Physiological effects of starvation interpreted as food allergyBMJ 1999; 319 doi: https://doi.org/10.1136/bmj.319.7205.304 (Published 31 July 1999) Cite this as: BMJ 1999;319:304
- a Regional Department of Immunology, Royal Victoria Infirmary, Newcastle upon Tyne NEI 3LP
- b Department of Immunology, Southmead Hospital, Bristol BS10 5NB
- Correspondence to: Dr Bethune
- Accepted 29 January 1999
Carbohydrate restriction can disturb glucose metabolism, and this may be interpreted falsely as food allergy
It is a common misconception that food allergy can cause a wide variety of symptoms. A population study in the English town of High Wycombe showed a perceived prevalence of food intolerance of almost 20%, but double blind food challenge (the definitive test) provided objective evidence of symptoms related to food in fewer than 2% of subjects.1 This popular belief is fuelled by extensive and often misleading reports on the subject, as well as the emergence of clinics offering unvalidated tests that claim to diagnose allergy.2
We describe three patients who undertook strict, low carbohydrate diets because of perceived allergy. Starvation may affect the response to an oral glucose challenge. It results in an initial excessive rise in the blood glucose concentration followed by hypoglycaemia, which can be accompanied by symptoms.3 4 All three of the patients described developed symptoms several hours after meals and attributed these to food allergy. Further restriction of carbohydrate intake exacerbated their problem Symptoms continued to occur after meals and were erroneously interpreted as further evidence of their carbohydrate allergy.
A 24 year old woman was referred with a 4 month history of diarrhoea and abdominal pain. A self prescribed wheat free diet had improved her symptoms at first, but as she restricted her diet further she began to complain of dizziness, weakness, nausea, and palpitations, which were most severe several hours after meals. She concluded that she was allergic to many foods, and cut down her intake further. The woman lost 5 kg in weight, and her body mass index fell to 16. She continued to have symptoms after any normal sized meals, and this reinforced her belief that she was allergic to several foods
Investigations excluded any important gastrointestinal abnormality, and the patient's symptoms were not reproduced by blinded food challenges However, a prolonged glucose tolerance test (50 g glucose taken orally) led to a reproduction of her symptoms and resulted in a blood (venous) glucose concentration of 1.8 mmol/l at 3.5 hours.
The patient ignored advice to reintroduce small amounts of carbohydrate regularly. By this stage she was convinced that she was allergic to all forms of carbohydrate, despite the lack of symptoms after blinded food challenge.
A 47 year old man was referred with a 3 year history of diarrhoea, dizziness, abdominal pain, myalgia, and weight loss of 7 kg. He had been investigated extensively by several doctors, had seen a psychiatrist, and had consulted numerous alternative therapists. He believed that he was allergic to carbohydrate and had adopted a low carbohydrate diet. This resulted in further weight loss; his body mass index fell to 17. The patient's symptoms of dizziness and weakness were reproduced by meals rich in carbohydrate, reinforcing his belief that food allergy was his problem.
A diagnosis of coeliac disease was made after the patient was found to be positive for IgA anti-endomysial antibodies, and this was confirmed by histological findings from a jejunal biopsy specimen. He was given advice about eating a gluten free diet. The patient combined the gluten free diet with restricted carbohydrates but found that the symptoms he had been experiencing after eating meals containing carbohydrate (gluten free) persisted. A prolonged oral glucose tolerance test resulted in symptomatic hypoglycaemia, and his venous glucose concentration was 1.5 mmol/l at 3 hours. The patient was advised to continue the gluten free diet and to incorporate carbohydrate regularly. He now eats a gluten free diet with normal amounts of carbohydrate; his symptoms have resolved and his weight has returned to normal.
A 44 year old man was referred with a 6 month history of fatigue, myalgia, cramp-like lower abdominal pain, and diarrhoea. An ultrasound scan showed that he had gall stones, and he underwent a cholecystectomy. As the patient's symptoms improved during the preoperative fast, he self diagnosed food allergy. He excluded all sugar and carbohydrates from his diet, but then developed dizziness, apathy, bloating of the abdomen, and nausea 3 to 4 hours after meals. The fact that these symptoms occurred after carbohydrate rich meals reinforced his belief that carbohydrate allergy was the cause of the problem and encouraged him to continue with the restricted diet. At the time of referral to the clinic he had lost 19 kg in weight in 1 year. An oral glucose tolerance test showed a rise in the blood glucose concentration to 12.7 mmol/l after 2 hours The test was not continued beyond this time. Such a pattern of impaired glucose tolerance is known to occur in diabetes and in normal patients on low carbohydrate diets.4 As follow up over 18 months has shown no progression to diabetes in this man, it is probable that the changes in glucose handling were caused by the low carbohydrate diet.
All three patients attributed their symptoms to carbohydrate allergy However, the most likely explanation is that their symptoms were caused by carbohydrate restriction, which led to symptomatic late hypoglycaemia. The patients described were on self imposed low carbohydrate diets, resulting in appreciable weight loss and abnormally low body mass indices. They believed that they were suffering from carbohydrate allergy as their symptoms occurred after carbohydrate rich meals, and they attempted to reduce their symptoms by further restricting their diets, leading to yet more weight loss. It is known that starvation diets lead to abnormalities of carbohydrate metabolism.4 Permutt et al showed that normal people who ate a carbohydrate restricted diet over 3 days developed reactive hypoglycaemia (measured by the prolonged oral glucose tolerance test), and that this was associated with symptoms in half of the group.3
In two cases the prolonged oral glucose tolerance test was used to show that symptoms could be reproduced by a glucose challenge. In cases 1 and 2, the symptoms occurred in conjunction with biochemical hypoglycaemia. The glucose intolerance in case 3 was probably associated with the delay in effective insulin secretion which results in hypoglycaemia and is seen in people on low carbohydrate diets and those with early diabetes.3 Carbohydrate restriction is known to cause abnormalities of glucose handling, and an abnormal rise in the blood glucose concentration after glucose challenge is well reported in people on low carbohydrate diets.4 Subsequent changes in insulin and fatty acid mobilisation may then result in reactive hypoglycaemia. In case 3, the glucose tolerance test was not continued beyond 2 hours, but the patient's symptoms were probably the result of disordered glucose handling caused by severe carbohydrate restriction.
Once patients have a fixed belief about a cause for their symptoms, it may be difficult to persuade them to entertain an alternative diagnosis. In case 1, negative results of blinded food challenges did not dissuade the patient from her belief that food allergy was the cause of her symptoms. Case 2 was more complicated as the initial symptoms described were due to undiagnosed coeliac disease. However, the gluten free diet did not relieve the patient's postprandial symptoms until it was combined with an adequate carbohydrate intake The patient in case 3 was initially thought to have early diabetes, but follow up over 18 months has shown no progression to diabetes, and it is thought that his restricted diet is the cause for his disturbed glucose handling.
When patients complain of food allergy it is important to take a detailed clinical history. The definitive investigation is the double blind food challenge. In individuals whose diets are severely restricted, symptoms that occur after meals may be caused by abnormal glucose metabolism secondary to the carbohydrate restriction itself. Showing that symptoms can be reproduced by an oral glucose tolerance test with documented abnormal glucose handling can sometimes help to break the cycle where patients believe that symptoms are relieved by dietary restriction when in fact the low carbohydrate diet is the cause of their symptoms.
We thank Mrs I Anderson and the dietetics department at Newcastle General Hospital for their help with this work.
Contributors: All authors were involved in the care of all three patients GPS will act as guarantor.