What's new in the other general journals
http://www.100md.com
《英国医生杂志》
Educating mothers improves nutrition and growth in infants
Malnutrition is probably responsible for half of all preventable infant deaths worldwide and causes anaemia and stunted growth in survivors. Educating poor communities about good nutrition can help, but local projects are often hard to sustain. Researchers from Peru got round the problem by recruiting government health centres to deliver practical, easy to understand, nutritional advice to new mothers in local shanty towns. They also introduced a system of accreditation to motivate staff.
A cluster randomised trial of the new approach found that it worked. At 18 months, infants from the six areas with a nutritionally accredited health centre or community hospital were better fed and were 1 cm taller, 200 g heavier, and three times less likely to have stunted growth than infants from six control areas (8/171 (5%) v 26/165 (16%); odds ratio 3.04, 95% CI 1.21 to 7.64; figure). Most infants in both groups did not get an adequate daily intake of iron and zinc: at 9 months, 99% of controls and 93% of intervention infants were not getting enough iron, and 87% of controls and 77% of intervention infants were not getting enough zinc.
Credit: LANCET
These results show that government run health facilities can be effective educators, making a real and potentially lasting difference to infants living nearby.
Lancet 2005 May 10; doi 10.1016/S0140-6736(05)66426-4
Targeted screening fails children with high serum concentrations of lead
Lead toxicity is still a big problem for American children. The US Centers for Disease Control and Prevention estimates that at least 2.2% of preschool children in the United States have serum concentrations of lead that exceed the defining threshold of 10 μg/dl. Despite the size of the problem, universal screening was abandoned in 1997 and replaced with targeted screening of vulnerable children, such as those living in poor or old housing. Even that is patchy and inadequate, according to a recent study from Michigan.
Researchers reviewed routine data from the state's Medicaid programme and found 3682 children who had been screened and found to have serum concentrations of lead above 10 μg/dl. About half (54%) had had a repeat test, the cornerstone of management and follow-up for these children. Of those who were tested again, 48% still had high serum concentrations of lead. Further analysis indicates that the neediest children were the most likely to miss out—children at high risk of lead exposure (relative risk 0.94, 95% CI 0.92 to 0.96), children from Hispanic or other cultural minority backgrounds (0.91, 0.87 to 0.94), and those living in urban areas (0.92, 0.89 to 96.0).
More than half of the children (59%) had had at least one encounter with the health services in the six months after their positive test, so why did they slip through the net? It's impossible to say from this observational study, but the authors suspect that communication problems between primary care doctors (responsible for repeat testing) and the local public health department (responsible for managing confirmed cases) could be partly to blame.
JAMA 2005;293: 2232-37
Higher dose vitamin D prevents first osteoporotic fracture
Vitamin D in the form of cholecalciferol reduces the risk of a first osteoporotic fracture in older people by about a quarter, but only when prescribed at a relatively high dose (700-800 IU/day), say researchers from the United States. A lower dose (400 IU/day), closer to current expert recommendations, failed to prevent hip or any other fractures in this meta-analysis of seven randomised trials. Five of the trials tested the high dose for at least a year and found that cholecalciferol 700-800 IU reduced the risk of hip fracture by 26% (relative risk 0.74, 95% CI 0.61 to 0.88) and of any non-vertebral fracture by 23% (0.77, 0.68 to 0.87) compared with placebo or calcium alone (figure)—translate to one less hip fracture for every 45 people treated for 24-60 months, or one less non-vertebral fracture for every 27 people treated for 12-60 months. In four of the five trials, participants taking vitamin D also took calcium.
Credit: JAMA
An estimated 1 in 3 women and 1 in 6 men will have a hip fracture by the time they reach 90, an increasing disease burden that already costs the US more than $7bn a year. These authors say their results are enough to justify general supplementation with higher dose vitamin D in older people. What they can't say is whether or not older people need to take calcium to make it work.
JAMA 2005;293: 2257-64
Extra heart sounds detected by computer are a poor marker for heart failure
A third or fourth heart sound, otherwise known as gallop rhythm, is a traditional sign of heart failure. To find out how it measures up as modern, computerised, diagnostic test, researchers looked for third or fourth heart sounds in 90 people who were also having a battery of sophisticated tests including cardiac catheterisation and transthoracic echocardiography. Using a computerised heart sound analyser, they found at least one extra heart sound in 41 (46%) of the patients.
