What's new in the other general journals
http://www.100md.com
《英国医生杂志》
Taking SSRIs or SNRIs in late pregnancy can harm newborns
Women who take selective serotonin or noradrenaline reuptake inhibitors (SSRIs or SNRIs) in late pregnancy risk giving their baby a neonatal syndrome characterised by jitteriness, increased muscle tone, feeding difficulties, and sometimes respiratory distress. But the symptoms are often mild, self limiting, and easily managed with supportive measures, say researchers. After a careful look for all reports of the syndrome, they found 13 published case reports, 9 cohort studies, 57 cases reported to the US Food and Drug Administration's adverse events database, and 74 cases reported to the World Health Organization.
Using data from the cohort studies, they estimated that babies whose mothers took these antidepressants during the last trimester were three times more likely to develop neonatal symptoms (risk ratio 3.0, 95% CI 2.0 to 4.4), 2.6 (1.4 to 4.7) times more likely to need nursing in a special care unit, and 2.3 (1.6 to 3.2) times more likely to have breathing problems than babies who were not exposed at all, or were exposed earlier in the pregnancy. Reported symptoms were fairly consistent (figure) and lasted no more than two weeks.
We still don't know the risks associated with individual drugs and we still don't know what causes the syndrome, or how to avoid it. The US Food and Drugs Administration recently added a warning label to all SSRIs and SNRIs describing the symptoms and suggesting that women consider tapering the dose in late pregnancy.
Credit: JAMA
JAMA 2005;293: 2372-83
New definition of fasting glucose is poor predictor of cardiovascular disease
In 2003, the American Diabetes Association reduced the lower limit of normal for fasting serum concentration of glucose. Men and women with fasting glucose between 5.6 mmol/l and 6.9 mmol/l are now deemed to have have impaired fasting glucose and a higher risk of diabetes. The new definition triples the number of men and women in the United States with impaired fasting glucose, but is the label a risk factor for cardiovascular disease? Not in women, according to a recent study.
Researchers followed a cohort of 2763 postmenopausal women for 6.8 years. The women already had established coronary heart disease, but having a serum concentration of glucose between 5.5 and 6.9 mmol/l did not increase their risk of further cardiovascular events, including non-fatal heart attack, cardiovascular death, hospital admission for congestive heart failure, stroke, or transient ischaemic attack (hazard ratio compared with normoglycaemic women, 1.09, 95% CI 0.90 to 1.34). Women with impaired fasting glucose according to the previous definition (6.1-6.9 mmol/l), were about 40% more likely than normoglycaemic women to have a non-fatal heart attack or die from cardiovascular disease (1.37, 1.08 to 1.74). Predictably, women with frank diabetes had the highest risks of all—about 75% higher than normoglycaemic women for all outcomes.
Annals of Internal Medicine 2005;142: 813-20
Timing of first cereal foods is linked to coeliac disease in vulnerable children
In a study from Denver, Colorado, infants introduced to wheat, barley, or rye before the age of 3 months were more likely to develop early coeliac disease than children who were first given cereals between 4 and 6 months (hazard ratio 5.17, 95% CI 1.44 to 18.57). Oddly, the risk went up again in children introduced to cereals later than 7 months (1.87, 0.97 to 3.60) (figure).
Credit: JAMA
These findings come from a cohort of 1560 children recruited for a study of the course of diabetes and coeliac disease. All of them had a higher than average risk of one or the other condition, defined by family history or genetic markers. Overall, 51 children developed coeliac disease autoantibodies, a sensitive and specific test for presymptomatic coeliac disease, and 25 had the disease confirmed by small bowel biopsy. The authors found a clear link between timing of first gluten and coeliac disease that was strongest in the subgroup of children with positive small bowel biopsies.
Although this is the first prospective study to look at this issue, it's hard to say what the results mean for most infants. The reported associations are confined to children who already have a high risk of coeliac disease, and the analysis is seriously limited in several places by small numbers. Longer follow-up might clarify the situation, but for now all we can say is that current recommendations to introduce cereals between the age of 4 and 6 months seem about right.
