Microbiology of LGE has not been fully defined. However, studies with microbiological analysis of LGE and periodontal disease in HIV/AIDS patients in general should be interpreted with caution. The major problem of arriving at a conclusion for the associated microbiota with LGE is the method of sampling. Various sampling methods have been used in the studies that examine the microbiological aetiology of LGE. The studies that considered LGE as a variant of periodontal disease have commonly examined dental plaque samples obtained from the site of the lesions to correlate the pathogenic microbiota with LGE. On the other hand, studies looking into the oral candidiasis have taken samples such as gingival scraping, mouth rinse or saliva. This should be kept in mind when correlating the microbiological data derived from different studies with LGE. In addition, definition of the clinical picture of “LGE” is somewhat ambiguous among studies resulting in contradictory reports in the literature. Moreover, some have used pooled samples from different sites with no clear demarcation of supra or subgingival plaque whereas others have used a site-specific sampling. Culturing methods and identification techniques vary among studies. All foregoing factors preclude arriving at a firm conclusion of the microbiological aetiology of LGE.

Composition of the gingival microbiota in HIV patients and comparative healthy counterparts has also been investigated, although the results are diverse and somewhat contradictory in nature. Some report that HIV-positive and healthy subjects have similar microbial profiles whilst others show the opposite. Aas et al 2007 examined the subgingival plaque microbiota of HIV-positive patients in order to describe and compare the predominant bacterial and fungal species associated with gingivitis, periodontitis and LGE using 16S and 18S rDNA sequencing (Aas et al. 2007). Bacterial profile of the LGE was quite different from other disease entities. Saccharomyces cerevisiae was the only fungal species detected in the LGE samples. In contrast, periodontitis patients had predominant Candida albicans. Association of classical “red-complex” periodontal pathogens such as Porphyromonas gingivalis, Tannerella forsythia and Treponema denticola with periodontal diseases in HIV patients including LGE was not as strong as in the healthy counterparts. Interestingly, while classical periodontal pathogens were sparse, other unconventional species like Gamella, Dialister, Streptococcus and Veillonella were predominant in HIV patients. Hence, authors suggested that unconventional bacterial species rather than classical periodontal pathogens are involved in the periodontal disease of the subjects with HIV (Aas et al. 2007). Another study compared the composition of subgingival microbiota of HIV-seropositive patients under HAART therapy and HIV-seronegative subjects with chronic periodontitis using DNA probes and checkerboard assays (Goncalves et al. 2007). They found significant association of Enterococcus faecalis and Acinetobacter baumannii with the periodontal disease of HIV patients corroborating the foregoing idea that classical periodontal pathogens are not usually detected in the periodontitis of HIV patients. Early studies on LGE and NPD have also suggested an association with viral pathogens. Viruses such as Cytomegalovirus, Epstein–Barr virus, human herpes viruses and papillomavirus were proposed to play a role in the initiation and progression of LGE and NPD (Umadevi et al. 2006). However, these claims have not been backed up by substantial data to prove such as an association. HIV-infected patients are likely to develop infections of concurrent viral, bacterial and fungal infections. This holds true for the HIV-associated periodontal disease too. Therefore, it is possible that multiple pathogens contribute to the development of LGE and NUG/NUP. Presence of these uncommon bacterial species may be a reflection of the immune-compromised status of the patients that lead to colonization of atypical pathogens.

The famous EC-Clearinghouse classification of the oral lesions associated with HIV infection in 1993 (Table 12.1) categorized LGE as a separate disease entity under Group 1 oral diseases, i.e. lesions most commonly associated with HIV infection (Axéll et al. 1993; van der Waal 1997). Additionally, few variants of oral candidiasis such as pseudomembranous candidiasis, erythematous candidiasis and angular chelitis were also categorized under Group 1 (Table 12.1). Due to the omission of LGE as a candidiasis variant in the EC-Clearinghouse classification, most of the subsequent studies on oral candidiasis of HIV subjects had not examined LGE.

Later studies showed that LGE could be associated with Candida infection. It is noteworthy that almost all HIV subjects have a heavy candidal carriage, which does not necessarily imply they suffer from Candida infections. One study evaluated subgingival colonization of Candida between HIV-positive and HIV-negative subjects taking oral mucosal samples and subgingival samples (Lamster et al. 1998). Identification and characterization of Candida was performed by culture-dependent enumeration as well as DNA fingerprinting. Higher percentage of HIV-positive subjects had Candida in the subgingival samples and Candida titre tends to be higher among them compared to HIV-negative subjects. Interestingly, subgingival Candida had a unique DNA pattern different from mucosal Candida indicating that gingival Candida colonizers are not a contamination from mucosal colonizers. Hence, Lamster et al. (1998) suggested presence of Candida in the subgingival plaque of the HIV patients may have an aetiological contribution to the development of LGE lesion. There are other studies showing higher Candida carriage more than 50 % in the subgingival plaque samples of HIV-infected patients suffering from HIV gingivitis (Zambon et al. 1990).

