Reactive hypoglycemia is a state characterised by sympathetic or neuroglycopenic symptoms associated with hypoglycaemia in the postprandial state resulting in considerable distress to the patient. It is our practice to carry out either extended glucose tolerance tests (eGTTs) or mixed meal tests in these patients. We describe two patients who experienced hypoglycaemic symptoms early and late during eGTT. The patient who experienced symptoms early, in contrast to the patient who presented with late symptoms, did not possess any characteristics of the metabolic syndrome. Based on clinical symptoms, glucose, insulin, and free fatty acid (FFA) levels, we speculate on possible mechanisms that may have accounted for each of their presentation patterns. We then discuss low glycaemic index diet which will be the mainstay of management.
Reactive hypoglycaemia is often considered in patients developing sympathetic or neuroglycopenic symptoms in the postprandial state. Prior to this diagnosis being reached, it is essential that hypoglycaemia must be demonstrated to be associated with the symptoms (
The results of the eGTT carried out in our metabolic investigation unit have shown differing clinical and biochemical patterns requiring an understanding of the relationship between glucose and factors influencing it such as insulin levels, insulin resistance, and counterregulatory measures. We have observed symptomatic hypoglycaemia both early and late during the eGTT. We present a couple of patients demonstrating possible reactive hypoglycaemia at different time points during the eGTT and then speculate on possible causes. This will be followed by possible actions that could be taken, focusing mainly on dietary measures.
Patients with possible postprandial reactive hypoglycaemia are often referred to the endocrine or metabolic clinics at Good Hope Hospital and are investigated in the metabolic investigation unit following clinical assessment. The eGTT is not recommended in patients with intercurrent illness, dumping syndrome, or postsurgery due to difficulty in the interpretation of results. It is carried out following a minimum 10 hours of fasting with only water permitted, preceded by 3 days or more of normal diet and activity. Polycal (113 mL which contains 75 g of anhydrous glucose, made up to 200 mL of water) is orally administered, followed by a further 100 mL of water. Blood samples (for glucose, insulin, c-peptide, and free fatty acids) are carried out at 0 minutes and at 30 minutes intervals for 4 hours and also at the point of patient’s symptoms. The investigation may require discontinuation if the patient is overtly symptomatic and requires rescue. All the samples are analysed for glucose with insulin, C-peptide, and free fatty acids estimated in the event of low blood sugar (≤3.5 mmol/L), hypoglycaemic symptoms, or retrospectively requested when deemed useful in reaching a diagnosis. Abnormal results are then discussed at a following multidisciplinary team meeting and future management and followup mapped out. The patients are then seen in the out-patient clinics, the results discussed, and possible treatments discussed with appropriate followup.
Glucose was measured using an Enzymatic Hexokinase/G-6-PDH assay on the Abbott Architect C system, the assay CV being 2.2%. FFA was estimated at the Birmingham Children’s Hospital using a colourimetric assay on the Olympus 640 analyser. Insulin was measured at the SAS Peptide Hormone Centre, Royal Surrey Hospital, Guilford, using the Mercodia Iso Insulin ELISA Immunoassay.
We present 2 patients presenting with hypoglycaemic symptoms during the eGTT at different time points and describe their clinical features, recovery, and biochemical changes.
Patient A, a 70-year-old woman (weight: 67 kg; BMI: 24.2) had experienced dizzy spells and collapse for 2 years. She did not possess any of the features of the metabolic syndrome [
It was decided to carry out an eGTT. The results of the relevant biochemistry are presented in Table
(a) Biochemistry parameters for patient A following the eGTT with 113 mL of Polycal. (b) Biochemistry parameters for patient A following the eGTT with 57 mL of Polycal.
Time |
Glucose |
Insulin |
FFA |
---|---|---|---|
0 | 4.6 | 32 | 497 |
30 | 11.9 | 1760 | |
60 | 11.9 | 2940 | |
90 | 6.2 | 640 | <50 |
120 | 2.3 | 211 | <50 |
130 | 2.5 |
Blood taken at 0 min showed normal renal, liver, and thyroid biochemistry. Cortisol level was 735 nmol/L.
Time |
Glucose |
Insulin |
FFA |
---|---|---|---|
0 | 4.6 | 18 | 727 |
30 | 11.9 | 1170 | 133 |
60 | 9.4 | 560 | <50 |
90 | 3.2 | 134 | <50 |
120 | 2.2 | 45 | 73 |
150 | 3.8 | 16 | 638 |
This demonstrates the changes observed in glucose and insulin levelsduring the eGTTs performed on the two patients. It provides a visual understanding of the relationship between blood glucose and insulin levels in the two patients with different presentation patterns.
