Daily Aspirin Use for Longer than 4 Years Associated with Reduced Long-Term Risk of Cancer Death
Aspirin has been linked to a reduction in the risk of certain colorectal cancers (N Engl J Med 2007 May 24;356(21):2131), but its effect on other cancers is unknown. A new systematic review assessed the relationship between aspirin use and overall cancer mortality using data from 8 randomized trials originally designed to evaluate aspirin for primary or secondary prevention of cardiovascular disease. A total of 25,570 patients were randomized to receive daily aspirin (≥ 75 mg) vs. placebo for ≥ 4 years. There were 674 total cancer deaths during the trials. Aspirin use was associated with reduced risk of cancer death (odds ratio 0.79, 95% CI 0.68-0.92), with an NNT of 106-429, based on 3% cancer mortality in the control groups (level 2 [mid-level] evidence). In a pooled analysis of individual patient data from 7 trials, there was no significant difference in cancer deaths between aspirin and control groups within the first 5 years of follow-up. However at ≥ 5 years, aspirin use was associated with reduced risk of death from any cancer (hazard ratio [HR] 0.66, 95% CI 0.5-0.87), from any gastrointestinal cancer (HR 0.46, 95% CI 0.27-0.77), and from any solid cancer (HR 0.64, 95% CI 0·49-0·85).
Three trials had 20-year follow-up data. In an analysis of 12,659 patients with 1,378 total cancer deaths, aspirin was associated with reduced 20-year risk of death from any gastrointestinal cancer (HR 0.65, 95% CI 0.54-0.78) and from any solid cancer (HR 0.8, 95% CI 0.72-0.88). Aspirin doses were 75 mg/day in 1 trial and ≥ 300 mg/day in the other 2 trials. The benefits of aspirin were similar for the different dosages. The greatest absolute reduction in risk of cancer death occurred in patients ≥ 65 years old (Lancet 2010 Dec 7 early online).
For more information, see the Aspirin topic in DynaMed. This article was featured in DynaMed Weekly Update 50.
_________________________________________________
Compression-Only CPR and Standard CPR Appear to be Similarly Effective for Out-of-Hospital Cardiac Arrest When Performed by Untrained Bystanders
In cases of out-of-hospital cardiac arrest, early cardiopulmonary resuscitation (CPR) initiated before the arrival of trained emergency services may improve a patient's chances of survival. However, untrained bystanders may be reluctant or unable to correctly perform the rescue breathing portion of standard CPR. Improper rescue breathing may reduce the effectiveness of standard CPR. Observational results have suggested that CPR with chest compression alone may be at least as effective as full CPR when performed by untrained people (Lancet 2007 Mar 17;369(9565):920).
Two recent randomized trials compared these two approaches in patients with suspected cardiac arrest. In both trials, bystanders with no previous CPR experience who were calling for emergency assistance were instructed by dispatchers to perform either compression-only CPR or standard CPR (compression and rescue breathing) prior to the arrival of paramedics. In 1 trial with an analysis of 1,941 patients, survival to hospital discharge was 12.5% in the compression-only group and 11% in the standard group (not significant), with 14.4% vs. 11.5% surviving with favorable neurologic status (not significant) (N Engl J Med 2010 Jul 29;363:423). In the second trial, with an analysis of 1,276 patients, 1-day survival was 24% for compression-only vs. 20.9% for standard CPR (not significant) and 30-day survival was 8.7% vs. 7% (not significant) (N Engl J Med 2010 Jul 29;363:434). The analyses in both trials included only about one-third of the randomized patients due to exclusions following randomization (level 2 [mid-level] evidence).
These results do not suggest that chest compression alone is a substitute for standard CPR performed by trained individuals. However, this simplified approach to resuscitation may make untrained people more likely to start CPR. Medical offices receiving emergency calls from families of patients might wish to consider compression-only CPR as an immediate intervention.
For more information, see the Cardiac arrest topic in DynaMed. This article was featured in DynaMed Weekly Update 31.
