Acrylamide in popular foods does not cause breast cancer
http://www.100md.com
《英国医生杂志》
Researchers have been looking for a link between acrylamide and human cancer since it was found in many of our favourite foods, including chips, bread, crisps, crackers, pancakes, and coffee. They haven't found one yet, and a recent cohort study that found no association between acrylamide intake and risk of breast cancer provides further reassurance. The study included 43 404 middle aged Swedish women who were followed up for 11 years after answering detailed questions about their diet. Overall they had a low dietary intake of acrylamide, most of which came from coffee (54% of intake) and chips (12%). After adjustment for a wide range of confounders, the women who consumed the most acrylamide were no more likely to get breast cancer than the women who consumed the least (figure) (relative risk for top fifth compared with bottom fifth 1.19, 95% CI 0.91 to 1.55)
Credit: JAMA
JAMA 2005;293: 1326-7
Low carbohydrate diet improves glycaemic control in obese people with type 2 diabetes
In a small metabolic study, 10 men and women with obesity and type 2 diabetes lost weight and improved their glycaemic control by going on a strict Atkins diet for 14 days. The participants were admitted to a hospital metabolic ward, where researchers then drastically reduced their carbohydrate intake to 21 g/day but allowed them free access to high protein and high fat foods such as beef burgers, turkey burgers, fish, butter, eggs, and cream. They were also allowed steamed or raw vegetables.
After 14 days the participants had each lost about 2 kg in weight, most of which was fat, and accounted for by a reduction in calorie intake of about a third. Their mean HbA1c fell from 7.3% to 6.8% (P = 0.006), and mean fasting plasma concentrations of glucose fell from 7.5 mmol/l to 6.3 mmol/l (P = 0.025). Insulin sensitivity improved by about 75%, and mean serum concentrations of triglyceride and cholesterol decreased by 35% and 10%, respectively.
The authors concede that their study is limited by its short duration, small size, and lack of a control group. But as a metabolic study of the Atkins diet in people with diabetes, it worked. The next step is larger, longer clinical trials to find out if the benefits last.
Annals of Internal Medicine 2005;142: 403-11
Inappropriate prescribing to older people is widespread across Europe
Nearly a fifth of older people who receive home care in Europe are taking at least one inappropriate drug, a survey in eight countries (2707 people) has found. Older people in the Czech Republic were most at risk; 41% of them were taking one or more drugs contraindicated in older people, compared with 15.8% in the other countries combined (figure). The 10 most problematic drugs included amitriptyline and diazepam in all countries; pentoxifylline, high dose digoxin, and chlordiazepoxide in the Czech Republic; ticlopidine and amiodarone in Italy; and unopposed oestrogens in Iceland.
Credit: JAMA
The authors say comparative data like theirs is hard to come by in Europe, not least because there's no official agreement between countries about which drugs are dangerous for older people. Patients who were poor, depressed, taking multiple drugs, or taking any anxiolytic were most likely to be inappropriately treated (relative risk for poverty 1.96, 95% CI 1.58 to 2.36).
The variation between countries should be no surprise given differences in drug regulation, prescribing habits, health, wealth, and expectations of the population. Ongoing efforts to harmonise drug policy across Europe should help, at least with the first two factors.
JAMA 2005;293: 1348-58
Hospital leaders don't like mandatory reporting of medical errors
By 2003, 21 American states had some kind of mandatory reporting system for medical errors, but a cross sectional survey shows that reporting is generally unpopular with hospital leaders. Public disclosure of details such as the hospital's name was particularly unpopular: 69% of responders said it would stop healthcare workers from reporting mistakes, 79% thought it encouraged lawsuits, and 73% thought public disclosure did not improve patients' safety.
The survey, which had a 63% response rate, included the chief executive or chief operating officer from 203 hospitals across six states: two had mandatory reporting systems that remained confidential, two allowed public access to reports, and two states did not have mandatory reporting at the time of the study. Four fifths of responders felt that the hospital and the professionals involved in an incident should be protected from public disclosure. However, responders already working in states with open systems were significantly happier for their hospital to be named (22% v 4-6%) (figure).
Credit: JAMA
Mandatory reporting is unlikely to go away, however unpopular it is. But the authors say the kind of pessimism expressed in their survey could threaten long term goals, including safer hospitals for patients and a healthier culture of transparency for doctors.
