What's new in the other general journals
http://www.100md.com
《英国医生杂志》
Chloramphenicol eye drops are unnecessary for uncomplicated conjunctivitis
Chloramphenicol eye drops are no better than placebo for children with conjunctivitis, a randomised trial has found. The authors think primary care doctors should stop prescribing chloramphenicol eye drops for children with uncomplicated conjunctivitis and should instead tell parents to keep the affected eye clean and to come back if it doesn't get better. The trial, which included 326 children aged 12 years or less, reported clinical cure rates of 83% (128/155) among children given placebo drops for seven days compared with 86% (140/162) among children given chloramphenicol. Four fifths of the children had bacterial conjunctivitis. Children given placebo drops were no more likely to relapse and no less likely to report side effects than the others, but they took about a third of a day longer to get better.
Credit: LANCET
This is the first trial done exclusively in primary care, where chloramphenicol eye drops are a standard treatment for conjunctivitis in children. Changing doctors' prescribing habits and patients' expectations will be difficult, particularly in countries such as the UK where schools and nurseries often exclude children until they are treated. The economic arguments are compelling, however. Every year one million children present to UK general practitioners with conjunctivitis; that's over a million consultations and potentially a million prescriptions.
Lancet 2005 June 22; doi 10.1016/S0140-6736(05)66709-8
Breast feeding is linked to better eyesight among children in Singapore
Researchers have found a link between having been breast fed and better eyesight among schoolchildren in Singapore. Their cross sectional study included 797 children aged 10-12 whose parents filled in a questionnaire one month before their child's routine eye test at school. The questionnaire included questions about breast feeding, but parents were unaware that this was the focus of the study. Overall, 521 children (65%) were shortsighted, but the prevalence of shortsightedness was lower among those who had been at least partly breast fed (259/418 (62.0%) v 262/379 (69.1%), P = 0.04). The association remained after adjustment for confounding factors (adjusted odds ratio 0.58; 95% confidence interval 0.39 to 0.84).
It's impossible to say from this study whether the link between breast feeding and better eyesight is causal, but there are biologically plausible explanations if it is. Breast milk is full of micronutrients that are generally missing from formula milk. One of them, docosahexaenoic acid, is important for the healthy development of cortical neurones and photoreceptors.
Shortsightedness is extremely common among children in Singapore, where this study was done. Even if the findings turn out to be clinically important, they may not be relevant in countries such as the UK where the incidence of childhood myopia is considerably lower.
JAMA 2005;293: 3001-2.
Antibiotics don't work for patients with cough
Most primary care doctors now think twice before prescribing antibiotics to patients with a simple cough, but the debate rumbles on because of limited evidence from decent randomised trials. In search of a definitive answer, researchers from the United Kingdom spent five years recruiting and studying 800 primary care patients with cough but no signs of pneumonia. Patients were given immediate antibiotics, no antibiotics, or a prescription they could pick up from the practice receptionist if they weren't better within two weeks.
Immediate antibiotics (10 days of amoxicillin or erythromycin) did not reduce the duration or severity of patients' coughs, although other symptoms such as wheeze and disturbed sleep got better about one day earlier than with no antibiotics. Since patients in this trial coughed for a mean of nine days before going to their doctor and for a mean of nearly 12 days afterwards, one day less of other symptoms is arguably a poor trade off against the well known risks of antibiotic resistance caused by liberal prescribing. Results for vulnerable subgroups such as elderly people and those with green sputum were no more convincing. If anything, older people did worse after immediate treatment with antibiotics.
Credit: JAMA
JAMA 2005;293: 3029-35
New score predicts risk of stroke in patients with transient ischaemic attack
Researchers from Oxford in the United Kingdom have developed a simple scoring system to help predict which patients with transient ischaemic attack are most likely to have a stroke soon afterwards. The system, called ABCD, scores patients up to a maximum of 6 points according to their age ( 60 years = 1), blood pressure (systolic > 140 mm Hg or diastolic 90, or both = 1), clinical picture (unilateral weakness = 2, speech disturbance without weakness = 1, other = 0) and duration of symptoms ( 60 minutes = 2, 10-59 = 1, < 10 = 0). When tested in a population based cohort of nearly 400 patients with suspected transient ischaemic attack, the ABCD score was highly predictive of early stroke (P < 0.0001); 19 out of the 20 strokes occurred in the 27% of patients with a score of 5 or more. Overall, the seven day risk of stroke was 0.4% for patients who scored less than 5, 12.1% for patients who scored 5, and 31.4% for patients who scored 6.
This preliminary validation looks promising but there's always room for more. In the meantime the authors hope their simple system will help primary care doctors assess patients for referral more rapidly and reliably, and help hospital specialists triage referrals for emergency investigations and treatment. A score of 6, say the authors, should be treated as a medical emergency.