_________________________________________________
Intensive Systolic Blood Pressure Control May Not Reduce Mortality in Patients with Diabetes
Current guidelines from the American Diabetes Association (ADA) and other organizations recommend a blood pressure target < 130/80 mm Hg for patients with diabetes (Diabetes Care 2010 Jan;33 Suppl 1:S11). To date, there has been little experimental data to guide blood pressure target recommendations, but a new trial directly compared 2 different blood pressure goals. The Action to Control Cardiovascular Risk in Diabetes blood pressure trial (ACCORD BP) compared intensive systolic blood pressure control (target < 120 mm Hg) vs. standard control (target < 140 mm Hg) in 4,733 patients. While the trial did not stipulate specific antihypertensive regimens, patients in both groups were required to receive a drug class associated with reduction in cardiovascular events in patients with diabetes (ACE inhibitors, angiotensin receptor blockers, beta blockers, calcium channel blockers, or diuretics). They could also receive other medications as necessary. Mean systolic blood pressure from the end of first year to end of follow-up was 119.3 mm Hg for the intensive group and 133.5 mm Hg for the standard group. Systolic blood pressure target < 120 mm Hg does not reduce mortality or myocardial infarction but does reduce nonfatal stroke compared to target < 140 mm Hg (level 1 [mid-level] evidence). During a mean follow-up of 4.7 years, cardiovascular mortality was 2.5% in each group. There were no significant differences in all-cause mortality (6.3% vs. 6.1%), nonfatal myocardial infarction (5.3% vs. 6.2%), or heart failure (3.5% vs. 3.8%). Stroke occurred in 1.7% of the intensive group compared to 2.6% of the standard group (p = 0.01, NNT 84). However, the risk of serious adverse events from treatment was increased for the intensive group (3.3% vs. 1.3%, p < 0.001, NNH 50). Adverse events reported (not all individually significant) included hypotension, syncope, bradycardia or other arrhythmia, hyperkalemia, angioedema, and renal failure (N Engl J Med. 2010 Apr 29;362(17):1575).
For more information, see the Hypertension treatment in patients with diabetes topic in DynaMed. This article was featured in DynaMed Weekly Update 13.
_________________________________________________
Glucosamine and Chondroitin either Separately or in Combination Appear to Lack Efficacy for Improving Pain in Osteoarthritis of Hip or Knee
In an earlier review of 25 randomized trials with significant statistical heterogeneity, there was inconsistent evidence for pain relief with glucosamine, but there was no significant improvement in analysis of 11 trials that had adequate allocation concealment (Cochrane Database Syst Rev 2009 Oct 7;(4):CD002946).
Glucosamine and chondroitin, alone and in combination, have become popular in recent years as both prescription and over-the-counter supplements for joint pain. A recent systematic review of 10 randomized trials evaluated their efficacy in 3,803 patients with hip or knee osteoarthritis. The review excluded trials with fewer than 100 patients in each arm. Neither supplement was found to produce clinically significant reductions in pain in any formulation (level 2 [mid-level] evidence). In all studies, pain was measured on a 10 cm visual analog scale, with a change of > 0.9 cm deemed clinically significant. The mean reductions compared to placebo were 0.4 cm (95% CI -0.7 to -0.1 cm) for glucosamine alone in analysis of 7 trials, 0.3 cm (95% CI -0.7 to 0 cm) for chondroitin alone in analysis of 4 trials, and 0.5 cm (95% CI -0.9 to 0 cm) for the combination in 1 trial. There were no significant differences in adverse events in comparisons of either drug to placebo (BMJ 2010 Sep 16;341:c4675full-text).
For more information, see the Degenerative joint disease of the hip and Degenerative joint disease of the knee topics in DynaMed. This article was featured in DynaMed Weekly Update 41.
_________________________________________________
Initial Oral Corticosteroid Treatment Appears as Effective as Initial IV Corticosteroids for Patients Hospitalized with Acute Exacerbation of COPD
High-dose intravenous corticosteroids are widely prescribed for patients admitted to the hospital with acute exacerbation of COPD, despite guidelines recommending oral steroids at lower doses (for a synthesis of 3 guidelines, see National Guideline Clearinghouse 2010 May 24:COPD_ACUTE10). A recent retrospective analysis of 79,985 hospitalized patients who received steroids for COPD exacerbation at 414 centers in the United States found that 92% were initially treated with IV steroids, while only 8% initially received oral steroids. However, outcome measures suggest that oral steroids for initial treatment may be at least as effective as IV steroids (level 2 [mid-level] evidence). Over the first 2 days, median doses (prednisone equivalents) were 60 mg in patients receiving oral steroids and 600 mg in patients receiving IV steroids. A total of 1,356 patients (22%) initially prescribed oral steroids were switched to IV steroids during hospitalization. The primary measure was treatment failure, defined as the composite of 3 outcomes: in-hospital mortality, initiation of mechanical ventilation after the second hospital day, and readmission for acute exacerbation of COPD within 30 days. Treatment failure occurred in 10.3% of the oral group and 10.9% of the IV group (not significant). In-hospital mortality was significantly lower in the oral group (1% vs. 1.4%, p = 0.01).
To control for potential baseline differences, a subgroup analysis compared patients from the 2 groups that were matched by "propensity scores" for initial treatment with oral steroids. Propensity scores were based on patient characteristics (including demographic and insurance factors), comorbidities, all other early treatments and diagnostic tests, and hospital characteristics. In this comparison, initial treatment with oral steroids was associated with reduction in risk of treatment failure (odds ratio 0.84, 95% CI 0.75-0.95), shorter hospital stay (odds ratio 0.9, 95% CI 0 0.88-0.91), and lower hospital costs (odds ratio 0.91, 95% CI 0.89-0.93) (JAMA 2010 Jun 16;303(23):2359).
