(Arch Dis Child 1994; 70: 95-98)

The AromaScan system was used to analyse vaginal swabs from 68 women attending a genitourinary clinic. Using clinical criteria, subjects were assessed for bacterial vaginosis. After training the AromaScan system to recognise patterns generated from four patients with and four patients without bacterial vaginosis, 16 of the 17 (94%) remaining subjects were correctly identified as having the condition. The positive predictive value of the test was 61.5%. These results indicate that the AromaScan technology may be of value as a screening test for bacterial vaginosis. (7 Clin Pathol 1997;50:790-791)

Endotoxin levels were measured in the blood and cerebrospinal fluid (CSF) of control individuals and 2 groups of patients with African sleeping sickness. Endotoxin levels were markedly elevated in the blood (infected groups mean endotoxin values 40.2 pg/ml and 53.8 pg/ml, compared to control 11.6 pg/ml, P < 0.0001 for both increases) and CSF (infected groups mean endotoxin values 45.8 pg/ml and 50.1 pg/ml compared to control 6.3 pg/ml, P < 0.0001 for both increases) of the patients. The levels were reduced 6 weeks following different drug treatments in the 2 groups (blood levels to mean 33.8 pg/ml and 28.5 pg/ml; CSF levels to 37.4 pg/ml and 27.0 pg/ml). The blood endotoxin values correlated with the CSF values before treatment (r = 0.74 and 0.57 for the 2 groups; P < 0.0001 for both) and after treatment (r = 0.57 and 0.56 for the 2 groups; P < 0.0001 for both). It is concluded that raised endotoxin equilibrates in the blood and CSF compartments, and may contribute significantly to the pathology of sleeping sickness.

A pilot study using electronic aroma detection was performed over a six-month period to assess the aroma of chronic non-healing venous leg ulcers and the effect of appropriate antibiotic therapy on modification of the aroma. Deep infection with pathogenic organisms was found on biopsy culture in 13 out of 15 patients. Odour analysis was performed at weekly intervals on the ulcer dressings using an AromaScan instrument. Data points on the aroma maps moved from their pre-treatment presentation. Alterations in aroma data correlated well with the progress of the ulcers. Aroma analysis is shown to be a potential tool in monitoring the progress towards healing of chronic venous ulcers.

and storage iron concentrations in life and between blood ferritin and liver iron concentrations after death.'0 It is reasonable to suppose that if infants who die of SIDS are born with excess tissue iron stores then this would be reflected in the blood ferritin measured soon after birth.The availability of blood spots on stored Guthrie cards enabled us to compare blood ferritin concentrations in the first week of life from victims ofSIDS and controls. A significant difference might have formed the basis of a useful screening test for infants at risk.Plasma ferritin concentrations in normal adults vary from 15 to 300 gg/l and reflect the variable concentrations ofstorage iron present.6There is a similar variability of ferritin concentrations in newborn infants so that wide variations in the mean value of the four control cards for each case are to be expected. The mean value of 93 gg/l in eight samples of cord blood is within the expected range.6 In the cord blood samples the plasma component accounts for a mean of 7% of the total blood ferritin-a smaller proportion than the mean of approximately 20% for adult blood.'0Preliminary studies showed that little ferritin could be extracted from the dried spots of blood on Guthrie cards stored for more than six months. However, with the extraction procedure described above, studies of blood spots stored at temperatures from 4 to 56°C indicated that a recovery of about 25% might be expected for cards stored for prolonged periods of time at room temperature.Our study is limited by the small numbers and because it would not be possible to detect a difference between a case and the control cards of less than 20%. Nevertheless