SciELO - Scientific Electronic Library Online

vol.73 número5Prevalence and aetiological factors of maxillofacial trauma in a rural district hospital in the Eastern CapeDetecting chronic apical periodontitis for improved endodontic success índice de autoresíndice de assuntospesquisa de artigos
Home Pagelista alfabética de periódicos  

Serviços Personalizados



Links relacionados

  • Em processo de indexaçãoCitado por Google
  • Em processo de indexaçãoSimilares em Google


South African Dental Journal

versão On-line ISSN 0375-1562
versão impressa ISSN 0011-8516

S. Afr. dent. j. vol.73 no.5 Johannesburg Jun. 2018 



Pathology of Candida infection in oral HIV-associated Kaposi sarcoma: A descriptive study



S MeerI; A SibdaII

IBChD (UWC), MDent (Wits), FCPath (SA), Associate Professor, Head Clinical Unit, Department of Oral Pathology, Faculty of Health Sciences, University of the Witwatersrand, Johannesburg, South Africa
IIBDS, MSc Dent (Wits). Research student, Department of Oral Pathology, Faculty of Health Sciences, University of the Witwatersrand, Johannesburg, South Africa





AIMS AND OBJECTIVES: To determine the frequency and histomorphology of secondary Candida infection in oral HIV-associated Kaposi sarcoma (HIV-KS) and to describe the demographics of patients with oral HIV-KS with and without secondary Candida infection.
MATERIALS AND METHODS: Haematoxylin and eosin, and periodic acid-Schiff stains of 32 oral HIV- KS were examined
histologically for intensity and morphology of Candida colonisation, depth of invasion, number of organisms, epithelial reactions and inflammatory response. Depth of Candida invasion and severity of infection were correlated with CD4 T-cell counts of HIV-positive patients.
RESULTS: Forty-one percent of oral HIV-KS were secondarily infected with Candida (n=13). Intensity varied from an isolated single pseudohyphus to matted colonies. Whilst in most cases, organisms did not invade beyond the parakeratin layer, pseudohyphae extended into stratum spinosum in two cases, and a single case showed a pseudohyphus within the lamina propria. Two cases showed pseudohyphae in the pyogenic membrane. Neutrophilic permeation of epithelium, commonly associated with Candida infection was frequently present even in absence of Candida infection
CONCLUSION: Oral HIV-KS is commonly secondarily infected with large numbers of Candida organisms. Morphological characteristics of secondary Candida infection in surface epithelium of HIV-KS suggest an altered pathogenetic pathway. Further studies are indicated.




The surface epithelium in oral human immunodeficiency virus-associated Kaposi sarcoma (HIV-KS) frequently shows secondary Candida infection with varying degrees of tissue invasion. The pathogenesis of such opportunistic fungal infections probably differs in oral HIV-KS, other HIV-related oral disease, and infection of normal oral mucosa in HIV-positive patients.1 Differences in pathogenetic pathways may vary in histomorphology, frequency or intensity of infection. The aim was to determine the frequency and histomorphology of secondary Candida infection of surface epithelium in oral HIV-KS, and to highlight patient demographics in oral HIV-KS with and without secondary Candida infection of overlying epithelium.



Thirty-two oral HIV-KS diagnosed in the Department of Oral Pathology, University of the Witwatersrand, South Africa over a five-year period were analysed. Data included patient age, gender, site of lesion, HIV status and CD4 counts. Ethics clearance was granted by the Human Research Ethics Committee (Medical), University of the Witwatersrand, Johannesburg (M00/08/29; M08.03.25).

Haematoxylin and eosin (H&E) and periodic acid Schiff (PAS) stained 4μ sections of HIV-seropositive oral KS (HIV-KS) were histologically examined for the criteria listed in Table 1. This is primarily a descriptive histomorphologic study with data being of a descriptive nature without statistical comparison.




The 32 patients who presented with oral HIV-KS ranged in age from 20 to 63 years and were mainly in the fourth decade (M:F=1:1.7). The age of males was not statistically different from females (p>0.5). The site most affected was the palate (37.5%) followed by the tongue (31.5%) (Table 2).