JAMA 2005;293: 2343-51
Fewer tobacco products appear in American movies after ban
In 1998, the tobacco industry signed an agreement with the US attorney general banning advertising by product placement movies. To find out if the agreement worked, two researchers watched 400 US box office movies made before the ban and 400 made afterwards, counting the appearance of branded tobacco products, such as cigarettes. They found a clear downward trend in blatant tobacco product placement during the five years following the agreement (figure), and a straightforward before and after comparison of all movies showed a significant drop in the number of times branded tobacco products appeared (83/400 (21%) v 42/400 (11%); odds ratio 0.45, 95% CI 0.29 to 0.68).
Credit: JAMA
The agreement worked best in movies rated R ("restricted" for people 17 and under) by the Motion Picture Association America (29.8% v 13.3%, P < 0.001), presumably because these movies were more likely to advertise tobacco products in the first place than those aimed at younger audiences.
More than two thirds of the tobacco products spotted by researchers in movies made after 1998 were manufactured by companies that were signed up to the ban, including Philip Morris, R J Reynolds, and Brown and Williamson. Marlboro, made by Philip Morris, appeared more often than any other brand.
JAMA 2005;293: 2341-2
Older treatment options are still cost effective for patients with chronic hepatitis B
In patients with chronic hepatitis B but no cirrhosis, doing nothing is the cheapest option, according to a cost effectiveness study, but it has no effect on the disease. Compared with doing nothing, treatment with interferon costs an extra $6337 for each life year gained, and a strategy that starts with lamivudine and switches to adefovir in the event of viral resistance costs a further $8446 for each life year gained. The authors say that both treatment options are affordable, but that older and therefore cheaper interferon might be more attractive to countries with cash limited health services.
In this analysis, neither lamivudine, which is well known for causing viral resistance, nor adefovir, which is well known for being expensive, was cost effective when used on its own.
These authors used data from a systematic review to construct a model of the costs and effects of treating 40 patients with chronic hepatitis B and raised serum concentrations of liver transaminases, but no cirrhosis. They computed estimates for two distinct groups of patient—those with and those without hepatitis B e antigen in their blood. Patients without circulating e antigen are harder to treat because their viral load tends to bounce back up when treatment stops. All the treatment strategies were more expensive and less effective in this group, pushing the cost of the adefovir salvage strategy up to $16 593 per life year gained, compared with interferon.
Annals of Internal Medicine 2005;142: 821-31
Asthma is a new risk factor for pneumococcal infection
Chronic obstructive pulmonary disease is a well known risk factor for serious pneumococcal infections, as are other common chronic diseases such as diabetes and sickle cell anaemia. Researchers from Tennessee want to add asthma to the list, after their case control study showed that asthma more than doubles the risk of invasive pneumococcal infection in both adults and children (adjusted odds ratio 2.4, 95% CI 1.9 to 3.1). The association held firm after adjustment for other high risk chronic diseases, age, ethnicity, and sex. But the researchers couldn't fully account for people's smoking habits, because the databases they mined for their study—Tennessee's Medicaid database and a surveillance programme for pneumococcal infection—did not record smoking reliably. Even so, children don't smoke (or at least not much), so the authors are confident the link between asthma and infection is not confounded entirely by smoking.
Finding risk factors for pneumococcal diseases is important because effective vaccines can be given to people who have them. The latest guidelines on vaccination from the US Centers for Disease Control and Prevention explicitly exclude people with asthma. The researchers calculate that including these people in vaccination programmes would prevent up to 30 severe pneumococcal infections for every 10 000 people over the next 10 years.
