Low-molecular-weight heparin (LMWH) is effective in the prevention of postoperative venous thromboembolism but does it have the safety advantages over standard heparin (SH) that have been claimed? In a multicentre randomised trial in 3809 patients undergoing major abdominal surgery (1894 LMWH, 1915 SH) heparin was given preoperatively and continued for at least 5 postoperative days. Patients were assessed in the postoperative period and were followed up for at least 4 weeks, the emphasis being on safety. Major bleeding events occurred in 69 (3.6%) patients in the LMWH group and 91 (4.8%) patients in the SH group (relative risk 0.77, 95% confidence interval 0.56-1.04; p = 0.10). 93 indices of major bleeding were observed in the 69 LMWH patients and 141 in the SH patients. (p = 0.058). Severe bleeding was less frequent in the LMWH group (1.0% vs 1.9%; p = 0.02), as was wound haematoma (1.4% vs 2.7%; p = 0.007). Bleeding episodes with LMWH were less likely to lead to further surgery to evacuate a haematoma or to control bleeding, and injection site bruising was also less common in the LMWH group. No significant differences were found in the efficacy of the two agents. Perioperative death rates were 3.3% in the LMWH group and 2.5% in the SH group; pulmonary emboli were detected in 0.7% and 0.7%; and deep-vein thrombosis was diagnosed in 0.6% of patients in each group. Follow-up was done on 91% of 3699 evaluable patients. There were 19 further deaths (10 LMWH, 9 SH group) and 25 patients with thromboembolic complications (15 and 10). Of the 3 patients with fatal pulmonary emboli during follow-up 2 had received LMWH and 1 SH. The two drugs were of similar efficacy. The primary end point, the frequency of major bleeding, showed a 23% reduction in the LMWH group, but this difference was not significant. The secondary safety end points revealed that LMWH was significantly better than SH. Fatal pulmonary embolism occurs rarely (0.09%) following discharge from hospital so the cost benefit ratio would not justify prolonged prophylaxis in this setting.

There is concern that feeding full-strength animal milk to infants aged less than 6 months with diarrhoea may have adverse consequences. We assessed the effects on clinical course of two feeding regimens in 159 Guatemalan and Brazilian infants aged 2 weeks to 6 months who had had acute diarrhoea for 120 h or less, showed signs of mild to moderate dehydration, and had no complications. After correction of dehydration, infants were assigned randomly to receive continued full-strength milk feeding or initial feeding with diluted milk with regrading to full-strength milk over 48 h. There were no significant differences between feeding groups in rate of treatment failures (-1%, 95% Cl -14 to 12%) or mean (SD) total stool output (full-strength milk 335 [268] g/kg, diluted milk 338 [354] g/kg) and duration of diarrhoea (92 [50] vs 92 [44] h). A significant association was found between presence of reducing substances in stools and treatment failure (OR 4.3, 95% CI 1.1 to 16.8), but reducing substances in stools were common both in treatment successes (61%) and in failures (87%). Our study supports the conclusion that, for infants under 6 months of age with diarrhoea whose only food is animal milk or formula, the milk or formula normally given should be provided in full strength as soon as dehydration has been corrected.

The measurement of gastric intramucosal pH (pHi) has been advocated to assist in decision-making for critically ill patients. To assess whether the information obtained from the measurement of pHi can be obtained from other measurements of metabolic acidosis, we studied 20 consecutive patients admitted to the intensive care unit. A mean of eight (range two to fourteen) data sets per patient were obtained, comprising measurement of arterial pH, pO2, pCO2, and oxygen saturation, tonometer balloon fluid pCO2, arterial pressures, and cardiac output. Bicarbonate concentration, base deficit or excess in blood and extracellular fluid, and pHi were calculated from these measurements. Relations between the variables and pHi were assessed by within-subject correlation comparisons. There were significant correlations (r > 0.6, p < 0.001) between markers of metabolic acidosis (base deficit in blood and extracellular fluid and bicarbonate concentration) and pHi. A blood base deficit of -4.65 or less and an extracellular-fluid base deficit of -6.13 or less could estimate pHi below 7.32 (lower limit of normal range) with sensitivity of at least 77% and specificity of at least 96%. There was no patient in whom either pHi or blood base deficit consistently reflected acidosis when the other variable did not. We conclude that the information that is obtained by gastric tonometry for pHi can be obtained more simply from measurements of metabolic acidosis; these variables can be calculated from routinely available blood-gas measurements.

