Do Yeasts Play Any Part in `Candida' (Fungal-type Dysbiosis)?: Discussion Paper


Gut dysbiotic symptoms with a characteristic clinical pattern have been known for many years. In the 1970s Truss advanced the theory that the syndrome is caused by Candida albicans. This has generated much medical controversy. In some centres the attribution has caused rejection of the syndrome itself. A number of studies have demonstrated organic changes in sufferers of what the author prefers to call fungal-type dysbiosis; however, over the past twenty years convincing data have not been published identifying causal yeasts. There is a further problem with a name involving Candida in that it may cause confusion with established syndromes known to be associated with that organism. This paper will review the literature and advance an alternative hypothesis that the syndrome could have a bacterial rather than a yeast cause.

Keywords: gut dysbiosis, Candida albicans, yeasts, bacteria.


Medical progress relies not only on new discoveries, but on advancing and testing new explanations for conditions which have previously been reported. In 1978 Truss [ 1] reported a condition, postulated its cause and set out a treatment programme. At the time it was thought to be a new condition, but it had previously been reported, and had fallen from favour [ 2, 3]. Problems arose because many accepted Truss's theory as if it had been proven, with a consequent adverse reaction from academic physicians who, in rejecting the explanation, also denied the existence of the syndrome [ 4].

The condition is that currently termed fungal-type dysbiosis. It will be described below. Evidence that it is an organic entity will be reviewed. As a result of recent progress the attribution to Candida albicans will be criticized. It is suggested that any name based on a Candidal cause should be abandoned, both as likely to delay acceptance and as contrary to the evidence. Indeed, even the concept that yeasts (unspecified) have any role in the condition is unproven and may be untrue.

Further problems which may arise from confusion with established diseases known to be due to Candidal infection will be discussed below.


There have been many accounts of this condition: in spite of this it remains difficult to summarize with brevity, which is one of the major bars to its acceptance. Clinically it is not a clear entity and cannot be separated by history from food intolerance [ 5]. Its only current justification is its good therapeutic response to a specific treatment programme [ 5]. This is difficult academically and will only be credible if supported by good research and a rigorous scientific attitude by all who work on it.

Patients display all or some of the following: symptoms of irritable bowel (bloating, abdominal pain, increased wind and constipation/diarrhoea), catarrhal symptoms and psychoneurological symptoms (minimal brain dysfunction) [ 5].

Its remarkable feature is the clearance of symptoms on a diet low in fermentable, yeasty and mould-containing foods, with or without antifungal drugs [ 6]. The effects of this on patients who have, in many cases, been unwell for many years and have had a number of other unsuccessful interventions are often dramatic. Such results are persuasive to those physicians who have tried them. For those who have not, and who require harder data before doing so, they can be dismissed as anecdotal. This is the second bar to general acceptance of the syndrome: its most impressive feature is clinical, and only seen by those who have accepted that it does exist and should be treated.


From the above it is clear that, without hard evidence, fungal-type dysbiosis must dwell in the shadowy lands of unorthodoxy: in such territory most physicians suspect psychosomatic disease. A small number of workers have contributed studies which provide data on physical changes and justify a conclusion that the problem has an organic nature. A seminal finding was that untreated patients, fasted and not consuming alcohol, produce excess blood ethanol after an intragastric enteric coated glucose challenge [ 7]. The amounts seen are of the order of 22-350 Mu mol L-1 in blood. The normal limit has been set at 22 Mu mol L-1, but it is probable that healthy individuals have a level of 0. Forensic levels in drink-driving cases are many orders of magnitude greater. The levels seen in fungal-type dysbiosis are compatible with a yeast/fungal fermentation phenomenon and this is how they were interpreted. It must be noted that ethanol production is not universal in these patients. Most physicians who treat the syndrome know of subjects who respond positively to diet and/or antifungals, but whose ethanol test is negative. However, the only publication to this effect shows that some sufferers produce excess urinary B-alanine excretion, normally regarded as a product of bacterial fermentation [ 8]. The author has used ethanol positivity as the criterion for inclusion in a number of studies of aspects of the condition. Patients were found to have low levels of B vitamins, zinc and magnesium [ 9]: similar findings have been recorded by Galland, who also found lowered essential fatty acids [ 10]. Increased gut permeability seems to be common, but although statistically significantly different from normal the levels were found to be the same as those in food intolerance [ 11]. In this study it was noted that over a two-year period patients with either fungal-type dysbiosis or food intolerance, with the alternative diagnosis having been excluded, could subsequently acquire the other condition. More recently our group have shown that urinary histidine excretion is equally lowered in classical atopic (type A) allergy, food intolerance and fungal-type dysbiosis [ 12]. This finding is likely to be due to common allergic mechanisms.

Recently ethanolic fermentation has been compared with breath hydrogen measurements [ 13]. Two groups were involved, one ethanol positive and the other ethanol negative. If the ethanol is generated by yeast/fungal activity rather than bacterial, the ethanol positive group should not produce excess hydrogen as this is not a fermentation product for yeasts. The ethanol negative group, presumed to be due to bacterial dysbiosis, should, however, show hydrogen excess. The results were that both groups show about 2/3 of subjects with hydrogen excess, with no statistical difference between them [ 13].

