Initial doses should be low (e.g. gliclazide 40 mg b.d.), with slow and gradual titration while monitoring glucose, HbA1c and, most importantly, hypoglycemia (see below). Dose–response relationships are weak and maximum glycemic effect is often seen at doses lower than the recommended maximum; conversely, increasing doses in the face of limited further improvement is not worthwhile. While the maximum effect of sulphonylureas is thought to be in the early stages of type 2 diabetes, when J3-cell reserve is at its maximum, it can sometimes be worthwhile reintroducing them later, for example when trying to discontinue insulin treatment using the newer agents; responses can sometimes be surprising. There is consistent evidence from UKPDS, ADOPT and RECORD studies that although the initial glycemic response to sulfonylureas is more dramatic than with either metformin or glitazones, with lower nadir HbA1c levels, the subsequent failure rate (secondary or sulphonylurea failure) is higher. In ADOPT (Fig. 6.1), using an arbitrary monotherapy failure threshold of 10 mmol/L (180 mg/ dL), the approximate annual failure rates were 3% for rosiglitazone, 4% for metformin, and 7% for glibenclamide.
Weight gain is common, UKPDS and ADOPT (the latter using glibenclamide 15 mg daily) reporting about a 2-kg gain, mostly in the first year of treatment.
The meglitinide analogues (repaglinide and nateglinide) bind to a different site of the sulphonylurea receptor, and have a much more rapid onset and offset of action. They are therefore suitable for taking immediately before a meal and for controlling predominant postprandial hyperglycemia (possibly beneficial – one study showed that for similar HbA1c reductions, increase in carotid intima–media thickness (CIMT) was less pronounced after repaglinide compared with glibenclamide ). Note, however, that in the NAVIGATOR study (2010) nateglinide if anything increased post- load glucose levels (in subjects with IGT, where the biochemical problem is predominantly post load), though it reduced fasting glucose levels. Mild hypoglycemia can occur with both agents but severe events are uncommon, and importantly they are unlikely to be prolonged, the major hazard with sulphonylureas. Weight gain is about the same as with sulphonylureas. They are suitable for patients with renal impairment, and the flexible dosing may be helpful during Ramadan fasting. The disadvantage is the need for multiple daily doses, which is likely to be less appropriate for the elderly who might particularly benefit from their short action. Repaglinide has similar glucose-lowering effects as sulphonylureas, though nateglinide is less potent (up to about 0.6% HbA1c lowering, 7 mmol/mol). Nonetheless, they are valuable agents that are probably underused in clinical practice.
Hypoglycemia occurs with any sulphonylurea (0.4–0.6% per year in UKPDS; 0.6% of patients in ADOPT had a serious hypoglycemic event, and nearly 40% reported hypoglycemia). Clinically it occurs about 4 hours after a morning dose, often in late morning after mild exertion and omitting any mid-morning snack (e.g. on shopping or golfing days). There is an impression that as glycemic targets continue to be driven down, significant sulphonylurea-related hypoglycemia is becoming more common. Discuss in detail the possibility of hypoglycemia when starting sulphonylurea therapy, outline the likely symptoms, and indicate clearly what procedure the patient must follow if hypoglycemia does occur (dose reduction, stopping treatment).
Profound and prolonged hypoglycemia, sometimes resulting in death or brain damage, can occur in the elderly, in those with intercurrent illness, malnutrition or alcoholism, and in those with impaired renal function. Admit, especially the elderly, and treat with a prolonged infusion of 10–20% glucose. Bolus injections of glucose and glucagon can stimulate residual insulin secretion in type 2 patients, hence the need for continuing glucose infusion with careful blood glucose monitoring. Monitor serum potassium frequently; it may fall with the high circulating insulin levels. Octreotide may be of value in very severe and prolonged hypoglycemia, and a bolus subcutaneous injection of 75 µg on admission, in addition to standard treatment, reduces the risk of relapsing hypoglycemia .