Ultrasound detection of insulin-induced lipohypertrophy in Type 1 and Type 2 diabetes

To define standard criteria for the detection of lipohypertrophy using ultrasonography and to determine the accuracy of this method.

Individuals using insulin therapy for ≥2 years with unknown lipohypertrophy status were enrolled at a diabetes education centre. A team of diabetes educator nurses performed a clinical examination for evidence of lipohypertrophy and a separate team of ultrasonographers examined participants in a blinded fashion.

The echo signature for lipohypertrophy consisted of location in the subcutaneous layer and lesions that were 1) well circumscribed either by hyperechoic foci with defined borders or a nodular shape with a hypoechoic halo, 2) heterogeneous in echotexture compared with surrounding tissue, 3) associated with distortion of surrounding connective tissue with 4) absence of vascularity and 5) absence of capsule. Ultrasonography identified individuals with lipohypertrophy significantly more frequently than inspection or palpation (P<0.0001). Inter-observer agreement was moderate (κ=0.50) and limited by the presence of subclinical lesions in 73% of the participants.

The ultrasound detection of lipohypertrophy is consistent with clinical examination and is reproducible using a defined echo signature. (ClinicalTrials.gov registration no: NCT02348099).