The results were discussed and an association between insulin levels and blood sugar was noted. The pattern observed appeared to fit a reactive pattern, albeit hypoglycaemia occurring early. It was considered that the raised insulin levels prevented the physiological counter-regulation; hence, the patient recovery was compromised. At this point, we decided to repeat the eGTT, but using half the glucose content to establish whether a different clinical, biochemical pattern and recovery took place. Although this modified eGTT was unconventional and may not be useful in determining the diagnosis, it was decided that it might yield important information when compared to the original eGTT. Further, it could be useful to see if smaller glucose load would lead to less severe symptoms. The results of the modified eGTT are presented in Table
Patient B, a 52-year-old man (weight: 118.2 kg; BMI: 38) had experienced severe dizzy spells a few hours following a meal. Although he was not diabetic, he was classified as having the metabolic syndrome (central weight distribution, hypertension, total cholesterol: 4.6 mmol/L; HDL-cholesterol: 0.8 mmol/L; triglycerides: 3.9 mmol/L). Due to the postprandial symptoms an eGTT was carried out and the relevant biochemistry test results are presented in Table
Biochemistry parameters for patient B following the eGTT with 113 mL of Polycal.
Time |
Glucose |
Insulin |
FFA |
---|---|---|---|
0 | 5.8 | ||
30 | 12.7 | 970 | 650 |
60 | 16.8 | 3320 | 430 |
90 | 13.6 | 4550 | 201 |
120 | 8.0 | 1710 | 119 |
150 | 6.9 | 65 | |
180 | 4.7 | 1360 | 86 |
210 | 3.5 | 920 | 124 |
We have presented 2 cases demonstrating differing patterns of postprandial reactive hypoglycaemia. Both patients experienced symptoms due to hypoglycaemia at different times following the glucose load. Patient A did not demonstrate any features of the metabolic syndrome in contrast to patient B. The blood glucose increased rapidly in both patients within 30 minutes; in patient A, it stopped increasing at that point, while in the insulin resistant patient B, it continued to rise. The plasma insulin levels also increased in both patients, albeit the patterns were different as seen in Figure
Reactive hypoglycaemia is a phenomenon that may be affected by exaggerated insulin release and insulin resistance [
The primary mechanism in Patient A, who did not demonstrate features of insulin resistance, could have been an exaggerated and rapid increase of insulin release. Rapid absorption of glucose could also have been a factor. Finally, factors influencing insulin and glucagon release, such as incretins and defects in the insulin release mechanism itself, must also be considered. Patient B, in contrast, demonstrated blood glucose and insulin levels during the eGTT more in agreement with insulin resistance.
Cellular entry of glucose is via the glucose transport (GLUT) proteins which are either insulin-independent or insulin-dependent ones [
A dysfunctional link between insulin and the targeting of GLUT 4 has been shown to contribute to insulin resistance [
Late reactive hypoglycaemias as part of the insulin resistance syndrome may thus be caused by delayed insulin secretion and thus delayed insertion of GLUT 4 [
Early reactive hypoglycaemia may be the function of an exaggerated incretin effect [
There was considerable benefit when both patients understood hypoglycaemia to be the probable reason for their symptoms. However, it was most important that management was tailored to the biochemical pattern observed during the eGTT. It was considered that both patients would benefit from smaller, more regular food intake consisting of carbohydrate of low glycaemic index (GI). Decreased insulin response, as it was hoped, would lead to symptomatic improvement. Currently, both patients have been advised accordingly and are experiencing some benefit.
Similarly, we would consider acarbose as second-line in both patients in order to decrease glucose absorption. We have previously started a patient demonstrating the late hypoglycaemic pattern on sitagliptin (starting at 25 mg and cautiously increasing the dose with patients consent) due to its possibly delaying gastric emptying. We recognise that increased endogenous GLP-1 following a gliptin could lead to even greater insulin levels worsening the symptoms; thus, careful monitoring of this patient was initiated. We would not consider this approach in patients exhibiting early postprandial hypoglycaemia as an exaggerated GLP-1 response which may be causative. We would have some concerns about metformin as it could possibly diminish the physiological response to hypoglycaemia, despite the potential to improve insulin sensitivity. Thus, the mainstay of treatment at present remains focused on reduced glycaemic load and a low GI diet and we will describe this further.
GI is a measure of the effects of carbohydrates on blood sugar levels. GI estimates how much each gram of consumed available carbohydrate (total carbohydrate minus fibre) raises an individual’s blood glucose level [
The rate of carbohydrate absorption after a meal, as quantified by GI, has significant effects on postprandial hormonal and metabolic responses. High GI meals produce an initial period of high blood glucose and insulin levels, followed in many individuals by reactive hypoglycaemia [
It has been demonstrated that a low GI meal resulted in significantly lower plasma glucose, serum insulin and plasma GLP-1 than the high GI meal [