New England Journal of Medicine 2005;352: 2082-90
Rifaximin prevents travellers' diarrhoea where E coli is the likeliest cause
The rifamycin antibacterial rifaximin is a non-absorbable antibiotic. When taken by mouth, 99.6% of the dose stays in the gut, making it an effective treatment for some bacterial forms of travellers' diarrhoea. A study of students visiting Mexico from the US suggests rifaximin also works as prophylaxis. The students took a placebo or rifaximin once, twice, or three times a day for 14 days during their visit to Guadalajara and recorded any symptoms and signs of travellers' diarrhoea in a diary. During treatment 54% (29/54) of the students taking a placebo had diarrhoea, usually caused by enterotoxigenic Escherichia coli, compared with only 15% (23/156) of the students taking rifaximin (rate ratio 0.27, 95% CI 0.17 to 0.43). All three dose regimens worked equally well (figure), and seemed to prevent even mild diarrhoea and other symptoms such as abdominal cramps and flatulence. There were few side effects in any group.
Credit: MEDICINE INTERNAL OF ANNALS
When the authors examined the students' normal gut flora they found few differences between those taking the antibiotic and those taking placebo, so rifaximin seems to prevent travellers' diarrhoea without killing resident harmless coliforms. The authors aren't sure how, but it's possible that rifaximin stays in the small bowel, killing pathogens, without affecting gut flora further down.
This study, which was double blind, was done in Mexico, where enterotoxigenic E coli is the commonest cause of travellers' diarrhoea. The authors plan further studies in places such as Asia, where other pathogens including Shigella, Salmonella, and Campylobacter are more prevalent.
Annals of Internal Medicine 2005;142: 805-12
Screening for colorectal cancer: women need colonoscopy
Consensus still hasn't been reached about how best to screen for colorectal cancer in people without symptoms, especially women. Some experts say that since women have a lower risk of cancer, flexible sigmoidoscopy should be enough to start with. But a recent study from the US indicates that this approach would miss two thirds of important lesions. Researchers screened 1463 asymptomatic women with colonoscopy and found 72 malignant or potentially malignant lesions, 47 of which would have been missed by a simple flexible sigmoidoscopy. More than four fifths of the cohort had no family history. Sigmoidoscopy for them would have picked up only 35% (19/54) of these lesions, a considerably smaller proportion than the 66% reported for matched men in a previous but related study (P < 0.001).
The authors conclude that colonoscopy should be the screening method of choice for asymptomatic women with an average risk of colorectal cancer.
Women who take selective serotonin or noradrenaline reuptake inhibitors (SSRIs or SNRIs) in late pregnancy risk giving their baby a neonatal syndrome characterised by jitteriness, increased muscle tone, feeding difficulties, and sometimes respiratory distress. But the symptoms are often mild, self limiting, and easily managed with supportive measures, say researchers. After a careful look for all reports of the syndrome, they found 13 published case reports, 9 cohort studies, 57 cases reported to the US Food and Drug Administration's adverse events database, and 74 cases reported to the World Health Organization.
Using data from the cohort studies, they estimated that babies whose mothers took these antidepressants during the last trimester were three times more likely to develop neonatal symptoms (risk ratio 3.0, 95% CI 2.0 to 4.4), 2.6 (1.4 to 4.7) times more likely to need nursing in a special care unit, and 2.3 (1.6 to 3.2) times more likely to have breathing problems than babies who were not exposed at all, or were exposed earlier in the pregnancy. Reported symptoms were fairly consistent (figure) and lasted no more than two weeks.
We still don't know the risks associated with individual drugs and we still don't know what causes the syndrome, or how to avoid it. The US Food and Drugs Administration recently added a warning label to all SSRIs and SNRIs describing the symptoms and suggesting that women consider tapering the dose in late pregnancy.
Credit: JAMA
JAMA 2005;293: 2372-83
New definition of fasting glucose is poor predictor of cardiovascular disease
In 2003, the American Diabetes Association reduced the lower limit of normal for fasting serum concentration of glucose. Men and women with fasting glucose between 5.6 mmol/l and 6.9 mmol/l are now deemed to have have impaired fasting glucose and a higher risk of diabetes. The new definition triples the number of men and women in the United States with impaired fasting glucose, but is the label a risk factor for cardiovascular disease? Not in women, according to a recent study.
