Article Text
PDF
Abstract
Aims To describe the clinical characteristics of ocular surface squamous neoplasia (OSSN) in a sub-Saharan referral hospital setting according to histopathological diagnosis and HIV status.
Methods All patients were enrolled who presented consecutively to the Kilimanjaro Christian Medical College eye department with lesions suspected to be OSSN from September 2005 to May 2007 and from February 2008 to September 2008. Clinical characteristics were documented on a standardised form, excision biopsies were performed and histopathological diagnosis was obtained on all cases. Data were analysed to look for associations among various factors.
Results 150 patients were enrolled. Histopathological study showed OSSN in 88% of cases. Of these, 128 (85.6%) were under the age of 50 years and 60% were HIV positive. The median CD4 cell count was 71 cells/μl among HIV-positive cases. Independent of size, the lesions of patients who were HIV positive were more likely to be higher grade malignancy than those who were HIV negative.
Conclusion In a sub-Saharan setting, OSSN occurs in persons who are younger than in industrialised countries and is often associated with HIV positivity. CD4 cell counts indicate that a majority of HIV-positive patients with OSSN are significantly immunosuppressed at presentation. Higher grade malignancy in this group could indicate a more aggressive course.
- CD4
- HIV
- neoplasia
- OSSN
- Tanzania
Statistics from Altmetric.com
Ocular surface squamous neoplasias (OSSN) represent a spectrum of clinical conditions ranging in severity from mild squamous dysplasia to invasive squamous cell carcinoma.1 2 Squamous cell carcinoma is the most frequently encountered malignant tumour of the conjunctiva.3 The precise aetiopathogenesis is unknown but heavy solar ultraviolet radiation, HIV and human papilloma virus infections are postulated risk factors.1–11 Morbidity is mainly related to local invasion of the conjunctiva and cornea. Ocular penetration, regional spread and distant metastasis are rare, but possible.12 The most common form of treatment is surgical excision, recommended with 4 mm tumour-free margins.12 13 This is also the standard treatment for ocular surface squamous neoplasia in Tanzania. The aim of this study was to learn more about the characteristics of OSSN in a clinic situation in sub-Saharan Africa. In particular, we wanted to know if there are clinical characteristics associated with the degree of malignancy, whether OSSN among those who are HIV positive differs clinically from those who are HIV negative, and whether the degree of immunosuppression among those who are HIV positive is associated with any clinical characteristics. Secondarily, we wanted to document the recurrence rate at 1 year in this typical sub-Saharan African referral hospital system.
Materials and methods
This was a prospective study, approved by the Kilimanjaro Christian Medical College Ethical Committee. All patients presenting to Kilimanjaro Christian Medical Centre (KCMC) Ophthalmology Department, with lesions suspected to be OSSN, were enrolled. An informed consent was provided by each enrolled patient. The time frame of the study was a total of 39 months from September 2005 to May 2007 then February 2008 to September 2008 with a break of 8 months from June 2007 to January 2008. All patients underwent history taking and a complete eye examination supervised by one or two ophthalmologists (IIM or WUM) in the eye clinic. Standardised examination forms were completed by the medical officer in attendance. HIV testing and counselling were provided by the nurse coordinator of the study. Patients underwent excision biopsy with an operating microscope. The surgical technique was left at the discretion of the surgeon. All specimens were examined histopathologically, 66% by two pathologists (BS and VAW). Cases in which there was disagreement were re-examined and a final diagnosis was assigned. Based on these diagnoses cases were grouped into three categories (mild/moderate dysplasia, severe dysplasia, carcinoma). Individuals who were positive for HIV were sent to the KCMC infectious disease clinic for measuring the CD4 cell count and consideration for starting on antiretroviral therapy. Following surgery, the patients were told to come back at regular intervals and were contacted by phone if they did not turn up on the follow-up date. The bus fare for follow-up, the cost of histological examinations and investigations at the infectious disease clinic were covered by the study. Data were entered from the standard forms into Excel then imported into STATA v.11 for analysis. Appropriate statistical tests were performed to look for associations as determined by the main aims of the study.