Grbic et al. (1995) examined the periodontal status of a cohort consisted of HIV-infected homosexual men and parenteral drug users with seronegative counterparts (Grbic et al. 1995). In addition, three variants of oral candidiasis, namely, pseudomembranous candidiasis, angular chelitis and erythematous candidiasis were also examined. Diagnosis of erythematous candidiasis was made from a smear taken from the erythematous area. Diagnosis criterion was presence of more than 10 hyphae of the lesion in Periodic Acid Schiff (PAS) stain. Interestingly, oral candidiasis as well as LGE was more prevalent in the subjects with decreased CD4+ lymphocytes. Hence, 42.9 % of the HIV-positive homosexual men having oral candidiasis had LGE whereas only 12.7 % of the subjects without candidiasis had LGE. Data from this study suggested a high likelihood of developing LGE in HIV-seropositive homosexuals if they have oral candidiasis. More interestingly, it was revealed, if the HIV patients had been treated with antifungals in the last 6 months, they are less likely to develop LGE, furnishing more evidence of candidal association with LGE.

Velegraki et al. (1999) demonstrated that LGE could be of candidal origin (Velegraki et al. 1999). By using CHORMagar and API32C methods, they identified Candida albicans in three paediatric HIV subjects and C. dubliniensis in one patient. Interesting, all lesions were healed upon treatment with antifungals showing a causal nature of Candida to LGE. Portela et al. in a series of studies on LGE demonstrated the possibility of Candida as the aetiological agent of LGE in HIV-infected patients. Subgingival plaque samples of HIV-infected children contained 42.3 % Candida species compared to 7.1 % HIV-negative children (Portela et al. 2004). C. albicans was the most commonly recovered species, but other non-albicans species such as Candida dubliniensis, Candida glabrata and Candida tropicalis and mixed species were also present in the subgingival plaque samples.

Another case report series evaluated the association of Candida with LGE in paediatric HIV patients, which were resistant to conventional plaque-removal therapy (Portela et al. 2012). They obtained the samples by frictionating a sterilized microbrush on the lesion (LGE) for mycological analysis. Subsequently, speciation of Candida was performed by CHROMagar Candida® and API 20C identification system demonstrating the presence of Candida in LGE lesions. All LGE lesions were successfully healed when treated with topical antifungal agents. This investigation provided strong evidence for the fungal aetiology of LGE in HIV-infected children. In another study of this group, Candida species such as C. albicans, C. tropicalis and C. parapsilosis were isolated from the saliva of HIV-infected children and upon the treatment with chlorhexidine gel the salivary candidal counts were significantly reduced, with the concomitant improvement of gingivitis (Machado et al. 2011). On the other hand, a study from Thai HIV-infected patients suggested that high prevalence of Candida in the periodontal pockets may be due to higher carriage of Candida in the saliva (Samaranayake et al. 2002). Serine proteases produced by the Candida isolated derived from LGE lesions in HIV patients have also been examined (Portela et al. 2010). It was found that SAPs secreted by Candida are able to cleave various components of host immune response and extracellular matrix proteins which facilitate the dissemination of Candida into deep organs protecting from the host immune response. In addition, histopathological studies on LGE lesions have observed Candida is able to penetrate the gingival epithelium in order to gain access to underlying soft tissues (Odden et al. 1994; Gomez et al. 1995). Therefore, there is substantial evidence to suggest a strong association of Candida with the LGE lesion seen in HIV-infected subjects.

HIV-infected patients are medically complex subjects. Therefore, prevalence of an oral manifestation like LGE may have many confounding factors. Patients’ data such as age, status of HIV infection, oral hygiene status, other medical problems, use of medication such as HAART and antimicrobials may significantly affect the presence of oral diseases including LGE in HIV-infected patients. In addition, definition of the clinical diagnosis, experience of the clinician, whether the diagnosis was purely of clinical basis or accompanied with microbiological and histopathological assays all affect the final diagnosis of the disease. Even in the post-HAART era, different institutions use various combinations of the anti-retroviral drugs; hence, these data cannot be directly compared. Unfortunately not all studies have clearly mentioned that information in their study cohorts. In addition, most of the studies in the literature on LGE are cross-sectional in nature. This paucity essentially precludes a proper comparison of the disease and data on the prevalence must be analysed with parsimony. Nevertheless, prevalence of the LGE in these past studies provides a general trend of the LGE in HIV patients (Table 12.2).