Only 40.6% of oral HIV-KS showed secondary Candida infection with PAS positive pseudohyphae and yeast cells in the surface epithelium or pyogenic membrane. The remaining 59.4% showed no Candida infection. The age of patients with oral HIV-KS secondarily infected with Candida ranged from 21-63 years; with most cases in the fourth and third decades (M:F=1:1.2). The site most affected was the tongue (46.2%) followed by the palate (30.8%) (Table 2). The age range of the 19 HIV-positive patients with oral HIV-KS not secondarily infected with Candida was 20-63 years. Most cases occurred in the fourth decade, followed by the third, fifth and seventh decades in both males and females, with a M:F ratio of 1:2.2. The most common site was the palate (42.1%) followed by the tongue (10.5%) (Table 2).

Of the 13 oral HIV-KS positive for Candida, pseudohyphae penetrated the parakeratin layer only (46.2%) (Figures 1a,b,c), both the pyogenic membrane and parakeratin layer simultaneously (7.7%), the superficial epithelium (7.7%) and deep stratum spinosum (15.4%) (Figures 2a,b). In 15.4% of cases, Candida pseudohyphae were noted in the necrotic slough only (Figure 1d). There was a single exceptional case of severe Candida colonisation with pseudohyphae penetrating the lamina propria and HIV-KS tissue (Figure 2c).





Secondary Candida infection ranged from severe (46.2%) to moderate (23.10%) to mild (30.8%) (Figures 1c,d). Mild infection showed penetration into tissues by only single, isolated Candida organisms, usually in the superficial parakeratin layer (Figure 1e). Moderately infected cases showed greater numbers of Candida pseudohyphae and yeasts. The parakeratin layer was thicker, with desquamation, and organisms mainly in the desquamated keratin and superficial parakeratin layers (Figures 1b,c).

Severe infection showed matted colonies of Candida pseudohyphae and yeasts in both the desquamated keratin and parakeratin layers (Figure 1c), sometimes reaching the stratum spinosum (Figures 2a,b). The yeasts always remained on the surface.

Neutrophilic micro-abscesses in superficial parakeratin were present in 53.8% of cases, however no Candida organisms were seen in close association to the micro-abscesses in 46.2% of cases. Surprisingly, Candida pseudohyphae were present without micro-abscesses in 53.8% of cases (Figures 2d,e,f).

Inflammation beneath Candida infected areas in the lamina propria ranged from severe (7.7%) to moderate (23.1%) to mild (53.8%), with areas having no inflammation (15.4%) (Figure 2). Epithelial hyperplasia (Figure 1a) was noted in 61.5% of infected cases, usually with severe infection. Candida positive cases were characterised by hyperparakeratosis and hyperplasia of surface epithelium compared to non-infected cases. Pseudoepitheliomatous hyperplasia was present in 7.7% of cases with severe infection.

An inverse relationship was noted between the CD4 count in the 13 oral HIV-KS secondarily infected with Candida and the presence, intensity and depth of penetration of Candida infection (Table 3). Cases with low CD4 counts showed deeper penetration of Candida pseudohyphae (Table 3). The 19 oral HIV-KS with no Candida growth showed equally low CD4 counts as the 13 oral HIV-KS secondarily infected with Candida; 9 cases had CD4 counts of < 100 cells/mm3.




HIV/AIDS has reached epidemic proportions in South Africa, with a dramatic increase in KS frequency.2-4 In 21% of HIV-positive people with KS, the initial presentation is in the mouth, while in 71.5% the mouth is affected at some time during the course of the disease. The number of oral KS cases in the Department of Oral Pathology increased from 84 cases over a 29-year period (1973-2002) to 133 new cases during the five-year study period (2003-2007).5 There has however been a 70-80% reduction in the risk of developing KS in South Africa, where up to 48% of HIV-infected adults are positive for HHV8.6

This study showed a secondary Candida infection rate of oral HIV-KS of only 40.6%. Many histopathologists anecdotally believe that the secondary Candida infection rate of HIV-KS is much higher. There are no comparable studies of infection rate or histomorphology of Candida secondary infection of oral HIV-KS to which the results of this study can be compared.

When considering reasons for the relatively low infection rate, it must be realised that the surface epithelium covering oral KS or oral HIV-KS may be unique and not comparable with other HIV-related lesions or even with normal epithelium in an HIV-positive patient. Oral HIV-KS may well influence the surface epithelium by induction from cytokines, prostaglandins and genetic influences, all of which influence the oral epithelial response to C. albicans penetration.7,8 Oral HIV-KS cells may create a unique micro-milieu not reproducible in other situations.

