR, DiMenna FJ, Wilkerson DP, Benjamin N, Jones AM. Dietary nitrate supplementation enhances muscle contractile efficiency during knee-extensor exercise in humans. J Appl Physiol. 2010 Jul;109(1):135-48. Epub 2010 May 13. Erratum in: J Appl Physiol. 2010 Sep;109(3):943.
- Lansley KE, Winyard PG, Fulford J, Vanhatalo A, Bailey SJ, Blackwell JR, Dimenna FJ, Gilchrist M, Benjamin N, Jones AM. Dietary nitrate supplementation reduces the O2 cost of walking and running: a placebo-controlled study. J Appl Physiol. 2010 Nov 11. [Epub ahead of print]
- Vanhatalo A, Bailey SJ, Blackwell JR, DiMenna FJ, Pavey TG, Wilkerson DP, Benjamin N, Winyard PG, Jones AM. Acute and chronic effects of dietary nitrate supplementation on blood pressure and the physiological responses to moderate-intensity and incremental exercise. Am J Physiol Regul Integr Comp Physiol. 2010 Oct;299(4):R1121-31. Epub 2010 Aug 11.
Top of Page
Ann Neurol 2010 Nov 12;
Hormone Therapy and Dementia Risk: A Critical Window?
A prospective cohort study reveals a decreased risk for dementia in women taking HT in midlife but an increased risk for those taking HT in later life.
Several observational studies have demonstrated a reduced risk for dementia in women who use hormone therapy (HT). However, well-designed interventional trials of estrogen–progesterone combinations have revealed an increased risk for dementia, cancer, and vascular events associated with HT use. This prospective cohort study was designed to further investigate the critical window theory that estrogen is beneficial only immediately before and immediately after menopause and may have deleterious effects in later life. More than 5000 women without dementia at baseline had been screened between ages 40 and 55 and reported whether they used HT (i.e., midlife use). Thirty years later, the authors identified participants' HT use (i.e., later-life use) from pharmacy databases. Starting when the cohort reached a median age of 80, the authors identified dementia from ICD-9 coding of diagnoses by neurologists, neuropsychologists, and internists; dementia assessment lasted 7.5 years.
During dementia assessment, 27% of the cohort received a diagnosis of dementia. Women who reported taking HT at midlife but had no evidence of later-life HT use had a significantly decreased risk for dementia diagnosis (adjusted hazard ratio, 0.74). In contrast, those who did not report midlife HT use but had evidence of later-life use had a significant increase in risk (AHR, 1.48). Women who had used HT in both midlife and later life had similar risks to those of women who had no evidence of HT use.
Comment: This study is potentially confounded by self-report data in its early phase and by the limitations inherent in the accuracy of pharmacy records and diagnostic coding in its later phase. Nonetheless, the findings unify data from observational and interventional studies of women using hormone therapy. Moreover, the findings are consistent with animal modeling of a critical window of estrogen-related neuroprotection, which is reportedly associated with improved cerebral blood flow and glucose use and with reduced amyloid deposition in the brain. The bottom line clinically is that HT use around menopause may help protect against dementia, but later use may increase risk.
— Brandy R. Matthews, MD Published in Journal Watch Neurology January 4, 2011
Citation(s):Whitmer RA et al. Timing of hormone therapy and dementia: The critical window theory revisited. Ann Neurol 2010 Nov 12; [e-pub ahead of print]. (http://dx.doi.org/10.1002/ana.22239)
Top of Page
Otitis Media Is Better Treated with Antibiotics Than Observation
In children with acute otitis media, two placebo-controlled trials establish that antibiotic treatment with amoxicillin-clavulanate is superior to watchful waiting. The trials appear in the New England Journal of Medicine.
Researchers in Finland and the U.S. separately randomized a total of some 600 children under age 3 years to roughly 10 days of double-blind treatment with either amoxicillin-clavulanate or placebo. All children had been diagnosed with acute otitis media according to strict standards.
In both trials, treatment failure was at least twice as frequent among placebo recipients. Rash and diarrhea were more common among those receiving antibiotics.
An editorialist writes that the trials provide "the best data yet" to answer the question: Is acute otitis media treatable? "The answer is yes," he writes.
http://www.nejm.org/doi/full/10.1056/NEJMoa0912254
Top of Page
Pediatrics 2010 Dec; 126:e1493
Lost in Spacer: How Many Breaths?
Children inhale as much aerosolized medication through a spacer with two or three regular inhalations as they do with more inhalations.
Asthma guidelines and manufacturer instructions recommend that children inhale from five to "several" times, using normal tidal breathing, when receiving a dose of medication from a pressurized metered dose inhaler (MDI) attached to a valved spacer. Alternatively, a single maximal inhalation is recommended for certain spacers if children using them are able to perform this maneuver. Using a custom-built device, investigators recorded drug delivery with two to nine tidal breaths and a single maximal inhalation in 118 Australian children (age range, 2–7 years) whose asthma treatment included inhaled steroids. The device simulated each child's breathing pattern from an MDI via various spacers: AeroChamber Plus (small volume, 149 mL), Funhaler (small volume, 225 mL), Volumatic (large volume, 750 mL), and a modified soft-drink bottle (valveless, large volume, 500 mL).
For the two small-volume spacers, a mean of about 40% of aerosolized MDI medication was delivered with both two and nine tidal breaths; for the Volumatic spacer, a mean of about 40% was delivered with three and nine breaths; for the soft-drink bottle spacer, about 50% was delivered with two and nine breaths. Most children aged 5 years and older were able to perform a single maximal inhalation, but this method resulted in the same or less drug delivery as the tidal breaths. In fact, the children's single tidal inhalations with a spacer were higher than predicted by their weight and were actually statistically similar in volume to single maximal inhalations.
Comment: To ensure adequate delivery of medication to the lungs, young children should use spacers with their MDIs. Getting some children to use spacers correctly and patiently can be difficult, and, without data on drug delivery, the number of inhalations has been arbitrary. This study gives pediatricians and caregivers practical information about optimal use: two normal tidal inhalations for small-volume spacers and three for large-volume spacers.
— Cornelius W. Van Niel, MD Published in Journal Watch Pediatrics and Adolescent Medicine January 12, 2011
Citation(s):Schultz A et al. Aerosol inhalation from spacers and valved holding chambers requires few tidal breaths for children. Pediatrics 2010 Dec; 126:e1493. PMID: 21078734