JAMA 2005:293: 1359-66
Paired test is best for diagnosing catheter related infection
Of the eight (or more) ways to diagnose a catheter related infection of the bloodstream, paired quantitative blood culture performed best in a recent meta-analysis, with a sensitivity of 0.87 (95% CI 0.83 to 0.91) and specificity of 0.98 (0.97 to 0.99). With this method, blood cultures are taken simultaneously through the suspected central catheter and percutaneously from a distant site. The test is positive if both samples grow bacteria and if the catheter sample grows at least three times as many bacteria as the peripheral blood sample.
The other seven tests included in the meta-analysis were less accurate, but most were still clinically useful. All but one had both a sensitivity and specificity above 0.75, and a negative predictive value of over 99%. Simple culture of a blood sample taken through the catheter (without quantifying the growth) was the second most accurate test overall, and simple culture of a catheter segment was the least accurate, with a specificity of 0.72 (0.66 to 0.78), and a sensitivity of 0.90 (0.83 to 0.97).
Annals of Internal Medicine 2005;142: 451-66
Vitamin E does not prevent heart disease or cancer
In animal experiments and observational research, vitamin E, a potent antioxidant, looks promising as a prophylactic against cardiovascular disease and cancer, but short term clinical trials have been disappointing. Hoping for better results from longer term treatment, researchers extended an existing placebo controlled trial for an extra four years. Still they found no reduction in cancers, cancer deaths, or cardiovascular events in the treated group.
Nearly 5000 patients with diabetes or vascular disease took part in the extension, 3994 of whom agreed to carry on taking their vitamin E or placebo. After a median follow-up of seven years, 552 cancers had occurred (11.6%) in the vitamin E group and 586 (12.3%) in the placebo group (relative risk 0.94, 95% CI 0.84 to 1.06). Also, 1022 (21.5%) major cardiovascular events occurred in the vitamin E group compared with 985 (20.6%) in the placebo group (1.04, 0.96 to 1.14).
The only positive finding was an increased risk of heart failure among patients who took vitamin E (1.13, 1.01 to 1.26), a result the authors describe as disturbing and probably real. A linked editorial (pp 1387-90) says this emphatic new analysis, along with many other stubbornly negative trials, signals the end for vitamin E as a general prophylactic against cancer and heart disease.
JAMA 2005;293: 1338-47
Borderline risk factors cause only a minority of heart attacks and coronary deaths in USA
Over 40% of women and 25% of men aged 35 to 74 have at least one borderline risk factor for coronary heart disease, according to an epidemiological study from the United States. But do borderline risk factors matter? The study's authors estimate that people with multiple borderline risk factors account for only about 8% of heart attacks or deaths from heart disease in the United States. Over 90% of coronary deaths and heart attacks occur in people with at least one true risk factor, such as high blood pressure, glucose intolerance, smoking, or a high serum concentration of LDL cholesterol (figure).
Credit: ANNALS OF INTERNAL MEDICINE
Borderline risk factors, defined as factors that are less than ideal but not bad enough to treat, seem more important in men than in women. But in this study, even men with a handful of borderline risk factors had an estimated risk of coronary death or a heart attack of less than 10% over 10 years.
These data, which were modelled from the Framingham study and a national US survey, indicate that borderline risk factors on their own contribute little to the overall burden of heart disease in the United States. Public health resources should be targeted at the millions of Americans with serious modifiable risks, such as smoking, instead.
Annals of Internal Medicine 2005;142: 393-402
Rabies virus from organ donor kills four transplant patients
In May last year four patients from Texas died of a mystery encephalitis soon after receiving a transplanted organ from the same donor. Two of them had received a kidney, one a liver, and one a vascular graft. All were dead less than eight weeks later. After initial investigations drew a blank, the Centers for Disease Control and Prevention was called in to help.
Investigators inoculated baby mice with samples from the patients, waited for them to die (which they quickly did), and then examined tissue from the central nervous system under an electron microscope. Rhabdovirus particles were clearly visible: the patients had died of rabies.
The donor, an apparently healthy young man, had died suddenly of a subarachnoid haemorrhage. Routine infection screens all had negative results and his organs were duly harvested and transplanted into five people (the lung recipient died of surgical complications before he had the chance to develop rabies). Only later did friends remember that the donor had recently been bitten by a bat, the commonest source of rabies infection in the United States.
New England Journal of Medicine 2005;352: 1103-11
Inflammatory marker and colorectal cancer in women are not linked
Researchers investigating the possible link between chronic inflammation and colon cancer found no association between the inflammatory marker C reactive protein and colon cancer in women. They tracked a cohort of 27 913 health professionals for up to 10 years after measuring their serum concentrations of C reactive protein at baseline: 169 of the women developed colorectal cancer, but those with the highest concentrations were no more likely to get cancer than those with the lowest (hazard ratio 0.66, 95% CI 0.43 to 1.03).