One cannot infer that mechanisms by which secondary Candida infection of oral HIV-KS occur are the same as those of primary Candida penetration in HIV-positive and HIV-negative patients or in other situations of oral Candida infection, and more importantly in other mucosal sites. For example, whilst oropharyngeal candidiasis occurs commonly in HIV-positive women Candida infection is rarely seen in vaginal mucosa of the same cohort.7

Oral Candida infection rates in HIV-positive patients vary from 26.3% (India)9 and 38% (Tanzania) to 94% (Zaire), and range from 37.8% to 63% in South Africa.10 Our findings of secondary Candida infection rate in oral HIV-KS (40.6%) are consistent with previous reports.

In South Africa, though KS was regarded as occurring predominantly in males, greater female involvement is recorded.2 The KS incidence has doubled in men and increased 7-fold in women resulting in a M:F ratio decline from 7:1 (1988) to 2:1 (2001).2 A previous study in our Department showed M:F ratio of 1.3:1 in 81 oral HIV-KS during 1997-2003.5 This was suggested to be due to differences in mode of HIV transmission, which is predominantly heterosexual in South Africa.5 The current study showed a prominent female predominance in oral HIV-KS, consistent with the dramatic increase in HIV infection of women in this population.11 In Africa across all ages, more females than males are infected with HIV, which may explain the M:F ratio reversal in favour of females. This gender imbalance of HIV-infected persons in Africa is most marked amongst the 15-24 age group, where M:F ratio is 1:4.11 The decrease in M:F ratio of HIV-KS in Africa due to ART is unlikely as the decline occurred before ART availability and HIV-KS incidence has not changed significantly since ART introduction.11

HIV-KS is reported in all ages; mainly in the third and fourth decades of life.5 Our study reported a predominance in third and fourth decades with no difference in age between HIV-KS patients secondarily infected with Candida and those who were not. Oral KS most frequently affects the palate, gingiva and dorsum of tongue.5 Our study demonstrated a similar predilection of oral HIV-KS for the palate followed by the tongue in HIV-positive patients regardless of secondary Candida infection.

Oral HIV-KS may develop at any stage of HIV infection and especially when CD4 T-cell counts fall below 200 cells/mm.12 Our study confirmed that most oral HIV-KS occurred at CD4 counts <100 cells/mm3. There was no difference in CD4 counts between lesions infected and those not secondarily infected with C. albicans. This implies that CD4 cells are not a determining factor in the pathogenesis of secondary candidiasis in oral HIV-KS. This remains to be confirmed with a larger study sample. A similar study showed 75% of 130 HIV-KS had CD4 counts of <200 cells/mm3 confirming that low CD4 counts are not a prerequisite for HIV-KS development.13 Previous reports show that CD4 T-cell numbers are markedly reduced in the oral mucosa of HIV-positive patients with or without candidiasis, and that C. albicans specific peripheral CD4 T-cells are depleted with HIV disease progression and concurrent oral candidiasis.14

This study is the first to establish the frequency of secondary Candida infection in oral HIV-KS and to describe the histomorphology. Candida pseudohyphae were present in epithelium overlying oral HIV-KS. Severity of Candida infection varied from single isolated pseudohyphae to matted colonies of fungal organisms. Morphology of less severe Candida infections closely mimicked that of immune competent hosts whereas severe Candida infections in oral HIV-KS showed more numerous and deeper penetration of Candida organisms than that in immune competent hosts.

Our study confirmed that Candida infection regardless of severity, rarely reaches the lamina propria by penetration through epithelium. In the single case where the organism penetrated connective tissue and KS tissue, it entered through the necrotic ulcerated surface, and not through intact epithelium. The reasons for this are unclear but perhaps the more anaerobic conditions found deep in the epithelium limit the depth of infiltration by organisms.

Further, Candida is epitheliotrophic and depends on attachment and penetrative biological processes to infect host tissue.7 It is uncertain whether these characteristics are inherent in either or both the organism and the host epithelium. Invasion of pyogenic membrane shows that epithelial factors are not absolutely essential for organism growth but that under exceptional circumstances organisms are capable of living in the fibrinopurulent exudate.