Researchers followed a cohort of 2763 postmenopausal women for 6.8 years. The women already had established coronary heart disease, but having a serum concentration of glucose between 5.5 and 6.9 mmol/l did not increase their risk of further cardiovascular events, including non-fatal heart attack, cardiovascular death, hospital admission for congestive heart failure, stroke, or transient ischaemic attack (hazard ratio compared with normoglycaemic women, 1.09, 95% CI 0.90 to 1.34). Women with impaired fasting glucose according to the previous definition (6.1-6.9 mmol/l), were about 40% more likely than normoglycaemic women to have a non-fatal heart attack or die from cardiovascular disease (1.37, 1.08 to 1.74). Predictably, women with frank diabetes had the highest risks of all—about 75% higher than normoglycaemic women for all outcomes.
Annals of Internal Medicine 2005;142: 813-20
Timing of first cereal foods is linked to coeliac disease in vulnerable children
In a study from Denver, Colorado, infants introduced to wheat, barley, or rye before the age of 3 months were more likely to develop early coeliac disease than children who were first given cereals between 4 and 6 months (hazard ratio 5.17, 95% CI 1.44 to 18.57). Oddly, the risk went up again in children introduced to cereals later than 7 months (1.87, 0.97 to 3.60) (figure).
Credit: JAMA
These findings come from a cohort of 1560 children recruited for a study of the course of diabetes and coeliac disease. All of them had a higher than average risk of one or the other condition, defined by family history or genetic markers. Overall, 51 children developed coeliac disease autoantibodies, a sensitive and specific test for presymptomatic coeliac disease, and 25 had the disease confirmed by small bowel biopsy. The authors found a clear link between timing of first gluten and coeliac disease that was strongest in the subgroup of children with positive small bowel biopsies.
Although this is the first prospective study to look at this issue, it's hard to say what the results mean for most infants. The reported associations are confined to children who already have a high risk of coeliac disease, and the analysis is seriously limited in several places by small numbers. Longer follow-up might clarify the situation, but for now all we can say is that current recommendations to introduce cereals between the age of 4 and 6 months seem about right.
JAMA 2005;293: 2343-51
Fewer tobacco products appear in American movies after ban
In 1998, the tobacco industry signed an agreement with the US attorney general banning advertising by product placement movies. To find out if the agreement worked, two researchers watched 400 US box office movies made before the ban and 400 made afterwards, counting the appearance of branded tobacco products, such as cigarettes. They found a clear downward trend in blatant tobacco product placement during the five years following the agreement (figure), and a straightforward before and after comparison of all movies showed a significant drop in the number of times branded tobacco products appeared (83/400 (21%) v 42/400 (11%); odds ratio 0.45, 95% CI 0.29 to 0.68).
Credit: JAMA
The agreement worked best in movies rated R ("restricted" for people 17 and under) by the Motion Picture Association America (29.8% v 13.3%, P < 0.001), presumably because these movies were more likely to advertise tobacco products in the first place than those aimed at younger audiences.
More than two thirds of the tobacco products spotted by researchers in movies made after 1998 were manufactured by companies that were signed up to the ban, including Philip Morris, R J Reynolds, and Brown and Williamson. Marlboro, made by Philip Morris, appeared more often than any other brand.
JAMA 2005;293: 2341-2
Older treatment options are still cost effective for patients with chronic hepatitis B
In patients with chronic hepatitis B but no cirrhosis, doing nothing is the cheapest option, according to a cost effectiveness study, but it has no effect on the disease. Compared with doing nothing, treatment with interferon costs an extra $6337 for each life year gained, and a strategy that starts with lamivudine and switches to adefovir in the event of viral resistance costs a further $8446 for each life year gained. The authors say that both treatment options are affordable, but that older and therefore cheaper interferon might be more attractive to countries with cash limited health services.
In this analysis, neither lamivudine, which is well known for causing viral resistance, nor adefovir, which is well known for being expensive, was cost effective when used on its own.