The authors are fairly confident of their results, even though they had blood samples from only 71% of the original recruits, and even though their findings disagree with the only other large study of C reactive protein and colorectal cancer. It's hard to explain why one study found an association while the other one didn't, although there are plenty of differences in the methods. The only thing we can say for certain is that we still have a lot to learn about the relation between cancer and chronic inflammation.
Credit: JAMA
JAMA 2005;293: 1326-7
Low carbohydrate diet improves glycaemic control in obese people with type 2 diabetes
In a small metabolic study, 10 men and women with obesity and type 2 diabetes lost weight and improved their glycaemic control by going on a strict Atkins diet for 14 days. The participants were admitted to a hospital metabolic ward, where researchers then drastically reduced their carbohydrate intake to 21 g/day but allowed them free access to high protein and high fat foods such as beef burgers, turkey burgers, fish, butter, eggs, and cream. They were also allowed steamed or raw vegetables.
After 14 days the participants had each lost about 2 kg in weight, most of which was fat, and accounted for by a reduction in calorie intake of about a third. Their mean HbA1c fell from 7.3% to 6.8% (P = 0.006), and mean fasting plasma concentrations of glucose fell from 7.5 mmol/l to 6.3 mmol/l (P = 0.025). Insulin sensitivity improved by about 75%, and mean serum concentrations of triglyceride and cholesterol decreased by 35% and 10%, respectively.
The authors concede that their study is limited by its short duration, small size, and lack of a control group. But as a metabolic study of the Atkins diet in people with diabetes, it worked. The next step is larger, longer clinical trials to find out if the benefits last.
Annals of Internal Medicine 2005;142: 403-11
Inappropriate prescribing to older people is widespread across Europe
Nearly a fifth of older people who receive home care in Europe are taking at least one inappropriate drug, a survey in eight countries (2707 people) has found. Older people in the Czech Republic were most at risk; 41% of them were taking one or more drugs contraindicated in older people, compared with 15.8% in the other countries combined (figure). The 10 most problematic drugs included amitriptyline and diazepam in all countries; pentoxifylline, high dose digoxin, and chlordiazepoxide in the Czech Republic; ticlopidine and amiodarone in Italy; and unopposed oestrogens in Iceland.
Credit: JAMA
The authors say comparative data like theirs is hard to come by in Europe, not least because there's no official agreement between countries about which drugs are dangerous for older people. Patients who were poor, depressed, taking multiple drugs, or taking any anxiolytic were most likely to be inappropriately treated (relative risk for poverty 1.96, 95% CI 1.58 to 2.36).
The variation between countries should be no surprise given differences in drug regulation, prescribing habits, health, wealth, and expectations of the population. Ongoing efforts to harmonise drug policy across Europe should help, at least with the first two factors.
JAMA 2005;293: 1348-58
Hospital leaders don't like mandatory reporting of medical errors
By 2003, 21 American states had some kind of mandatory reporting system for medical errors, but a cross sectional survey shows that reporting is generally unpopular with hospital leaders. Public disclosure of details such as the hospital's name was particularly unpopular: 69% of responders said it would stop healthcare workers from reporting mistakes, 79% thought it encouraged lawsuits, and 73% thought public disclosure did not improve patients' safety.
The survey, which had a 63% response rate, included the chief executive or chief operating officer from 203 hospitals across six states: two had mandatory reporting systems that remained confidential, two allowed public access to reports, and two states did not have mandatory reporting at the time of the study. Four fifths of responders felt that the hospital and the professionals involved in an incident should be protected from public disclosure. However, responders already working in states with open systems were significantly happier for their hospital to be named (22% v 4-6%) (figure).
Credit: JAMA
Mandatory reporting is unlikely to go away, however unpopular it is. But the authors say the kind of pessimism expressed in their survey could threaten long term goals, including safer hospitals for patients and a healthier culture of transparency for doctors.
JAMA 2005:293: 1359-66
Paired test is best for diagnosing catheter related infection
Of the eight (or more) ways to diagnose a catheter related infection of the bloodstream, paired quantitative blood culture performed best in a recent meta-analysis, with a sensitivity of 0.87 (95% CI 0.83 to 0.91) and specificity of 0.98 (0.97 to 0.99). With this method, blood cultures are taken simultaneously through the suspected central catheter and percutaneously from a distant site. The test is positive if both samples grow bacteria and if the catheter sample grows at least three times as many bacteria as the peripheral blood sample.
