A 45-year-old man with poorly controlled type 2 diabetes (T2DM) (HbA1c 87 mmol/mol) despite 100 units of insulin per day and severe obesity (BMI 40.2 kg/m2) was referred for bariatric intervention. He declined bariatric surgery or GLP1 agonist therapy. Initially, his glycaemic control improved with dietary modification and better adherence to insulin therapy, but he gained weight. We started a low-energy liquid diet, with 2.2 L of semi-skimmed milk (equivalent to 1012 kcal) per day for 8 weeks (along with micronutrient, salt and fibre supplementation) followed by 16 weeks of phased reintroduction of a normal diet. His insulin was stopped within a week of starting this programme, and over 6 months, he lost 20.6 kg and his HbA1c normalised. However, 1 year later, despite further weight loss, his HbA1c deteriorated dramatically, requiring introduction of linagliptin and canagliflozin, with good response. Five years after initial presentation, his BMI remains elevated but improved at 35.5 kg/m2 and his glycaemic control is excellent with a HbA1c of 50 mmol/mol and he is off insulin therapy. Whether semi-skimmed milk is a safe, effective substrate for carefully selected patients with severe obesity complicated by T2DM remains to be determined. Such patients would need frequent monitoring by an experienced multidisciplinary team.
Meal replacement programmes are an emerging therapeutic strategy to allow severely obese type 2 diabetes patients to achieve clinically impactful weight loss.
Using semi-skimmed milk as a meal replacement substrate might be less costly than commercially available programmes, but is likely to require intensive multidisciplinary bariatric clinical follow-up.
For severely obese adults with poor diabetes control who decline bariatric surgery or GLP1 agonist therapy, a milk-based meal replacement programme may be an option.
Milk-based meal replacement in patients with insulin requiring type 2 diabetes causes rapid and profound reductions in insulin requirements, so rigorous monitoring of glucose levels by patients and their clinicians is necessary.
In carefully selected and adequately monitored patients, the response to oral antidiabetic medications may help to differentiate between absolute and relative insulin deficiency.