The safety and efficacy of ivermectin in the prevention of blindness from onchocerciasis have been established in many studies that have addressed the drug's effects on the front of the eye. We undertook a study with sufficient statistical power to detect an effect on optic nerve disease (OND), probably the main cause of blindness in the disorder. The trial was based in 34 mesoendemic communities in Kaduna State, Nigeria. Villagers aged 5 years and older were randomly assigned annual dosing with ivermectin or placebo for 3 years. Participants underwent medical and ophthalmological examinations before the first, third, and fourth treatments. 3522 villagers aged 15 and older were re-examined at least once. Skin-snip samples were taken at baseline for calculation of microfilarial load. The outcome measure was development of disc pallor accompanied by objective evidence of deterioration in visual function; 116 subjects (45 ivermectin-treated, 71 placebo-treated) showed such changes during the trial. The incidence rate ratio (ivermectin vs placebo) was 0.90 (95% CI 0.54-1.51) for subjects with loads of 0-10 mf (microfilariae) per mg skin and 0.52 (0.29-0.93) for subjects with more than 10 mf/mg. The incidence rate ratio varied little when account was taken of age, sex, presence of pre-existing disc pallor in one eye, previous use of diethylcarbamazine citrate, or doses of ivermectin or placebo received. There was evidence that ivermectin reduced the incidence of OND in subjects with microfilarial loads above 10 mf/mg but had little effect in those with lower loads. Sustained annual delivery of ivermectin could prevent a substantial proportion of onchocercal blindness in mesoendemic communities.

Women with low bone density in the radius or calcaneus are at increased risk of hip fracture. To see whether bone density of the hip measured by dual X-ray absorptiometry is a better predictor of hip fracture than measurements of other bones, we assessed bone density at several sites in 8134 women aged 65 years or more. 65 women had hip fractures during a mean follow-up of 1.8 years. Each SD decrease in femoral neck bone density increased the age-adjusted risk of hip fracture 2.6 times (95% CL 1.9, 3.6). Women with bone density in the lowest quartile had an 8.5-fold greater risk of hip fracture than those in the highest quartile. Bone density of the femoral neck was a better predictor than measurements of the spine (p < 0.0001), radius (p < 0.002), and moderately better than the calcaneus (p = 0.10). Low hip bone density is a stronger predictor of hip fracture than bone density at other sites. Efforts to prevent hip fractures should focus on women with low hip bone density.

Although vitamin A deficiency in children seems to increase susceptibility to infection and community trials have shown that vitamin A supplementation can reduce childhood mortality from infectious diseases, the underlying biological mechanisms are largely unknown. We conducted a randomised, double-masked, placebo-controlled clinical trial among children in West Java, Indonesia, to determine whether vitamin A deficiency is associated with abnormalities in T-cell subsets and whether vitamin A supplementation affects T-cell subsets. We studied 55 children aged 3-6 years--30 with xerophthalmia and 25 without. Acutely malnourished children (< 80% of reference weight-for-height) were excluded. CD4/CD8 ratios and the proportions of circulating CD4 naive, CD4 memory, CD8, CD45RA, and CD8, CD45RO T-cell subsets were measured. Children with xerophthalmia had lower CD4/CD8 ratios (p < 0.08), lower proportions of CD4 naive T cells (p < 0.03), and higher proportions of CD8, CD45RO T cells (p < 0.04) than those without xerophthalmia. 26 children were given vitamin A supplementation (60 mg retinol equivalent) and 29 received placebo. 5 weeks later the vitamin A group had higher CD4/CD8 ratios (p < 0.001), higher proportions of CD4 naive T cells (p < 0.01), and lower proportions of CD8, CD45RO T cells (p < 0.05) than the placebo group. Vitamin-A-deficient children have underlying immune abnormalities in T-cell subsets and these abnormalities are reversible with vitamin A supplementation.

Islet transplantation has been slow to develop as a therapy for type I diabetes mellitus. Conventional immunosuppression does not protect islet allografts from early failure and by current techniques the yield of purified islets from a single pancreas is inadequate or only marginally in excess of the number needed to sustain normoglycaemia. We transplanted unpurified islets from a single pancreas concomitantly with a kidney to two uraemic diabetic patients. The novel agent 15-deoxyspergualin, along with antilymphocyte globulin, was used for induction immunosuppression, and azathioprine, prednisone, and cyclosporin for maintenance. Islet function has been sustained in both, and the second patient is insulin-independent and euglycaemic more than 6 months after transplantation.

Most hip fractures seem to be related to trauma near the hip, so a controlled trial was conducted to investigate the effect of external hip protectors on the prevention of such fractures in residents of a nursing home. 10 of the 28 wards in the nursing home were randomised to receive external hip protectors; thus 167 women and 80 men were given protectors and 277 and 141 men no protectors. A fall register was set up for 2 treatment wards (45 residents) and 2 control wards (76 residents). There were 8 hip and 15 non-hip fractures in the hip-protector group and 31 hip and 27 non-hip fractures in the control group. The relative risk of hip fractures among women and men in the intervention group was 0.44 (95% CI 0.21-0.94). None of the 8 residents in the intervention group who had a hip fracture was wearing the device at the time of the fracture. 154 falls were registered and 20% of these falls produced a direct impact to the hip. In 25 falls direct impact to the hip was sustained at a time when hip protectors were not being worn, and 6 fractures were produced. The study indicates that external hip protectors can prevent hip fractures in nursing-home residents.