Results
One hundred and fifty patients were enrolled, of whom 48 (32%) were men and 102 (68%) were women. Ages ranged from 18 and 86 years with a mean of 38.7 (SD 12.1) and a median of 36. Histological examination showed that 132 patients (88%) had lesions that were dysplastic or neoplastic (OSSN) and 18 patients (12%) had non-malignant lesions. Associations between histopathological grade and other factors are shown in table 1 for OSSN lesions.
Associations between histopathological grade and other factors
A regression model (analysis of variance) with covariables HIV status, age and size showed that HIV status (p=0.035) and size (p=0.002) were both independently associated with malignancy grade. Out of the 132 OSSN cases, 79 (59.8%) patients were positive for HIV. Fifty-two (39.4%) were negative and in one case (0.8%) the HIV test was not performed. Associations between HIV status and clinical factors are shown in table 2, which shows that longer duration, larger lesions, leukoplakia, presence of a feeder vessel and recurrence after surgery were all associated with positive HIV status.
Associations between HIV status and other factors
A logistic regression model including duration and size of lesion, leukoplakia and feeder vessels revealed that larger size, the presence of leukoplakia or a feeder vessel all independently predicted HIV status as positive.
CD4 cell counts were determined for 76 HIV-positive patients. Values ranged from 5 to 394 cells/μl with a mean of 103.27 (SD 88) and a median of 71. In 65 patients (85%) the CD4 cell count was lower than 200 cells/μl. Eleven patients (15%) had CD4 cell counts equal to or higher than 200 cells/μl. Among the 132 patients with OSSN 22 patients (16.6%) had recurrence within a year.
Discussion
We included in this study patients with conjunctival lesions suspicious of OSSN. Eighty-eight per cent of the suspicious cases turned out to be OSSN on histological examination. This agrees with findings reported by McKelvie et al,14 but is higher than other reports of accurate diagnosis of 30%1 and 41%.12 Among the patients with OSSN, 85.6% were less than 50 years old, which is similar to other reports from Africa,3 7 15 but in marked contrast to North American and European studies.12 16 17
Sixty per cent of the patients with malignant lesions in our study were HIV positive. This is slightly lower than reports from Uganda and Malawi.5 7 We found that there were more women than men among HIV-positive patients, the larger proportion of women (69%) being consistent with other reports.3 15
A new finding in this study is that HIV-positive patients tend to have lesions of a higher malignancy grade than HIV-negative patients (χ2=7.43, p=0.02). The regression model indicated that this was independent of age or the size of lesion.
A number of clinical factors was associated with HIV status. A longer duration between onset and diagnosis in HIV-infected patients was suggested by Kaimbo Wa Kaimbo et al,15 although this predictor dropped out in regression analysis of our data. The association between HIV status and the presence of leukoplakia and feeder vessels is interesting but it is not clear what the underlying pathophysiology might be.
The CD4 cell count was determined in 96% of HIV-positive patients and 85% of patients had a CD4 cell count lower than 200 cells/μl. The median CD4 cell count of the patients at the time of presentation was 71 cells/μl, well into immunosuppression. In the east African setting, with the financial and technological constraints, OSSN might be considered a criteria to commence antiretroviral treatment in HIV-positive patients.
The recurrence rate within 1 year was 16.6%, much higher than the recurrence rate reported by Waddell et al18 in Uganda (3.3%) and it is interesting that recurrence was more frequent among HIV-infected patients. It is difficult to interpret this, however, because the surgical excision procedure in our study was not standardised and only some of the histopathological reports commented on whether the tumour had been totally removed. It suggests that the meticulous removal procedure described by Waddell et al18 is critical in avoiding recurrences, but the question of whether HIV-positive patients are more likely to have recurrences than HIV-negative patients has not yet been answered.
References
Footnotes
Funding Fogarty International Center paid for the cost of the histopathological examinations.
Competing interests None.
Ethics approval Ethics approval was received from the Kilimanjaro Christian Medical College Ethical Committee.
Provenance and peer review Commissioned; externally peer reviewed.
Linked Articles
- At a glance