Clinicopathologic presentation of malignant orbito-ocular tumors in Kano, Nigeria: A prospective multicenter study.

Background: Malignant orbito-ocular tumors are major causes of morbidity and mortality in the developing countries. Their presentation, morphologic type, and management are challenging, and earlier reports were mostly retrospective surveys. Histological analysis is essential in making correct diagnosis. Here, we determined the clinical presentations and histopathologic types of orbito-ocular tumors in some tertiary hospitals in Kano, Nigeria. Materials and
Methods: This was a prospective cross-sectional multicenter clinical and pathologic study conducted between June 2012 and May 2013. A structured pro forma was used to record findings, patient's biodata, clinical presentation, type of surgical intervention, and histological diagnosis. Data were analyzed with STATA version 11.0.
Results: Sixty-seven eye tumors were observed among 61 patients. There were 43 males and 18 females (male-to-female ratio = 2.4:1). Patients had a mean of 13.4 ± 18.3 with a median of 4.0 (0.5-70) years with a double peak in age distribution. The most common features were visual loss (90.2%), leukocoria (59%), proptosis (55%), and fungating mass (49.2%). The symptoms were for ≥6 months in 57.4% and unilateral in 82.1%. B-scan showed retinal masses and calcifications in 13.5% and 4.5%, respectively. Confirmation by histology was achieved in 91% with the most common tumor being retinoblastoma, conjunctival squamous cell carcinoma (SCC), and rhabdomyosarcoma with frequencies of 38 (56.6%), 11 (16.4%), and 6 (9.0%), respectively. Six patients were HIV positive and all had SCC. Conclusions: Retinoblastoma was the most common malignant orbito-ocular tumor observed in this study. Patients often present late to the hospital with visual loss.

INTRODUCTION

Tumor as defined by Powell White “is a mass of cells, tissues or organs resembling those normally present but arranged atypically. It grows at the expense of the organism, without sub-serving any function.”[1] Histopathology remains the mainstay of confirming diagnosis and determining the malignant potential of tumors.[2]

In the underdeveloped nations, malignant tumors account for most of the orbito-ocular lesions.[34] Studies[567] from some African countries reported an average annual incidence rate of between 0.5 and 1.4 per 100,000 population.

A study in Singapore reported ocular malignancies accounting for 0.15% of all cancers with an average annual incidence rate of 1.89 per million male residents and 1.81 per million female residents.[8] In Canada, incidence rates of malignant tumors have been reported as 6.0 and 8.1 per million among females and males, respectively.[9]

Studies[1011121314] from different parts of Nigeria recorded frequencies of malignancy in ocular tumors from 56.6% to as high as 73.8%. This may be related to genetic or environmental factors.

Some studies[101112131415] showed age range of 1–80 years, whereas a lower mean age was reported in a study[15] in Lagos. Malignant childhood tumors are life threatening and could be associated with visual loss. The most common orbital malignancy in children is rhabdomyosarcoma, while the most common intraocular malignant tumor is retinoblastoma.[16]

A study[17] conducted in Enugu found that about 70% of pediatric patients presented at 5 years and below. Of these tumors, 38% were intraocular, 34% orbital, and 18% and 10% in the eyelid and conjunctiva, respectively.

The most common malignant orbito-ocular tumor in Asian and African adults is squamous cell carcinoma (SCC).[1819]

MATERIALS AND METHODS

The study was adhered to the Tenets of the Helsinki Declaration. Ethical approval was obtained from the Ethics Review Board of Aminu Kano Teaching Hospital. Permission was also obtained from the Medical Directors of Makkah Eye Hospital, ECWA Eye Hospital, and Murtala Muhammad Specialist Hospital. This was a cross-sectional multicenter prospective study that was carried out over a period of 12 months.

Written and informed consent was obtained from all patients or their parents, and assent was obtained from children after detailed explanation (in English and Hausa where necessary) of the nature of the study. Criteria for inclusion were patients of all age groups who had a clinical diagnosis of orbito-ocular tumor that consented to participate in the study. Exclusion criteria included proptosis from other causes, refusal of surgical intervention, and benign histology report.