For more information, see the Acute exacerbation of chronic bronchitis topic in DynaMed. This article was featured in DynaMed Weekly Update 25.
_________________________________________________
Lenient Rate Control and Strict Rate Control for Atrial Fibrillation Appear to have Similar Cardiovascular Outcomes
Rate control and rhythm control strategies have been associated with similar rates of stroke and death in patients with chronic atrial fibrillation (Arch Intern Med 2005 Feb 14;165(3):258). A target heart rate of 60-80 beats per minute has been suggested by the American College of Cardiology, but this has not been based on studies with clinical outcomes. To address this issue, a randomized trial compared "lenient" rate control (resting heart rate < 110 beats/minute) vs. "strict" rate control (resting heart rate < 80 beats/minute and heart rate during moderate exercise < 110 beats/minute) in patients with permanent atrial fibrillation. The desired rate control was achieved through the use of beta blockers, calcium channel blockers, and digoxin, alone or in combination. The primary outcome was a composite of death, hospitalization for heart failure, stroke, major bleeding, or significant arrhythmic events. The 3-year estimated cumulative incidence of the primary composite outcome was 12.9% with lenient control vs. 14.9% with strict control (level 2 [mid-level] evidence). There were no significant differences between groups in rates of dyspnea, fatigue, palpitations, or New York Heart Association (NYHA) functional class. The authors concluded that lenient rate control was as effective as strict rate control for the prevention of major cardiovascular events (N Engl J Med. 2010 Apr 15;362(15):1363).
For more information, see the Rate control in atrial fibrillation topic in DynaMed. This article was featured in DynaMed Weekly Update 12.
_________________________________________________
Probiotic Lactobacillus reuteri May Reduce Crying Time in Infants with Colic Symptoms
Colic can be stressful for newborns and new parents alike. Previous research has suggested that L. reuteri may reduce symptoms compared to simethicone in colicky infants (Pediatrics 2007 Jan;119(1):e124). A new blinded, placebo-controlled trial with 50 infants provides further support for the efficacy of probiotics. Infants aged 2-16 weeks old with fussy crying episodes lasting at least 3 hours over at least 3 days (modified Wessel's criteria for colic diagnosis) were randomized to L. reuteri vs. placebo for 21 days. L. reuteri was given as 108 colony-forming units in 5 drops of oil suspension, 30 minutes before morning feeding. All infants had been exclusively breastfed, and mothers were asked to avoid cow's milk during the trial. At baseline, the median crying time was 370 minutes per day for the L. reuteri group and 300 minutes per day for controls (not significant). At 21 days' follow-up, the L. reuteri group had significantly reduced crying time (median total crying time 35 minutes per day vs. 90 minutes per day, p = 0.022) (level 2 [mid-level] evidence). The proportion of infants with at least 50% reduction in crying time from baseline was significantly higher in the L. reuteri group at 7, 14, and 21 days (Pediatrics 2010 Sep;126(3):e526).
For more information, see the Infantile colic topic in DynaMed. This article was featured in DynaMed Weekly Update 34.
_________________________________________________
Short-Acting Antibiotics for 5 Days Might Be Less Effective than Longer Course Antibiotic Treatments for Children with AOM
Acute otitis media (AOM) is the most common reason children are prescribed antibiotics in the United States. While the optimal treatment duration is unclear, previous studies, including a Cochrane review from 2000, have suggested that short courses of treatment (< 7 days) may be as effective as longer courses. A recent update of that Cochrane review provides evidence that treatment with short-acting antibiotics for 5 days is associated with significantly higher rates of treatment failure than longer course treatments (level 2 [mid-level] evidence). The review examined 49 moderate-quality randomized trials comparing short vs. long course antibiotic regimens in 12,000 children (aged 1 month to 18 years) with a clinical diagnosis of acute otitis media. The primary outcome was treatment failure within 1 month, defined as absence of clinical resolution, relapse, or recurrence of AOM following initiation of therapy. In an analysis of 16 trials of short-acting antibiotics (e.g. amoxicillin), treatment for 5 days was associated with increased risk of treatment failure (OR 1.34, 95% CI 1.15-1.55) compared to 8-10 days of treatment. This corresponds to an NNH of 11-41, assuming treatment failure in 16% of children receiving longer treatment. This result might be due to differences in efficacy for children with and without a history of ear infections. Five-day treatment had a higher failure rate in analysis of 6 trials that included children with a history of recurrent AOM, but not in 4 trials that excluded these patients. For oral azithromycin, treatment for 3-5 days was associated with increased treatment failure at 8-19 days' follow-up compared to longer treatments (OR 1.27, 95% CI 1.04-1.55), but there was no significant difference by 20-30 days' follow-up. The risks of gastrointestinal adverse effects were significantly reduced with short course treatments for both short-acting antibiotics and azithromycin. The authors suggest that clinicians must weigh the relative benefits and harms of longer treatment durations (Cochrane Database Syst Rev 2010 Sep 08;(9):CD001095).