Lancet 2005 June 21; doi 10.1061/S0140-6736(05)66702-5
Monoclonal antibodies for diabetes: benefits offset by serious side effects
Ever since diabetes was recognised as an autoimmune disease, researchers have been looking for a safe immunosuppressant to help with treatment. The latest candidate is ChAglyCD3, a monoclonal antibody against CD3 T cells. This was recently tested in a small and select group of 80 young people with new type 1 diabetes. In a placebo controlled trial, the antibody helped preserve residual cell function and stop insulin requirements from rising for the first 18 months after diagnosis. The antibody worked best for patients with reasonable cell function at the start of the trial (above the 50th centile). In this subgroup, patients given the antibody needed significantly less insulin 18 months later than patients not given the antibody (0.22 IU/kg/day v 0.61 IU/kg/day, P < 0.001).
Credit: NEW ENGLAND JOURNAL OF MEDICINE
Despite this small success, it's still early days for monoclonal antibodies as a treatment for diabetes, not least because of serious side effects. All the patients given this antibody were ill during treatment with fever, sore joints and muscles, and headache. Half of them developed a rash, and three quarters developed an acute fever with sore throat and cervical lymphadenopathy at 16-21 days after treatment.
New England Journal of Medicine 2005;352: 2598-608
Postnatal depression in fathers linked to behavioural problems in their young sons
Men get postnatal depression too, but in contrast to postnatal depression in women, we know very little about it. In one large cohort from Bristol in the United Kingdom, 4% of over 8000 men were depressed eight weeks after the birth of their baby. Three and a half years later, their children were twice as likely to have emotional and behavioural problems as other children in the cohort (odds ratio 2.09, 95% confidence interval 1.42 to 3.08). Fathers' postnatal depression seemed particularly hard on boys in this study, more than doubling their risk of conduct problems at the age of three and a half (2.66, 1.67 to 4.25). As expected, maternal postnatal depression also increased the risk of psychological problems in young children, but the effects were similar in boys and girls.
These findings come from a long running cohort of parents and their children, originally recruited between 1991 and 1992. The results were adjusted for social class, mothers' education, and depression in the other parent, so the authors are fairly confident that the associations they found were real, independent, and possibly due to depression preventing fathers from interacting normally with their infants. They plan further studies to find out why a depressed father might be more of a problem for boys than girls.
Lancet 2005;365: 2201-5
Women with SLE can take HRT
Doctors generally avoid prescribing hormone replacement therapy to postmenopausal women with systemic lupus erythematosus because of a widespread belief that exogenous oestrogens make the disease worse. A large placebo controlled trial now reports that hormone replacement therapy with oestrogen and medroxyprogesterone does not cause serious flare-ups of lupus, but does increase the risk of a mild or moderate flare-up by about a third (1.14 flares/person-year v 0.86 flares/person-year, relative risk 1.34, P = 0.01).
Of the 351 postmenopausal women who participated in the trial, a fifth had stable but active disease and the rest were in remission. They took active treatment (0.625 mg of conjugated oestrogen daily plus 5 mg of medroxyprogesterone for 12 days each month) or placebo for one year, during which time 13 women (7.5%) taking hormone therapy and eight women (4.5%) taking placebo had a severe flare-up. About half the women had a mild or moderate flare-up during the trial, but the incidence was significantly higher among women taking hormone therapy (59% v 50%).
Recruitment to this trial was stopped early by the publication of landmark data linking hormone therapy to breast cancer, stroke, and heart disease. However, the authors say that a short course may still be appropriate for women with systemic lupus erythematosus who have menopausal symptoms.
Annals of Internal Medicine 2005;142: 953-62
Prophylactic antibiotics improve survival after chemotherapy
Prophylactic antibiotics can prevent infections and save lives in patients who develop neutropenia after chemotherapy for cancer, say researchers from Israel. Their meta-analysis of 95 randomised trials showed a clear survival advantage for patients given prophylaxis, (relative risk of death from all causes 0.67; 95% confidence interval 0.55 to 0.81) a finding that could overturn current guidelines.
Quinolones such as ciprofloxacin looked particularly effective, reducing mortality from all causes by about half compared with placebo (relative risk 0.52; 0.35 to 0.77). Most of the 9283 patients in these trials had haematological cancers such as acute leukaemia, lymphomas, or multiple myeloma, and in most trials patients were given antibiotics with their chemotherapy.
Credit: ANNALS OF INTERNAL MEDICINE
This is not the first meta-analysis to look at this issue, but it is the first to report a clear result for mortality in favour of antibiotic prophylaxis, possibly because this meta-analysis is at least four times bigger than previous attempts.