Only less than half of the cases of oral HIV-KS were secondarily infected with Candida. This may be due to the fact that many oral HIV-KS are covered by a pyogenic membrane. Furthermore, Candida is strongly epitheliotrophic and Candida infection in the fibrinopurulent exudate overlying oral HIV-KS is unusual. The relatively low frequency of secondary Candida infection of oral HIV-KS, its deep tissue penetration, and its presence in the absence of inflammation requires further investigation.


The authors thank Ms Amina Kaskar for her excellent technical assistance.



1. Williams DW, Jordan RP, Wei XQ, et al. Interactions of Candida albicans with host epithelial surfaces. J Oral Microbiol. 2013; 5; doi:10.3402/jom.v5i0.22434.         [ Links ]

2. Sitas F, Newton R. Kaposi's sarcoma in South Africa. J Natl Cancer Inst Monogr. 2001; 28: 1-4.         [ Links ]

3. Rees CA, Keating EM, Lukolyo H, et al. Mapping the epidemiology of Kaposi sarcoma and Non-Hodgkin lymphoma among children in Sub-Saharan Africa: a review. Pediatr Blood Cancer 2016; 63(8): 1325-31.         [ Links ]

4. Meer S, Altini M. Cytomegalovirus co-infection in AIDS-associated oral Kaposi's sarcoma. Adv Dent Res. 2006; 19(1): 96-8.         [ Links ]

5. Lager I, Altini M, Coleman H, Ali H. Oral Kaposi's sarcoma: a clinicopathologic study from South Africa. Oral Surg Oral Med Oral Pathol Oral Radiol Endod. 2003; 96(6): 701-10.         [ Links ]

6. Bohlius J, Valeri F, Maskew M, et al. Kaposi's sarcoma in HIV-infected patients in South Africa: multicohort study in the antiretroviral therapy era. Int J Cancer 2014; 135(11): 2644-52.         [ Links ]

7. De Repentigny L, Lewandowski D, Jolicoeur P. Immunopatho-genesis of oropharyngeal candidiasis in human immunodeficiency virus infection. Clin Microbiol Rev. 2004; 17(4): 729-59.         [ Links ]

8. Jivan V, Meer S. Quantification of oral palatine Langerhans cells in HIV/AIDS associated oral Kaposi sarcoma with and without oral candidiasis. J Cancer Res Ther. 2016; 12(2): 705-11.         [ Links ]

9. SanadhyaYK, Sanadhya S, Nagarajappa R, Jain S, Aapaliya P, Sharma N. Correlation between oral lesions and opportunistic infections among human immunodeficiency virus-infected individuals in Indian population. Int Marit Health 2014; 65(3): 124-30.         [ Links ]

10. Naidoo S, Chikte U. HIV/AIDS - the evolving pandemic and its impact on oral health in sub-Saharan Africa. SADJ. 1999;.54(12): 616-30.         [ Links ]

11. UN Joint Programme on HIV/AIDS (UNAIDS). Global AIDS Update - 2016, [accessed 24 September 2017]        [ Links ]

12. Wojcicki JM, Newton R, Urban MI, et al. Risk factors for high anti-HHV-8 antibody titres (> or =1:51,200) in black, HIV-1 negative South African cancer patients: a case control study. BMC Infect Dis. 2003; 3: 21-5.         [ Links ]

13. Jung AC, Paauw DS. Diagnosing HIV-related disease : using the CD4 count as a guide. J Gen Intern Med. 1998; 13(2): 131-6.         [ Links ]

14. Kunkl A, Mortara L, Valle MT, et al. Recognition of antigenic clusters of Candida albicans by T lymphocytes from human immunodeficiency virus-infected persons. J Infect Dis. 1998; 178(2): 488-96.         [ Links ]



Shabnum Meer
Associate Professor, Department of Oral Patholog
Faculty of Health Sciences, Private Bag 3
University of the Witwatersrand
Johannesburg, South Africa, 2050
Tel: +27 (0)11 717 2523. Fax: +27 (0)11 717 2146
E-mail: or

Creative Commons License Todo o conteúdo deste periódico, exceto onde está identificado, está licenciado sob uma Licença Creative Commons