Patients underwent history taking and routine ophthalmic and systemic examination. After evaluating the patient, the pro forma was filled. The presenting unaided visual acuity (VA) was assessed for distance (at a distance of 6 m) using the Snellen's chart for literate patients and illiterate “E” chart for illiterate patients. Near V/A was measured in adults with the Jaeger chart.

The Snellen's illiterate “E” or the pictorial charts were used for children >3 years at a distance of 6 m in a brightly illuminated room. For children <3 years, V/A was measured by observing fixation and following of light. In patients with fungating tumor, the eye was categorized under nil perception of light.

Anterior segment examination was undertaken with a pen torch and slit-lamp biomicroscope (where applicable). In those presenting with proptosis, extent was assessed with Hertel exophthalmometer. Pupillary reaction was tested and recorded. In cases where the pupil was not visible, it was recorded as “cannot be tested.” Posterior segment was examined by direct and indirect ophthalmoscopy with the binocular indirect ophthalmoscope.

Investigations done included skull X-ray, computed tomography (CT) scan, ocular B-scan, HIV test, full blood count, urea/electrolyte, and creatinine. Patients and guardians were counseled, and consent for surgery was obtained.

The surgical specimen was taken for histology after excision. Further management was based on the histology report warranting management by the oncologist or other specialists as the case may be.

Data analysis

Data obtained were entered into the Microsoft Excel and analyzed with STATA/IC version 11.0 manufactured by Quantec 2009-TEXAS USA. The association between two categorical variables (type of tumor and HIV status) was tested using Fisher's exact test. P ≤ 0.01 for Fisher's exact test was considered statistically significant.

RESULTS

A total of 61 patients were recruited into the study; 67 eyes were examined. There were 43 males and 18 females (male-to-female ratio = 2.4:1). Figure 1 gives the summary of age distribution. Their ages ranged from 6 months to 70 years; the age distribution is skewed. The mean age was 13.4 years with a median of 4.0 years and standard deviation (SD) of 18.3 years.

Figure 1

Age distribution of patients with malignant orbito-ocular tumor

Hausa ethnic group comprised 41 patients (67.2%), while Fulani and others constituted 10 patients (16.4%) each.

Most patients presented with more than one clinical presentation [Table 1].

Table 1

Clinical presentations in 61 patients with malignant orbito-ocular tumors

Clinical presentations	Frequency (number of eyes) (%)
Visual loss	55 (90.2)
Leukocoria	36 (59.0)
Proptosis	34 (55.0)
Fungating mass	30 (49.2)
Conjunctival inflammation	14 (22.9)
Conjunctival mass	7 (11.5)
Eyelid mass	6 (9.8)
Squint	5 (8.2)

The tumor was unilateral in 55 (82.1%) eyes and bilateral in 12 eyes (17.9%), 10 of which had retinoblastoma [Table 2].

Table 2

Laterality of orbito-ocular tumor

Laterality of the tumor	Frequency (%)
Right eye	34 (50.7)
Left eye	21 (31.4)
Bilateral	12 (17.9)
Total	67 (100)

Of the 61 patients enrolled for the study, 47 (77.1%) were seronegative, while 8 (13.1%) declined consent to do the test. All the 6 (9.8%) seropositive patients were adult and had conjunctival SCC. Only a few patients did a CT scan due to financial constraints [Table 3].

Table 3

Investigations in patients with malignant orbito-ocular tumor

Investigations	Frequencies (%)
HIV status
Positive	6 (9.8)
Negative	47 (77.1)
Not tested	8 (13.1)
Ocular B-scan
Retinal masses	9 (13.5)
Calcification	3 (4.5)
Normal	22 (32.8)
Not done	33 (49.2)
Skull X-ray
Soft-tissue shadow	30 (49.2)
Normal	31 (50.8)

Eleven eyes of 11 patients had SCC of the conjunctiva. Of these, 6 patients (54.5%) were HIV positive, and the relationship was statistically significant (P < 0.0001; Fisher's exact test, P < 0.0001).