Overall, these patients had worse left ventricular function (a higher left ventricular end diastolic pressure or a lower left ventricular ejection fraction) than patients without extra diastolic heart sounds (figure). But heart sounds detected by computer were too insensitive to be useful as a test to rule ourt heart failure. A third heart sound had a sensitivity of only 52% for detecting a low left ventricular ejection fraction (< 50%), and a sensitivity of only 41% for detecting a raised left ventricular end diastolic pressure (> 15 mm Hg). A fourth heart sound had sensitivities below 50% for both, meaning that patients without any extra diastolic heart sounds could still have substantial heart failure. Specificities for the two heart sounds were between 80% and 92%.
Credit: JAMA
JAMA 2005;293: 2238-44
Meta-analyses are top of the citation league table
For years, experts having been telling us that meta-analyses are the best kind of research evidence and that case reports are irrelevant. A study looking at the changing pattern of citations for different types of research shows that we believe them: meta-analyses are now more commonly cited in medical literature than any other type of study (P < 0.001), closely followed by randomised trials. Case reports are hardly cited at all.
Using the Science Citation Index, the authors tracked citation patterns for two groups of studies (2646 in total), selected because the study design was in the title. Studies in one group were published in 1991, studies in the other group were published 10 years later. Citations were higher across the board between 1991 and 2001, probably because of an increase in the number of studies published. Over the same period, meta-analyses made substantial gains over their closest rivals, randomised trials, which in turn made substantial gains over epidemiological studies. Economic and decision analyses have lagged behind randomised trials in the citation stakes and are now cited about as often as narrative reviews. Case reports are very firmly at the bottom of the hierarchy for both periods.
JAMA 2005;293: 2362-6
Osteogenic cells found in peripheral blood
Scientists have found large numbers of previously elusive osteoblastic precursor cells in the peripheral circulation, adding a new dimension to bone growth in humans. The cells, which are thought to mature in to osteoblasts, have previously been seen only in tiny numbers, probably because the technique used to look for them required that the cells stick to plastic, and now it seems that they don't.
Using a new approach, the scientists looked for bone related proteins on the surface of circulating cells instead and found that 1-2% of mononuclear cells in 11 adult male volunteers were positive for osteocalcin. The cells expressed osteogenic genes, formed bone when transplanted in to mice, and mineralised nodules in an in vitro experiment. They were also found in much greater numbers in 11 boys having an adolescent growth spurt and three adult volunteers with recent fractures.
Taken together, all this points to a new circulatory component to bone formation that might have therapeutic potential—for example, in people with poorly healing fractures, say the authors.
New England Journal of Medicine 2005;352: 1959-66
Radical surgery reduces mortality in men with early prostate cancer
Three years ago, a trial reported that men with early, localised prostate cancer did better after a radical prostatectomy than with "watchful waiting." Over about six years of follow-up, radical prostatectomy reduced the risk of death due to prostate cancer by 50% and the risk of distant metastasis by 37%. An extended follow-up (mean 8 years) now shows that radical prostatectomy reduces all cause mortality too. In the latest analysis, 83/347 men who had surgery and 106/348 men who had watchful waiting died from any cause (relative risk 0.74, 95% CI 0.56 to 0.99; P = 0.04). Men who had surgery were less likely to die of prostate cancer (8.6% v 14.4%; P = 0.01) or develop metastases (19.2% v 44.3%) than men who did not have surgery, and the difference for both outcomes had increased (figure).
Credit: NEW ENGLAND JOURNAL OF MEDICINE
Although these results seem clear, the absolute survival gains from surgery are only moderate. The authors say that watchful waiting could become a better option if combined with regular tests for prostate specific antigen. Men could then be offered radical surgery if their test results showed a worsening trend.
New England Journal of Medicine 2005;352: 1977-84
Laparoscopic surgery for colorectal cancer looks safe in the short term
Laparoscopic surgery for colorectal cancer is as safe and effective as open surgery, at least in the first three months after the procedure, according to initial findings from a randomised trial in 794 patients. In an intention to treat analysis, which means analysing patients according to the treatment they were allocated and not necessarily the treatment they received, the groups did not differ in the rate of intraoperative complications (10% in the open group v 10% in the laparoscopy group), postoperative complications (32% v 33%), transfusion requirements (15% v 20%), mortality in hospital (5% v 4%), or quality of life up to three months after surgery. The two groups of patients were also equally likely to have resection margins that were clear of cancer, with the possible exception of patients having an anterior resection for rectal cancer, in whom margins were slightly more likely to be clear after an open operation (94% v 88%).
Nearly a third of patients allocated to laparoscopic surgery had to convert to open surgery in this trial (143/488, 29%). These patients were significantly more likely to have intraoperative and postoperative complications and to need a transfusion.
These results, although important, aren't complete without the longer term outcomes, including recurrence rates and mortality, that will be reported later. In the meantime, more work needs to be done to identify patients at high risk of conversion.