Chloramphenicol eye drops are no better than placebo for children with conjunctivitis, a randomised trial has found. The authors think primary care doctors should stop prescribing chloramphenicol eye drops for children with uncomplicated conjunctivitis and should instead tell parents to keep the affected eye clean and to come back if it doesn't get better. The trial, which included 326 children aged 12 years or less, reported clinical cure rates of 83% (128/155) among children given placebo drops for seven days compared with 86% (140/162) among children given chloramphenicol. Four fifths of the children had bacterial conjunctivitis. Children given placebo drops were no more likely to relapse and no less likely to report side effects than the others, but they took about a third of a day longer to get better.
Credit: LANCET
This is the first trial done exclusively in primary care, where chloramphenicol eye drops are a standard treatment for conjunctivitis in children. Changing doctors' prescribing habits and patients' expectations will be difficult, particularly in countries such as the UK where schools and nurseries often exclude children until they are treated. The economic arguments are compelling, however. Every year one million children present to UK general practitioners with conjunctivitis; that's over a million consultations and potentially a million prescriptions.
Lancet 2005 June 22; doi 10.1016/S0140-6736(05)66709-8
Breast feeding is linked to better eyesight among children in Singapore
Researchers have found a link between having been breast fed and better eyesight among schoolchildren in Singapore. Their cross sectional study included 797 children aged 10-12 whose parents filled in a questionnaire one month before their child's routine eye test at school. The questionnaire included questions about breast feeding, but parents were unaware that this was the focus of the study. Overall, 521 children (65%) were shortsighted, but the prevalence of shortsightedness was lower among those who had been at least partly breast fed (259/418 (62.0%) v 262/379 (69.1%), P = 0.04). The association remained after adjustment for confounding factors (adjusted odds ratio 0.58; 95% confidence interval 0.39 to 0.84).
It's impossible to say from this study whether the link between breast feeding and better eyesight is causal, but there are biologically plausible explanations if it is. Breast milk is full of micronutrients that are generally missing from formula milk. One of them, docosahexaenoic acid, is important for the healthy development of cortical neurones and photoreceptors.
Shortsightedness is extremely common among children in Singapore, where this study was done. Even if the findings turn out to be clinically important, they may not be relevant in countries such as the UK where the incidence of childhood myopia is considerably lower.
JAMA 2005;293: 3001-2.
Antibiotics don't work for patients with cough
Most primary care doctors now think twice before prescribing antibiotics to patients with a simple cough, but the debate rumbles on because of limited evidence from decent randomised trials. In search of a definitive answer, researchers from the United Kingdom spent five years recruiting and studying 800 primary care patients with cough but no signs of pneumonia. Patients were given immediate antibiotics, no antibiotics, or a prescription they could pick up from the practice receptionist if they weren't better within two weeks.
Immediate antibiotics (10 days of amoxicillin or erythromycin) did not reduce the duration or severity of patients' coughs, although other symptoms such as wheeze and disturbed sleep got better about one day earlier than with no antibiotics. Since patients in this trial coughed for a mean of nine days before going to their doctor and for a mean of nearly 12 days afterwards, one day less of other symptoms is arguably a poor trade off against the well known risks of antibiotic resistance caused by liberal prescribing. Results for vulnerable subgroups such as elderly people and those with green sputum were no more convincing. If anything, older people did worse after immediate treatment with antibiotics.
Credit: JAMA
JAMA 2005;293: 3029-35
New score predicts risk of stroke in patients with transient ischaemic attack
Researchers from Oxford in the United Kingdom have developed a simple scoring system to help predict which patients with transient ischaemic attack are most likely to have a stroke soon afterwards. The system, called ABCD, scores patients up to a maximum of 6 points according to their age ( 60 years = 1), blood pressure (systolic > 140 mm Hg or diastolic 90, or both = 1), clinical picture (unilateral weakness = 2, speech disturbance without weakness = 1, other = 0) and duration of symptoms ( 60 minutes = 2, 10-59 = 1, < 10 = 0). When tested in a population based cohort of nearly 400 patients with suspected transient ischaemic attack, the ABCD score was highly predictive of early stroke (P < 0.0001); 19 out of the 20 strokes occurred in the 27% of patients with a score of 5 or more. Overall, the seven day risk of stroke was 0.4% for patients who scored less than 5, 12.1% for patients who scored 5, and 31.4% for patients who scored 6.
This preliminary validation looks promising but there's always room for more. In the meantime the authors hope their simple system will help primary care doctors assess patients for referral more rapidly and reliably, and help hospital specialists triage referrals for emergency investigations and treatment. A score of 6, say the authors, should be treated as a medical emergency.
