Pregnancy outcome in Type 1 diabetes mellitus treated with insulin lispro (Humalog).

AIMS: The use of insulin lispro in pregnancy has not been systematically investigated despite its increasing use. Pooled data from seven centres with experience in the use of insulin lispro were accumulated to evaluate pregnancy outcome in women with Type 1 diabetes. METHODS: Seven units with specialist obstetric diabetes services were recruited to describe their total experience with insulin lispro in pregnancy. Outcomes with respect to the rate of miscarriage, congenital abnormality, perinatal mortality and maternal parameters were recorded in a standardized format. RESULTS: Outcomes on 71 babies from 76 pregnancies were documented. There were six (7.8%) early miscarriages. All 71 babies were liveborn with a mean gestational age of 37.2 weeks, and median birthweight of 3230 g. Seven babies weighed > 4 kg. There were four congenital abnormalities (5.6%). There was a 72% increase in the mean insulin dose (0.75-1.29 IU/kg per day). Maternal glycaemic control improved throughout pregnancy. No women developed retinopathy de novo during pregnancy and six with established retinopathy required laser therapy during pregnancy. CONCLUSIONS: The use of insulin lispro in Type 1 diabetes during pregnancy results in outcomes comparable to other large studies of diabetic pregnancy.

Magnetic resonance imaging techniques demonstrate soft tissue damage in the diabetic foot.

AIMS: Our objective was to assess the qualitative soft tissue changes which occur in the diabetic neuropathic foot, which may predispose to ulceration, using a specific magnetic resonance imaging (MRI) contrast sequence, magnetization transfer (MT) which produces contrast based on exchange between water bound to macromolecules (e.g. collagen) and free water (e.g. extracellular fluid). METHODS: The first metatarsal head of 19 diabetic neuropathic subjects and 11 diabetic non-neuropathic controls was studied using a 'targeted' radiofrequency coil. Neuropathy was classified using vibration perception threshold (VPT) (< or > 25 V), cold threshold (< 1 degree C or > 4 degrees C) and Michigan neuropathy score (< 5 or > 15). Peripheral vascular disease was excluded. Results were expressed as percentage of tissue MT activity in a cross-sectional area. At autopsy full thickness biopsies were taken from the plantar fat pad of 10 unrelated subjects with diabetic neuropathy. RESULTS: Healthy muscle displays high MT activity, whereas adipose tissue induces little activity. Muscle MT activity was considerably reduced (75+/-20%, 30+/-24%, P<0.001) and fat pad MT activity was considerably increased in subjects with neuropathy (37+/-17% 68+/-21%, P<0.001). Muscle fibre atrophy decreases MT activity, whereas fibrous infiltration of the fat pad increases MT activity, fibro-atrophic post-mortem histological changes were found in the plantar fat pads of all neuropathic subjects examined (n = 10). CONCLUSIONS: Changes in MT activity reflect qualitative structural changes which this study reveals are extensive in the diabetic neuropathic foot. Fibrotic atrophy of the plantar fat pad may affect its ability to dissipate the increased weight-bearing forces associated with diabetic neuropathy.

Magnetic resonance imaging reveals micro-haemorrhage in the feet of diabetic patients with a history of ulceration.

Soft tissue haemorrhage in the foot is a possible precursor of ulceration in patients with diabetic peripheral neuropathy. High resolution 'targetted' magnetic resonance imaging was used to scan the forefoot. Neuropathic patients with and without previous ulceration were matched for degree of neuropathy, mean vibration perception threshold 33.5 +/- 4.2 V (previous ulcer) vs 31.0 +/- 6.9 V (no ulcer), age, sex, and duration of diabetes against non-neuropathic controls. There were nine patients in each category. Paramagnetic materials, e.g. iron compounds, cause a signal void ('drop-out') on gradient-echo images which disappear on spin-echo images. Evidence of haemorrhage was seen in 6 patients with previous ulceration, and none in the other groups (p = 0.009, chi square test). Autologous injection of 20 microliters of blood into the foot of a healthy volunteer produced similar images, a 'drop-out' 1 cm across being visible on magnetic resonance scanning 3 days later. Peak vertical forefoot pressures were not significantly different in the neuropathic groups 0.67 +/- 0.20 vs 0.60 +/- 0.13 Pa but were lower in the non-neuropathic group, 0.43 +/- 0.11 Pa (p = 0.0004, Mann-Whitney), and do not explain the appearance of these haemorrhages. Magnetic resonance imaging provides a sensitive way of detecting micro-haemorrhage and its presence may predict an increased risk of foot ulceration.