There are several possible mechanisms, probably all acting together. The brownish deposits on teeth and tongue are mostly caused by the fact that bacterial proteins are denatured when the bacterial cell membranes are destroyed by high concentrations of chlorhexidine and disulphide functions are reduced to thiol functions, which form dark-coloured complexes with the iron(III) ions of saliva.
Other discolourations could occur because monosaccharides dissolved in saliva, such as glucose and fructose, react with the amine functions of bacterial proteins, the so called Maillard reaction.
There is great individual variation in the degree of staining from person to person, this makes explanation more difficult as it may be caused by intrinsic factors, differences in extrinsic factors or both. No longer accepted theories of stain formation with chlorhexidine include breakdown of chlorhexidine in the oral cavity to form parachloraniline and also that chlorhexidine may reduce bacterial activity such that partly metabolised sugars were broken down and then degraded over time to produce brown-coloured compounds. Most recent debate has centred around three possible mechanisms.
Non-enzymatic browning reactions: Berk suggested that the protein and carbohydrate in the acquired pellicle could undergo a series of condensation and polymerisation reactions leading to discolouration of the acquired pellicle. Chlorhexidine may accelerate formation of the acquired pellicle and also catalyze steps in the Maillard reaction. Observation of furfurals, intermediate products in Maillard reactions, in brown-discoloured pellicle has leant support to the theory, but the evidence is inconclusive. Moreover, these authors did not consider at all the same staining phenomenon observed with the numerous other antiseptics.
The formation of the pigmented sulphides of iron and tin: this theory suggests that chlorhexidine denatures the acquired pellicle to expose sulphur radicals. The exposed radicals would then be able to react with the metal ions to form the metal sulphide. Warner and coworkers have shown increased levels of iron in chlorhexidine treated individuals compared with water controls, no evidence was shown for tin. They then went on to conclude that the chromophore was not a sulphide, but a sulphur containing organic compound and metal ion complex and that chlorhexidine promoted the deposition of sulphate proteins. However, somewhat anomalously although the amount of stain and iron levels were increased, the levels of sulphide were reduced. Studies in vitro have contradicted aspects of the metal sulphide/denaturation theory. For instance, dietary staining of chlorhexidine treated tooth substance and acrylic occurred in the absence of salivary pellicle. More importantly pellicle coated surfaces exposed to protein denaturants or chlorhexidine did not stain when subsequently exposed to salts of iron and tin. Staining of saliva-coated tooth and acrylic occurred only when the chlorhexidine treatment was followed by a dietary chromogen such as tea. This has to some extent been replicated in vivo, where reciprocal rinsing with chlorhexidine and iron sulphate produced no staining in volunteers who abstained from food and beverages. However, chlorhexidine or iron sulphate followed by tea rinse produced immediately the characteristic brown and black discolouration of the teeth and tongue reported for chlorhexidine and iron respectively.
Precipitation of dietary chromogens by chlorhexidine: Plaque inhibition is dependent upon adsorption of chlorhexidine onto the tooth surface. Davies et al. suggested that locally adsorped chlorhexidine complexed with ions from the oral environment and showed this in vitro with the colour produced between chlorhexidine and food dyes. Following this observation, in vitro and in vivo experiments showed that chlorhexidine and other antiseptics known to cause staining in vivo could bind dietary chromogens to surfaces to produce staining.
One objection to the dietary chromogen theory was that there is no known correlation between chromogenic staining and dietary consumption of beverages. However, tea and coffee and red wine are not the only drinks to contain chromogenic polyphenols capable of interacting with chlorhexidine or polyvalent metal ions. Thus, the fact that staining can be produced in rabbits and dogs, which do not usually imbibe human beverages, can be explained by the presence of other polyphenols within the diet which are able to interact with chlorhexidine. Nevertheless, it was of interest to note that Leonard et al. demonstrated that staining was exaggerated in beagle dogs when tea and coffee was provided in conjunction with chlorhexidine rinses.
Most evidence indicates that the likely cause of staining is the precipitation of anionic dietary chromogens onto adsorped cations. Thus, polyphenols found in dietary substances, being anionic, are able to react with cations adsorped to surfaces such as the cationic antiseptics or polyvalent metal ions to produce staining. The difference in the potential of various cationic antiseptics to produce staining in vivo can be explained by their differing substantivity, which is consistent with the dietary aetiology. The apparent individual variation in staining noted in particular with chlorhexidine is of interest. It is worthy of note that from the diet controlled studies this variation can not be explained solely as a difference in the quantities of chromogenic agents in any one individual's diet even though abstinence from tea, coffee or red wine virtually eliminates staining from everyone. Clearly differences do exist in the propensity of individuals to produce stain and it is worthy of further investigation as it would be relevant to the need to use cosmetic tooth whitening products. There is no evidence to show that chlorhexidine is any less effective in people with a low susceptibility to staining.
A Watts & M Addy: "Tooth discolouration and staining: a review of the literature", British Dental Journal volume 190, pages 309–316 (24 March 2001); PDF, DOI
L. G. Hjeljord, G. Rølla, P. Bonesvoll: Chlorhexidine-protein interactions. In: J Periodont Res. 8 (Suppl 12), 1973, S. 11–16. PMID 4269593
H. F. Gilbert: Molecular and Cellular Aspects of Thiol-Disulfide Exchange. In: Advances in Enzymology and Related Areas of Molecular Biology. 63, 1990, S. 69–172. doi:10.1002/9780470123096.ch2
S. K. Grandhee, V. M. Monnier: Mechanism of formation of the Maillard protein cross-link pentosidine. In: J Biol Chem. 266(18), 1991, S. 11649–11653. PMID 4269593.
These unwelcome effects might be reduced a bit:
Bernardi F, Pincelli MR, Carloni S, Gatto MR, Montebugnoli L (August 2004). "Chlorhexidine with an Anti Discoloration System. A comparative study". Int J Dent Hyg. 2 (3): 122–26. doi:10.1111/j.1601-5037.2004.00083.x. PMID 16451475.