The other seven tests included in the meta-analysis were less accurate, but most were still clinically useful. All but one had both a sensitivity and specificity above 0.75, and a negative predictive value of over 99%. Simple culture of a blood sample taken through the catheter (without quantifying the growth) was the second most accurate test overall, and simple culture of a catheter segment was the least accurate, with a specificity of 0.72 (0.66 to 0.78), and a sensitivity of 0.90 (0.83 to 0.97).
Annals of Internal Medicine 2005;142: 451-66
Vitamin E does not prevent heart disease or cancer
In animal experiments and observational research, vitamin E, a potent antioxidant, looks promising as a prophylactic against cardiovascular disease and cancer, but short term clinical trials have been disappointing. Hoping for better results from longer term treatment, researchers extended an existing placebo controlled trial for an extra four years. Still they found no reduction in cancers, cancer deaths, or cardiovascular events in the treated group.
Nearly 5000 patients with diabetes or vascular disease took part in the extension, 3994 of whom agreed to carry on taking their vitamin E or placebo. After a median follow-up of seven years, 552 cancers had occurred (11.6%) in the vitamin E group and 586 (12.3%) in the placebo group (relative risk 0.94, 95% CI 0.84 to 1.06). Also, 1022 (21.5%) major cardiovascular events occurred in the vitamin E group compared with 985 (20.6%) in the placebo group (1.04, 0.96 to 1.14).
The only positive finding was an increased risk of heart failure among patients who took vitamin E (1.13, 1.01 to 1.26), a result the authors describe as disturbing and probably real. A linked editorial (pp 1387-90) says this emphatic new analysis, along with many other stubbornly negative trials, signals the end for vitamin E as a general prophylactic against cancer and heart disease.
JAMA 2005;293: 1338-47
Borderline risk factors cause only a minority of heart attacks and coronary deaths in USA
Over 40% of women and 25% of men aged 35 to 74 have at least one borderline risk factor for coronary heart disease, according to an epidemiological study from the United States. But do borderline risk factors matter? The study's authors estimate that people with multiple borderline risk factors account for only about 8% of heart attacks or deaths from heart disease in the United States. Over 90% of coronary deaths and heart attacks occur in people with at least one true risk factor, such as high blood pressure, glucose intolerance, smoking, or a high serum concentration of LDL cholesterol (figure).
Credit: ANNALS OF INTERNAL MEDICINE
Borderline risk factors, defined as factors that are less than ideal but not bad enough to treat, seem more important in men than in women. But in this study, even men with a handful of borderline risk factors had an estimated risk of coronary death or a heart attack of less than 10% over 10 years.
These data, which were modelled from the Framingham study and a national US survey, indicate that borderline risk factors on their own contribute little to the overall burden of heart disease in the United States. Public health resources should be targeted at the millions of Americans with serious modifiable risks, such as smoking, instead.
Annals of Internal Medicine 2005;142: 393-402
Rabies virus from organ donor kills four transplant patients
In May last year four patients from Texas died of a mystery encephalitis soon after receiving a transplanted organ from the same donor. Two of them had received a kidney, one a liver, and one a vascular graft. All were dead less than eight weeks later. After initial investigations drew a blank, the Centers for Disease Control and Prevention was called in to help.
Investigators inoculated baby mice with samples from the patients, waited for them to die (which they quickly did), and then examined tissue from the central nervous system under an electron microscope. Rhabdovirus particles were clearly visible: the patients had died of rabies.
The donor, an apparently healthy young man, had died suddenly of a subarachnoid haemorrhage. Routine infection screens all had negative results and his organs were duly harvested and transplanted into five people (the lung recipient died of surgical complications before he had the chance to develop rabies). Only later did friends remember that the donor had recently been bitten by a bat, the commonest source of rabies infection in the United States.
New England Journal of Medicine 2005;352: 1103-11
Inflammatory marker and colorectal cancer in women are not linked
Researchers investigating the possible link between chronic inflammation and colon cancer found no association between the inflammatory marker C reactive protein and colon cancer in women. They tracked a cohort of 27 913 health professionals for up to 10 years after measuring their serum concentrations of C reactive protein at baseline: 169 of the women developed colorectal cancer, but those with the highest concentrations were no more likely to get cancer than those with the lowest (hazard ratio 0.66, 95% CI 0.43 to 1.03).
The authors are fairly confident of their results, even though they had blood samples from only 71% of the original recruits, and even though their findings disagree with the only other large study of C reactive protein and colorectal cancer. It's hard to explain why one study found an association while the other one didn't, although there are plenty of differences in the methods. The only thing we can say for certain is that we still have a lot to learn about the relation between cancer and chronic inflammation.