Malnutrition is probably responsible for half of all preventable infant deaths worldwide and causes anaemia and stunted growth in survivors. Educating poor communities about good nutrition can help, but local projects are often hard to sustain. Researchers from Peru got round the problem by recruiting government health centres to deliver practical, easy to understand, nutritional advice to new mothers in local shanty towns. They also introduced a system of accreditation to motivate staff.
A cluster randomised trial of the new approach found that it worked. At 18 months, infants from the six areas with a nutritionally accredited health centre or community hospital were better fed and were 1 cm taller, 200 g heavier, and three times less likely to have stunted growth than infants from six control areas (8/171 (5%) v 26/165 (16%); odds ratio 3.04, 95% CI 1.21 to 7.64; figure). Most infants in both groups did not get an adequate daily intake of iron and zinc: at 9 months, 99% of controls and 93% of intervention infants were not getting enough iron, and 87% of controls and 77% of intervention infants were not getting enough zinc.
Credit: LANCET
These results show that government run health facilities can be effective educators, making a real and potentially lasting difference to infants living nearby.
Lancet 2005 May 10; doi 10.1016/S0140-6736(05)66426-4
Targeted screening fails children with high serum concentrations of lead
Lead toxicity is still a big problem for American children. The US Centers for Disease Control and Prevention estimates that at least 2.2% of preschool children in the United States have serum concentrations of lead that exceed the defining threshold of 10 μg/dl. Despite the size of the problem, universal screening was abandoned in 1997 and replaced with targeted screening of vulnerable children, such as those living in poor or old housing. Even that is patchy and inadequate, according to a recent study from Michigan.
Researchers reviewed routine data from the state's Medicaid programme and found 3682 children who had been screened and found to have serum concentrations of lead above 10 μg/dl. About half (54%) had had a repeat test, the cornerstone of management and follow-up for these children. Of those who were tested again, 48% still had high serum concentrations of lead. Further analysis indicates that the neediest children were the most likely to miss out—children at high risk of lead exposure (relative risk 0.94, 95% CI 0.92 to 0.96), children from Hispanic or other cultural minority backgrounds (0.91, 0.87 to 0.94), and those living in urban areas (0.92, 0.89 to 96.0).
More than half of the children (59%) had had at least one encounter with the health services in the six months after their positive test, so why did they slip through the net? It's impossible to say from this observational study, but the authors suspect that communication problems between primary care doctors (responsible for repeat testing) and the local public health department (responsible for managing confirmed cases) could be partly to blame.
JAMA 2005;293: 2232-37
Higher dose vitamin D prevents first osteoporotic fracture
Vitamin D in the form of cholecalciferol reduces the risk of a first osteoporotic fracture in older people by about a quarter, but only when prescribed at a relatively high dose (700-800 IU/day), say researchers from the United States. A lower dose (400 IU/day), closer to current expert recommendations, failed to prevent hip or any other fractures in this meta-analysis of seven randomised trials. Five of the trials tested the high dose for at least a year and found that cholecalciferol 700-800 IU reduced the risk of hip fracture by 26% (relative risk 0.74, 95% CI 0.61 to 0.88) and of any non-vertebral fracture by 23% (0.77, 0.68 to 0.87) compared with placebo or calcium alone (figure)—translate to one less hip fracture for every 45 people treated for 24-60 months, or one less non-vertebral fracture for every 27 people treated for 12-60 months. In four of the five trials, participants taking vitamin D also took calcium.
Credit: JAMA
An estimated 1 in 3 women and 1 in 6 men will have a hip fracture by the time they reach 90, an increasing disease burden that already costs the US more than $7bn a year. These authors say their results are enough to justify general supplementation with higher dose vitamin D in older people. What they can't say is whether or not older people need to take calcium to make it work.
JAMA 2005;293: 2257-64
Extra heart sounds detected by computer are a poor marker for heart failure
A third or fourth heart sound, otherwise known as gallop rhythm, is a traditional sign of heart failure. To find out how it measures up as modern, computerised, diagnostic test, researchers looked for third or fourth heart sounds in 90 people who were also having a battery of sophisticated tests including cardiac catheterisation and transthoracic echocardiography. Using a computerised heart sound analyser, they found at least one extra heart sound in 41 (46%) of the patients.