We also have data on bacterial/yeast cultures on a 58-year-old female patient who had a total colectomy for an unrelated condition 41 years ago and has had fungal-type dysbiosis for 12 years. This enables us to sample ileal contents directly. The results, shown in Table 1, do not reveal any positivity for yeasts or pathogenic bacteria in 3 cultures, one of which was taken at the time of an exacerbation of the fungal-type dysbiosis. In addition, 2 routine stool cultures were done by another laboratory which would not have identified any yeast/fungal overgrowth; both were reported as normal. It is germane to note that the patient, whose symptoms of fungal-type dysbiosis are typical of the condition, remarks that the seat of this condition cannot be colonic as she does not have a colon.


The unexpected findings of these latest studies have made us reconsider what we know about the cause of fungal-type dysbiosis. It is easy to become attached to a plausible disease model and continue to support it, even when new evidence renders it less tenable, or even indefensible. We must remember that one single fact which is reliably established and contradicts an hypothesis must mean that the whole edifice crumbles. Many of us in environmental medicine have had occasion to criticize academic colleagues for their unwillingness to accept data from our field that disproves their practices. Here we may have a situation where we have to discard our own postulates when they do not stand up to inspection. It will be a measure of the maturity or otherwise of our discipline whether we accept or fail this challenge.

Is there any evidence of Candidal involvement? To answer this we must decide what data are acceptable. A recent project attempted to compare the past and present prevalences of classic fungal diseases in patients with fungal-type dysbiosis and in the general population [ 14]. No reliable population figures could be found: consequently any clinical suspicion of concurrence remains anecdotal. The core of any such attribution should be microbiological studies. Here the major problem is that the clinical pattern of fungal-type dysbiosis suggests that it is a small bowel condition. Attempts to justify a Candidal cause by stool cultures [ 15] must be rejected as irrelevant unless it can be shown that large bowel Candidal overgrowth cannot occur. In view of known colonic colonization with Candida [ 16] stool cultures cannot currently justify a causal attribution. Equally, antibody studies do not provide reliable justification as Candida antibody positivity in healthy subjects as an immunological marker of normal response to environmental exposure has been well known since the 1960s [ 17]. Only samples obtained directly from the small bowel can reflect activity therein. Under normal circumstances such material is difficult and invasive to harvest and likely to require active cooperation from a sympathetic gastroenterologist and the services of a skilled and motivated microbiologist. Currently, the only direct ileal samples that have been cultured known to us are ours, reported here. These are from a single patient: more data are clearly desirable. Our patient showed no yeast growth on 3 cultures. Had it been present we believe that our methods would have shown a positive result.

None of these findings proves that Candida albicans is not involved in the condition. However, there is no evidence that it is: until the position changes there is no justification for either a causal attribution or for using any name which implies that there is. Does the evidence justify a yeast involvement at all? Clinicians report improvement on diets low in (yeast) fermentable, yeasty and mould-containing foods, with or without antifungal drugs [ 5, 14]. Details of the implications of these findings are discussed elsewhere [ 12, 13]: there are other explanations for all these phenomena which need not involve yeast activity. Evidence of other preceding or concurrent yeast conditions remains anecdotal [ 14]. The finding of elevated hydrogen production and B-alanine excretion points to bacterial and not yeast involvement [ 7, 13]. Only the high levels of ethanol are more characteristic of yeasts than bacteria [ 18]. Even here bacteria such as Clostridia, common gut pathogens, are ethanol fermenters [ 19], although generally in modest amounts. However, food spoilage organisms such as Zymomonas mobilis [ 20] are capable of producing the amounts seen in this syndrome.


Candida albicans is well known in certain sites as a pathogen, so much so that it would be invidious to quote references to its causal relationship with oral or vaginal thrush. True Candidal allergy has also been documented [ 21]. Intestinal Candidiasis is generally seen in immune compromised patients, where it generally responds to a week's therapy with an antifungal drug. No dietary measures are required [ 22]. By contrast in one trial, albeit of poor methodology, nystatin without diet had no significant benefit in fungal-type dysbiosis [ 23].

We cause further needless and inappropriate confusion with these established diseases by invoking Candida in fungal-type dysbiosis where no evidence exists.


The evidence does not at present support a role for Candida albicans in fungal-type dysbiosis. It may even be plausibly argued that there is insufficient data to support any yeast/fungal attribution, although common sense still suggests that there is some link. However, the use of any name based on a Candidal cause for the syndrome is at present unjustified, offensive to many, and should be abandoned. To continue to use such terms places us on a level with those who have refused to accept the well-documented advances in allergy and nutritional and environmental medicine [ 24].


The author is grateful to Professor Richard Lacey and colleagues at the Parascope Laboratory, Department of Microbiology, Chapel Allerton Hospital, Chapel Allerton, Leeds LS7 4SA, UK for performing the ileal cultures, and to his patient for her unstinting cooperation.