All the other patients who had other malignant orbito-ocular tumors were HIV negative.

Most of the patients had modified exenteration and enucleation.

Of the 61 (91%) clinically diagnosed cases of malignant orbito-ocular tumors, there was agreement with the histopathology assessment. The most common tumors were retinoblastoma and SCC. Other tumors include fibrosarcoma, astrocytoma, and lacrimal gland tumor. Among the other tumors, fibrosarcoma and lacrimal gland tumor were clinically misdiagnosis [Table 4].

Table 4

Comparison of clinical diagnosis with histological diagnosis in 67 eyes

	Frequency (%)
Clinical diagnosis
Retinoblastoma	38 (56.6)
SCC	11 (16.4)
Rhabdomyosarcoma	6 (9.0)
Burkitt’s lymphoma	3 (4.5)
Basal-cell carcinoma	2 (3.0)
Metastatic tumors	2 (3.0)
Neuroblastoma	2 (3.0)
Other tumors	3 (4.5)
Histological diagnosis
Retinoblastoma	38 (56.6)
SCC	11 (16.4)
Rhabdomyosarcoma	6 (9.0)
BL	3 (4.5)
Basal cell carcinoma	2 (3.0)
Metastatic tumors	0
Neuroblastoma	2 (3.0)
Other tumors	5 (7.5)

SCC=Squamous cell carcinoma, BL=Burkitt’s lymphoma

DISCUSSION

There were more males than females in this study. The gender disparity in presentation of patients with tumors to the hospital could be a reflection of the health-seeking behavior prevalent in the study environment. The males have liberty to seek for their health needs, whereas the females depend on their husbands for finances and also for permission to attend hospital. Similarly, some studies[202122] from Northern Nigeria also had a preponderance of male patients. However, a study[23] from Southeastern Nigeria showed female preponderance. This difference within the same country but in different regions may be due to differences in sociocultural setup, such as religious beliefs, health-seeking behavior, and taboos.

Most of the patients were of Hausa ethnic group which are the major ethnic group in the environmentthe study was conducted.

More than half of the patients with malignant orbito-ocular tumor were children; hence, the data are skewed and account for the low mean age compared to the SD [Figure 1]. This observation was similarly reported in other studies[17202324] from various parts of Nigeria and India.

Two peaks were observed in the age distribution, one in the first decade of life and the other in patients in the third to seventh decades. The early peak is for patients with retinoblastoma and the second for patients with SCC. A study[17] in India also reported two peaks in the age distribution of their patients with orbito-ocular tumors. Retinoblastoma is more common in children from birth to 5 years of age, with the peak age being 3 years. Bilateral retinoblastoma may even present at birth or shortly after birth.[25] This was observed in this study with more than half of the patients with bilateral retinoblastoma presenting before 1 year of age.

The most common clinical presentations are illustrated in Table 1. In a hospital-based study[26] in Thailand on orbito-ocular tumors, fungating mass and proptosis were the most common presentations. A study[27] in Kano on childhood orbito-ocular tumors also reported fungating mass, proptosis, and leukocoria as the most common clinical presentations. In another study[14] in Ilorin, proptosis and chemosis occurred in 84.6% and leukocoria in 61.50%. These studies[1427] indicate the pattern of presentation of the disease in that environment.

An earlier study[27] in Kano reported that the duration of symptoms before seeking medical care was 6 months or more. Late presentation to the hospital could be due to the lack of awareness, misconception of the disease, and poverty. Furthermore, the sociocultural belief and probably poor accessibility to tertiary eye institution could have contributed to the delay in presentation. [Figure 2]

Figure 2

More than 50% of the patients had a duration of ≥ 6 months before presentation to the hospital

Orbito-ocular tumor involved the right eye more than the left with a small number having bilateral manifestation similar to some studies[1828] in Nigeria and Nepal. More than half of the tumors were retinoblastoma and may be sporadic in nature. The bilateral cases of this condition in the study are most likely of genetic origin. Further confirmation of this could be carried out by genetic studies.