These authors used data from a systematic review to construct a model of the costs and effects of treating 40 patients with chronic hepatitis B and raised serum concentrations of liver transaminases, but no cirrhosis. They computed estimates for two distinct groups of patient—those with and those without hepatitis B e antigen in their blood. Patients without circulating e antigen are harder to treat because their viral load tends to bounce back up when treatment stops. All the treatment strategies were more expensive and less effective in this group, pushing the cost of the adefovir salvage strategy up to $16 593 per life year gained, compared with interferon.
Annals of Internal Medicine 2005;142: 821-31
Asthma is a new risk factor for pneumococcal infection
Chronic obstructive pulmonary disease is a well known risk factor for serious pneumococcal infections, as are other common chronic diseases such as diabetes and sickle cell anaemia. Researchers from Tennessee want to add asthma to the list, after their case control study showed that asthma more than doubles the risk of invasive pneumococcal infection in both adults and children (adjusted odds ratio 2.4, 95% CI 1.9 to 3.1). The association held firm after adjustment for other high risk chronic diseases, age, ethnicity, and sex. But the researchers couldn't fully account for people's smoking habits, because the databases they mined for their study—Tennessee's Medicaid database and a surveillance programme for pneumococcal infection—did not record smoking reliably. Even so, children don't smoke (or at least not much), so the authors are confident the link between asthma and infection is not confounded entirely by smoking.
Finding risk factors for pneumococcal diseases is important because effective vaccines can be given to people who have them. The latest guidelines on vaccination from the US Centers for Disease Control and Prevention explicitly exclude people with asthma. The researchers calculate that including these people in vaccination programmes would prevent up to 30 severe pneumococcal infections for every 10 000 people over the next 10 years.
New England Journal of Medicine 2005;352: 2082-90
Rifaximin prevents travellers' diarrhoea where E coli is the likeliest cause
The rifamycin antibacterial rifaximin is a non-absorbable antibiotic. When taken by mouth, 99.6% of the dose stays in the gut, making it an effective treatment for some bacterial forms of travellers' diarrhoea. A study of students visiting Mexico from the US suggests rifaximin also works as prophylaxis. The students took a placebo or rifaximin once, twice, or three times a day for 14 days during their visit to Guadalajara and recorded any symptoms and signs of travellers' diarrhoea in a diary. During treatment 54% (29/54) of the students taking a placebo had diarrhoea, usually caused by enterotoxigenic Escherichia coli, compared with only 15% (23/156) of the students taking rifaximin (rate ratio 0.27, 95% CI 0.17 to 0.43). All three dose regimens worked equally well (figure), and seemed to prevent even mild diarrhoea and other symptoms such as abdominal cramps and flatulence. There were few side effects in any group.
Credit: MEDICINE INTERNAL OF ANNALS
When the authors examined the students' normal gut flora they found few differences between those taking the antibiotic and those taking placebo, so rifaximin seems to prevent travellers' diarrhoea without killing resident harmless coliforms. The authors aren't sure how, but it's possible that rifaximin stays in the small bowel, killing pathogens, without affecting gut flora further down.
This study, which was double blind, was done in Mexico, where enterotoxigenic E coli is the commonest cause of travellers' diarrhoea. The authors plan further studies in places such as Asia, where other pathogens including Shigella, Salmonella, and Campylobacter are more prevalent.
Annals of Internal Medicine 2005;142: 805-12
Screening for colorectal cancer: women need colonoscopy
Consensus still hasn't been reached about how best to screen for colorectal cancer in people without symptoms, especially women. Some experts say that since women have a lower risk of cancer, flexible sigmoidoscopy should be enough to start with. But a recent study from the US indicates that this approach would miss two thirds of important lesions. Researchers screened 1463 asymptomatic women with colonoscopy and found 72 malignant or potentially malignant lesions, 47 of which would have been missed by a simple flexible sigmoidoscopy. More than four fifths of the cohort had no family history. Sigmoidoscopy for them would have picked up only 35% (19/54) of these lesions, a considerably smaller proportion than the 66% reported for matched men in a previous but related study (P < 0.001).
The authors conclude that colonoscopy should be the screening method of choice for asymptomatic women with an average risk of colorectal cancer.