Lancet 2005 June 21; doi 10.1061/S0140-6736(05)66702-5
Monoclonal antibodies for diabetes: benefits offset by serious side effects
Ever since diabetes was recognised as an autoimmune disease, researchers have been looking for a safe immunosuppressant to help with treatment. The latest candidate is ChAglyCD3, a monoclonal antibody against CD3 T cells. This was recently tested in a small and select group of 80 young people with new type 1 diabetes. In a placebo controlled trial, the antibody helped preserve residual cell function and stop insulin requirements from rising for the first 18 months after diagnosis. The antibody worked best for patients with reasonable cell function at the start of the trial (above the 50th centile). In this subgroup, patients given the antibody needed significantly less insulin 18 months later than patients not given the antibody (0.22 IU/kg/day v 0.61 IU/kg/day, P < 0.001).
Credit: NEW ENGLAND JOURNAL OF MEDICINE
Despite this small success, it's still early days for monoclonal antibodies as a treatment for diabetes, not least because of serious side effects. All the patients given this antibody were ill during treatment with fever, sore joints and muscles, and headache. Half of them developed a rash, and three quarters developed an acute fever with sore throat and cervical lymphadenopathy at 16-21 days after treatment.
New England Journal of Medicine 2005;352: 2598-608
Postnatal depression in fathers linked to behavioural problems in their young sons
Men get postnatal depression too, but in contrast to postnatal depression in women, we know very little about it. In one large cohort from Bristol in the United Kingdom, 4% of over 8000 men were depressed eight weeks after the birth of their baby. Three and a half years later, their children were twice as likely to have emotional and behavioural problems as other children in the cohort (odds ratio 2.09, 95% confidence interval 1.42 to 3.08). Fathers' postnatal depression seemed particularly hard on boys in this study, more than doubling their risk of conduct problems at the age of three and a half (2.66, 1.67 to 4.25). As expected, maternal postnatal depression also increased the risk of psychological problems in young children, but the effects were similar in boys and girls.
These findings come from a long running cohort of parents and their children, originally recruited between 1991 and 1992. The results were adjusted for social class, mothers' education, and depression in the other parent, so the authors are fairly confident that the associations they found were real, independent, and possibly due to depression preventing fathers from interacting normally with their infants. They plan further studies to find out why a depressed father might be more of a problem for boys than girls.
Lancet 2005;365: 2201-5
Women with SLE can take HRT
Doctors generally avoid prescribing hormone replacement therapy to postmenopausal women with systemic lupus erythematosus because of a widespread belief that exogenous oestrogens make the disease worse. A large placebo controlled trial now reports that hormone replacement therapy with oestrogen and medroxyprogesterone does not cause serious flare-ups of lupus, but does increase the risk of a mild or moderate flare-up by about a third (1.14 flares/person-year v 0.86 flares/person-year, relative risk 1.34, P = 0.01).
Of the 351 postmenopausal women who participated in the trial, a fifth had stable but active disease and the rest were in remission. They took active treatment (0.625 mg of conjugated oestrogen daily plus 5 mg of medroxyprogesterone for 12 days each month) or placebo for one year, during which time 13 women (7.5%) taking hormone therapy and eight women (4.5%) taking placebo had a severe flare-up. About half the women had a mild or moderate flare-up during the trial, but the incidence was significantly higher among women taking hormone therapy (59% v 50%).
Recruitment to this trial was stopped early by the publication of landmark data linking hormone therapy to breast cancer, stroke, and heart disease. However, the authors say that a short course may still be appropriate for women with systemic lupus erythematosus who have menopausal symptoms.
Annals of Internal Medicine 2005;142: 953-62
Prophylactic antibiotics improve survival after chemotherapy
Prophylactic antibiotics can prevent infections and save lives in patients who develop neutropenia after chemotherapy for cancer, say researchers from Israel. Their meta-analysis of 95 randomised trials showed a clear survival advantage for patients given prophylaxis, (relative risk of death from all causes 0.67; 95% confidence interval 0.55 to 0.81) a finding that could overturn current guidelines.
Quinolones such as ciprofloxacin looked particularly effective, reducing mortality from all causes by about half compared with placebo (relative risk 0.52; 0.35 to 0.77). Most of the 9283 patients in these trials had haematological cancers such as acute leukaemia, lymphomas, or multiple myeloma, and in most trials patients were given antibiotics with their chemotherapy.
Credit: ANNALS OF INTERNAL MEDICINE
This is not the first meta-analysis to look at this issue, but it is the first to report a clear result for mortality in favour of antibiotic prophylaxis, possibly because this meta-analysis is at least four times bigger than previous attempts.