Seventy-seven (77%) of the patients were HIV seronegative and 9.8% were seropositive, while 13.1% declined consent to do the test because they are apprehensive of knowing their HIV status.

Some studies[2930] found HIV seropositivity in patients with SCC to be 75%. However, in this study, HIV positivity among patients with conjunctival SCC was 54.6%, and the Fischer's test demonstrates a statistically significant association between HIV positivity and SCC. The difference in incidence may be due to the sociocultural practices, exposure to the risk factors, and also the small sample size.

Among the patients who had ocular B-scan [Table 5], some had retrolental mass, a few had calcification, while other patients had normal findings. Skull X-ray showed that soft-tissue shadow in some patients but no orbital calcification, or bone destruction were seen. Furthermore, a study[27] showed retinal masses and calcification in 88.9% of the patients who were scanned. Hence, a high-resolution B-scan can be helpful in early detection of retinoblastoma, particularly those in the anterior segment.[31] Ocular B-scan appears to be more useful than skull X-ray in detecting intraocular calcification.

Table 5

Types of surgical procedure performed in 67 eyes with malignant orbito-ocular tumors

Type of surgery	Frequency (%)
Excisional biopsy	3 (4.5)
Enucleation	22 (32.8)
Modified exenteration	42 (62.7)
Total	67 (100)

Modified exenteration was the most common surgery performed for advanced tumors comparable with other studies from Northern Nigeria where it was performed in 74% and 73.4% of patients, respectively.[2227]

Retinoblastoma was the most common orbito-ocular malignancy similar to findings reported in other studies[4111920] from the southwest, northeast, and northwest zones of the country. However, a study[17] from the southeast zone had an incidence (38.1%) slightly lower than that from the northwest and southwest zones. This difference in incidence in the same country may be due to genetic and environmental factors and probably culture. Studies[832] from Singapore and Nepal also reported a high incidence of retinoblastoma in their series. The second most common orbito-ocular tumor seen in this study was conjunctival SCC. This finding is similarly reported in other studies[41923] with incidence rates of 11.6%–18.9%, respectively. In Singapore, SCC accounted for 11.2% of malignant orbito-ocular tumors.[8] Increased frequency of this tumor may be due to the HIV pandemic.

The most common orbital lymphoma was Burkitt's lymphoma (BL) similar to what was observed in Northeastern Nigeria[20] and Congo.[33] The reported incidence of BL was 2.6% and 3% of the patients in those series. This study reported frequency much lower than 9.6% in Zaria. The difference between the incidence rates of BL in comparable environment may be due to difference in the study design and time period covered. In studies[1023] from Southwestern and Southeastern Nigeria, BL constituted 21.4% and 9.3% of malignant orbito-ocular tumors, although the identified predisposing factors such as malaria, heavy rainfall, and malnutrition are similar for both the regions.

In most cases, clinical diagnosis accurately correlated with the histological diagnosis. Orbito-ocular tumors can be diagnosed clinically with a high degree of accuracy as shown in this study. This finding may justify initial chemotherapy/chemoreduction in patients with advanced retinoblastoma before surgery is performed where indicated. An earlier study[27] reported histological confirmation in 98% of patients with retinoblastoma.

Histopathology still plays a vital role in confirmation of orbito-ocular tumors. However, with improved diagnostic skill and high index of suspicion, chemoreduction and laser treatment of some ocular and orbital tumors could be given without waiting for histopathological confirmation, especially for patients under the care of an ocular oncology team.

CONCLUSIONS

The most common presenting symptom was visual loss often with late presentation. Retinoblastoma and conjunctival SCC were the most common malignant orbito-ocular tumors.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.