Guidelines from the American Academy of Pediatrics recommend antibiotics for 10 days in children < 6 years old and children with severe disease and for 5-7 days in children ≥ 6 years old with mild to moderate disease (Pediatrics. 2004 May;113(5):1451).
For more information, see the Acute otitis media (AOM) topic in DynaMed.
This article was featured in DynaMed Weekly Update 39.
_________________________________________________
Sentinel Lymph Node Resection Alone Is As Effective As Axillary Lymph Node Dissection For Women with Node-Negative Invasive Breast Cancer
Sentinel lymph node (SLN) resection has been advocated as an option to reduce morbidity for women undergoing surgery for breast cancer. When the sentinel node is negative for breast cancer, SLN resection without axillary lymph node dissection (ALND) should minimize the adverse effects, such as arm swelling, numbness, and tingling. New reports from the National Surgical Adjuvant Breast and Bowel Project (NSABP) B-32 trial show that SLN resection is as effective for survival as ALND and may reduce post-surgical morbidity and symptoms in node-negative women. A total of 5,611 women with invasive breast cancer and clinically negative nodes were randomized to SLN resection with ALND only if SLN-positive vs. immediate ALND without waiting for biopsy results. The primary analysis was performed on 3,986 women (71%) who were SLN-negative. There were no significant differences between groups in overall survival (90.3% vs. 91.8%) or progression-free survival (81.5% vs. 82.4%) over 8 years (level 1 [likely reliable] evidence). There were also no significant differences in local recurrence (2.4% vs. 2.7%), distant metastases (3.2% vs. 2.8%), cancer in the opposite breast (2.2% vs. 2.8%), or second non-breast cancer (5.4% vs. 4.5%) (Lancet Oncol 2010 Oct;11(10):927).
In a report of 3-year post-surgical follow-up, SLN resection alone was associated with reductions in residual arm tingling (7.5% vs. 13.5%, p < 0.001, NNT 17), numbness (8.1% vs. 31.1%, p < 0.001, NNT 5), lower rates of significant shoulder abduction deficits (5.7% vs. 9%, p < 0.001, NNT 31), and arm volume differences (7.5% vs. 14.3%, p < 0.001, NNT 15) (level 2 [mid-level] evidence) (J Surg Oncol 2010 Aug 1;102(2):111). In a subgroup of 749 women with negative SLNs who completed symptom questionnaires, SLN resection alone was associated with reduced rates of bothersome arm symptoms, restricted work and social activity, and impaired quality of life (level 2 [mid-level] evidence) (J Clin Oncol 2010 Sep 1;28(25):3929).
For more information, see the Breast cancer (female) topic in DynaMed. This article was featured in DynaMed Weekly Update 43.
_________________________________________________
Preliminary Data Suggest that Low-Dose Computed Tomography (CT) Screening May Reduce Lung Cancer Mortality and All-Cause Mortality
Guidelines from USPSTF and Canadian Task Force for Preventive Health Care do not recommend lung cancer screening, even in persons who smoke. The National Lung Screening Trial (NLST) evaluated the efficacy of low-dose CT for lung cancer screening in 53,454 current and former smokers aged 55-74 years with a history of > 30 pack-years. Patients were randomized to low-dose CT vs. chest radiography offered at baseline and at 2 annual follow-up visits. Findings were announced in a preliminary, unpublished report from the trial's Data and Safety Monitoring Board. Initial screens were positive for abnormality in 27.3% vs. 9.2%. Lung cancers were detected in 649 patients by screening CT and in 279 patients by chest x-ray. Lung cancer mortality rates were 245.7 per 100,000 patient-years vs. 308.3 per 100,000 patient-years (p < 0.05). The DynaMed Editors estimate the overall lung cancer mortality at 1.345% vs. 1.697% (NNS 284, 95% CI 179-706). All-cause mortality rates were 1,117.2 per 100,000 patient-years with CT screening vs. 1,200 per 100,000 patient-years with radiography (p < 0.05), with estimated all-cause mortality 7.106% vs. 7.662% (NNS 178, 95% CI 100-926) (level 2 [mid-level] evidence). The trial was terminated early at an interim analysis in October 2010 due to a significant difference between groups that met a pre-specified stopping rule (NLST Data and Safety Monitoring Board statement 2010 Oct 28, National Cancer Institute Press Release 2010 Nov 4).
For more information, see the Lung cancer topic in DynaMed. This article was featured in DynaMed Weekly Update 45.
