We read with great interest the comments of Van de Velde and Lapauw on our evidence-based guideline (International consensus on the diagnosis and management of dumping syndrome. Nat. Rev. Endocrinol. 16, 448–466 (2020)1). The authors raise a number of issues related to the recommendations of an international consensus panel (Late dumping syndrome or postprandial reactive hypoglycaemic syndrome after bariatric surgery. Nat. Rev. Endocrinol. https://doi.org/10.1038/s41574-021-00473-6 (2021)2).

They suggest the terminology ‘late dumping syndrome’ should be abandoned and replaced by ‘postprandial reactive hypoglycaemic syndrome’. In our opinion, narrowing down the chain of events to reactive hypoglycaemia is an undesirable simplification. Indeed, early postprandial hyperglycaemia is a prerequisite for late hypoglycaemia1,3,4,5. The observation that early hyperglycaemia is accompanied by a rise in pulse rate and haematocrit1,3,4, as well as the therapeutic benefit of acarbose, the SGLT2 inhibitor empagliflozin and agents that increase meal viscosity (such as guar gum, pectin and glucomannan)6, supports the sequence of events where initial overload of the small bowel with nutrients, including carbohydrates, triggers late hypoglycaemia through excessive release of insulin and incretins. Several cohort studies of patients who have undergone bariatric surgery confirm the association between (early) dumping syndrome and hypoglycaemia5,7,8,9. The idiom dumping syndrome, with its early and late component, covers this sequence.

The authors query the proposed threshold of glycaemia <2.8 mmol/l during a modified oral glucose tolerance test (OGTT), as this might occur after a challenge in healthy controls6. The evidence-based guideline only advocates performing OGTT in patients with suspected dumping syndrome1. Moreover, several studies do not report hypoglycaemia after OGTT in controls, even when using the less stringent threshold of 3 mmol/l (refs2,10,11,12). We consider a low glycaemia, either spontaneous or during a dumping provocation test, a prerequisite for initiating treatment approaches with potential adverse effects or high costs, such as somatostatin analogues1,3,4,5. The threshold for abnormal glycaemia is debatable, as we clearly stated1. In a joint position statement, the American Diabetes Association and the European Association for the Study of Diabetes considered 3.0 mmol/l and 2.8 mmol/l as unequivocally hypoglycaemic values, associated with impaired cognitive functioning13. Furthermore, our guideline clearly indicates that the diagnostic work-up should also consider other aspects, such as the rise in haematocrit and pulse1,2,4.

We agree that the relationship between symptoms and hypoglycaemia is inconsistent. One important factor is hypoglycemia unawareness14,15. In addition, the early dumping component is also associated with cardiovascular and neurological symptoms, in the absence of hypoglycaemia1,3. Furthermore, in patients with a history of abdominal surgery, adhesions, altered anatomy and anastomosis, problems might also trigger abdominal and generalized symptoms1,3.

Hence, while we recognize that additional information such as sensitivity, reproducibility and optimal cut-off for the OGTT are desirable, we see many benefits to its use. It enables confirmation of early or late dumping syndrome and consequently directs targeted therapy. The test has also been successfully applied in establishing the spectrum of efficacy of older and newer treatments of dumping syndrome. This complex interplay of fluid volume shifts, altered peptide release, glycaemia shifts and associated symptoms is best covered by the term dumping syndrome and its subsidiaries early and late dumping syndrome.