Overall, these patients had worse left ventricular function (a higher left ventricular end diastolic pressure or a lower left ventricular ejection fraction) than patients without extra diastolic heart sounds (figure). But heart sounds detected by computer were too insensitive to be useful as a test to rule ourt heart failure. A third heart sound had a sensitivity of only 52% for detecting a low left ventricular ejection fraction (< 50%), and a sensitivity of only 41% for detecting a raised left ventricular end diastolic pressure (> 15 mm Hg). A fourth heart sound had sensitivities below 50% for both, meaning that patients without any extra diastolic heart sounds could still have substantial heart failure. Specificities for the two heart sounds were between 80% and 92%.
Credit: JAMA
JAMA 2005;293: 2238-44
Meta-analyses are top of the citation league table
For years, experts having been telling us that meta-analyses are the best kind of research evidence and that case reports are irrelevant. A study looking at the changing pattern of citations for different types of research shows that we believe them: meta-analyses are now more commonly cited in medical literature than any other type of study (P < 0.001), closely followed by randomised trials. Case reports are hardly cited at all.
Using the Science Citation Index, the authors tracked citation patterns for two groups of studies (2646 in total), selected because the study design was in the title. Studies in one group were published in 1991, studies in the other group were published 10 years later. Citations were higher across the board between 1991 and 2001, probably because of an increase in the number of studies published. Over the same period, meta-analyses made substantial gains over their closest rivals, randomised trials, which in turn made substantial gains over epidemiological studies. Economic and decision analyses have lagged behind randomised trials in the citation stakes and are now cited about as often as narrative reviews. Case reports are very firmly at the bottom of the hierarchy for both periods.
JAMA 2005;293: 2362-6
Osteogenic cells found in peripheral blood
Scientists have found large numbers of previously elusive osteoblastic precursor cells in the peripheral circulation, adding a new dimension to bone growth in humans. The cells, which are thought to mature in to osteoblasts, have previously been seen only in tiny numbers, probably because the technique used to look for them required that the cells stick to plastic, and now it seems that they don't.
Using a new approach, the scientists looked for bone related proteins on the surface of circulating cells instead and found that 1-2% of mononuclear cells in 11 adult male volunteers were positive for osteocalcin. The cells expressed osteogenic genes, formed bone when transplanted in to mice, and mineralised nodules in an in vitro experiment. They were also found in much greater numbers in 11 boys having an adolescent growth spurt and three adult volunteers with recent fractures.
Taken together, all this points to a new circulatory component to bone formation that might have therapeutic potential—for example, in people with poorly healing fractures, say the authors.
New England Journal of Medicine 2005;352: 1959-66
Radical surgery reduces mortality in men with early prostate cancer
Three years ago, a trial reported that men with early, localised prostate cancer did better after a radical prostatectomy than with "watchful waiting." Over about six years of follow-up, radical prostatectomy reduced the risk of death due to prostate cancer by 50% and the risk of distant metastasis by 37%. An extended follow-up (mean 8 years) now shows that radical prostatectomy reduces all cause mortality too. In the latest analysis, 83/347 men who had surgery and 106/348 men who had watchful waiting died from any cause (relative risk 0.74, 95% CI 0.56 to 0.99; P = 0.04). Men who had surgery were less likely to die of prostate cancer (8.6% v 14.4%; P = 0.01) or develop metastases (19.2% v 44.3%) than men who did not have surgery, and the difference for both outcomes had increased (figure).
Credit: NEW ENGLAND JOURNAL OF MEDICINE
Although these results seem clear, the absolute survival gains from surgery are only moderate. The authors say that watchful waiting could become a better option if combined with regular tests for prostate specific antigen. Men could then be offered radical surgery if their test results showed a worsening trend.
New England Journal of Medicine 2005;352: 1977-84
Laparoscopic surgery for colorectal cancer looks safe in the short term
Laparoscopic surgery for colorectal cancer is as safe and effective as open surgery, at least in the first three months after the procedure, according to initial findings from a randomised trial in 794 patients. In an intention to treat analysis, which means analysing patients according to the treatment they were allocated and not necessarily the treatment they received, the groups did not differ in the rate of intraoperative complications (10% in the open group v 10% in the laparoscopy group), postoperative complications (32% v 33%), transfusion requirements (15% v 20%), mortality in hospital (5% v 4%), or quality of life up to three months after surgery. The two groups of patients were also equally likely to have resection margins that were clear of cancer, with the possible exception of patients having an anterior resection for rectal cancer, in whom margins were slightly more likely to be clear after an open operation (94% v 88%).
Nearly a third of patients allocated to laparoscopic surgery had to convert to open surgery in this trial (143/488, 29%). These patients were significantly more likely to have intraoperative and postoperative complications and to need a transfusion.
These results, although important, aren't complete without the longer term outcomes, including recurrence rates and mortality, that will be reported later. In the meantime, more work needs to be done to identify patients at high risk of conversion.