TABLE 1. Patient ileostomy sample results
Legend for Chart:

A - Date
B - Culture for bacterial pathogens
C - Comment on enteroflora
D - Microscopy
E - Culture for yeasts



02 Sept 1991

Negative No imbalance Vegetable fibres + + Negative
Starch granules + +

14 May 1997

Negative No imbalance No abnormalities Negative

25 July 1998

Negative No imbalance No abnormalities Negative
[1] Truss CO. Tissue injury induced by Candida albicans: mental and neurological manifestations. J Orthomolecular Psychiatry 1978; 7: 17-37.

[2] Hurst AF, Knott FA. Intestinal Carbohydrate Dyspepsia. QJ Med 1930-31; 24: 171-80.

[3] Berk JE. Intestinal carbohydrate dyspepsia. In: HL Bockus (ed.) Gastroenterology, vol. 2, 1st Erin. Saunders, Philadelphia. 1944; 248-57.

[4] Hunter J. Food allergy and intolerance. Prescriber's J 1997; 37: 193-8.

[5] Anthony H, Birtwistle S, Eaton K, Maberly J. Environmental Medicine in Clinical Practice. BSAENM Publications. Southampton. 1997: 141-57.

[6] Eaton KK. Gut fermentation: a reappraisal of an old clinical condition with diagnostic tests and management: discussion paper. J R Soc Med 1991; 84: 669-71.

[7] Hunnisett A, Howard J, Davies S. Gut fermentation (or the "autobrewery" syndrome): a clinical test with initial observations and discussion. J Nutr Med 1990; 1: 33-8.

[8] Eaton KK, Howard M, Hunnisett A. Urinary B-alanine is a marker of abnormal as well as normal gut fermentation. J Nutr Med 1994; 4: 157--63.

[9] Eaton KK, McLaren-Howard J, Hunnisett A, Harris M. Abnormal gut fermentation: laboratory studies reveal deficiencies of B vitamins, Zinc and magnesium. J Nutr Biochem 1993; 4: 636-8.

[10] Galland L. Nutrition and candidiasis. J Orthomolecular Psychiatry 1984; 13: 50--60.

[11] Eaton KK, Howard M, McLaren-Howard JM. Gut permeability measured by polyethylene glycol absorption in abnormal gut fermentation as compared with food intolerance. J R Soc Med 1995; 88: 63-6.

[12] Eaton KK, Howard M, Hunnisett A. Urinary histidine excretion in patients with classical allergy (type A allergy), food intolerance (type B allergy) and fungal-type dysbiosis. J Nutr Biochem 1998; 9: 586-90.

[13] Eaton KK, Chan R, Howard M, McLaren-Howard J. A comparison of lactulose breath hydrogen measurements with gut fermentation profiles in patients with fungal-type dysbiosis. Paper submitted for publication, 1998.

[14] Eaton KK, Howard M. Fungal-type dysbiosis of the gut: the occurrence of fungal diseases and the response to challenge with yeasty and mould-containing foods. J Nutr Environ Med 1998; 8: 247-55.

[15] Hauss R. Gastrointestinal mycoses: insights and clinical pearls from the German perspective. 31st Annual Meeting Am Acad Environ Med 1996 (abstracts): 281-5.

[16] Al-Doory Y. Laboratory Medical Mycology. Lea and Febiger. Philadelphia. 1980: 212.

[17] Longbottom JL, Murray IG, Pepys J. Diagnosis of fungal diseases. In: Clinical Aspects of immunology. PGH Gell, RRA Coombs, (eds.). Blackwell, Oxford. 1968 (2nd edn.): 80-4.

[18] James SA. National Collection of Yeast Cultures. Norwich. Personal Communication, 1998.

[19] Balows A, Truper HG, Dworkin M, Harder W, Schliefer K-H (eds.). The Prokaryotes. A handbook on the Biology of Bacteria: Isolation, Identification, Applications. Springer-Verlag, New York. 1981 (2nd edn.), vols 2 and 3: 1820, 1824-5, 1836, 2290 and 1909.

[20] Green PN. Curator, National Collection of Industrial and Marine Bacteria, Aberdeen. Personal Comunication, 1998.

[21] Holti G. Candida allergy. In: Symposium on Candida infections. HI Winner, R Hurley (eds.). Livingstone, Edinburgh. 1966; 84: 74-81.

[22] Carpenter CCJ. Infectious Diseases: Candidiasis. In: Cecil Textbook of Medicine. JB Wyngaarden, LH Smith (eds.): Saunders, Philadelphia. 1985 (17th edn.) 1768-70.

[23] Dismukes JE, Wade JS, Lee JY et al. A randomised double blind controlled trial of nystatin therapy for the candidiasis sensitivity syndrome. N Engl J Med 1090; 323: 1707-23.

[24] Davies S. Nutritional flat earthers. Editorial. J Nutr Med 1990; 1: 167-70.


By K. K. Eaton, LRCPE, LRCSE, LRFPSG, The Princess Margaret Hospital, Osborne Road, Windsor, Berks SL4 3SJ, UK


Share this with your friends