The persistent view in the literature is that the relative frequency of ameloblastomas is higher in the black population than in Caucasians. The aim of this study was to determine the relative frequency of all odontogenic tumours (OT) in a 100% black population and to compare our findings with those of previous studies. A prospective study was undertaken of all patients presenting with OT to all 16 Nigerian departments of oral and maxillofacial surgery over a 4-year period. The following data were obtained: patient demographics, delay to presentation, extent of the lesion, and histological diagnosis. Six hundred and twenty-two cases were studied. A slight male preponderance was observed (male to female ratio 1.17:1). Patients ranged in age from 5 to 89 years, with a peak incidence in the third decade. The relative frequency of OT was 0.99 per million and that of ameloblastoma was 0.76 per million. Ameloblastoma was the most prevalent OT (76.5%), followed by adenomatoid odontogenic tumours (5.6%), odontogenic myxoma (4.5%), and keratocystic odontogenic tumours (KCOT) (3.1%). The relative frequency of ameloblastoma among Nigerians was not different from frequencies reported previously among Caucasian and Tanzanian black populations. KCOTs were, however, rarely diagnosed in Nigerians as compared to the white population in the Western world.
The notion that the relative frequency of ameloblastomas is higher in the black population as compared to Caucasians was particularly emphasized in a study by Shear and Singh, who found considerable differences in the mixed South African population. Sawyer et al., confirmed this hypothesis by comparing the data from Lagos, Nigeria with those from Virginia, USA. Several retrospective studies from the African continent, in which large numbers of odontogenic tumours were reported, have appeared to support this view, although a possible harvesting phenomenon was mentioned in these studies. A prospective study from Tanzania, however, revealed a relative frequency of 0.68 per million, which is similar to figures reported by several European studies.
Another striking difference between African reports and studies from Caucasian sources is that keratocystic odontogenic tumour (KCOT) seems to be far less prevalent in the black population. In the Caucasian population, KCOT is reportedly diagnosed about 10 times more frequently than ameloblastoma. This assumption is based on the fact that eight ameloblastomas were seen in the same 25-year period within the Arnhem region as compared to 78 KCOTs. Some studies from Africa suggest that KCOTs are at least five to ten times less often diagnosed than ameloblastomas. No satisfactory explanation has been given for either of these suggested differences, assuming that the embryological development of the odontogenic apparatus is the same in both types of people.
In order to challenge or confirm the findings of the aforementioned studies, it was decided to perform a prospective study over a period of at least 4 years in a West African country with a population large enough to collect sufficient data to arrive at a well supported conclusion, enabling comparison with the study of Simon et al. Nigeria, with a population of approximately 150 million and a structured health system, complemented by a sufficient number of trained oral and maxillofacial surgeons, seemed to meet these requirements. Hence, a project was started in 2009 with the goal of prospectively collecting all data from patients with odontogenic tumours from the whole of Nigeria. The aim was to describe the relative frequency of all odontogenic tumours in a 100% black population and to compare these figures with those from previous studies and in particular with studies from Caucasian and Asian sources.
Materials and methods
All existing centres with a department of oral and maxillofacial surgery ( n = 11) were approached in November 2008 after they had already agreed to cooperate in the proposed study. They were asked to keep detailed records of all Nigerian patients who presented with an odontogenic tumour, as defined in the latest World Health Organization (WHO) classification (2005). They also received a structured questionnaire that was designed to gather information including gender and age of the patient, location of the lesion, and the estimated time-span over which the lesion had been present. A schematic drawing of the jaws was also provided to map the tumour diagnosed (see Appendix ).
In order to avoid bias by inclusion of patients from neighbouring countries who were seeking treatment in Nigeria, all patients included in the study had to have their nationality verified; thus, only black Nigerians were included in this study.
Since five new departments of oral and maxillofacial surgery were established during the 4-year study period, these departments were approached in the same manner as described previously and all agreed to cooperate with the project. Hence, at the end of the study period a total of 16 departments involved in the management of odontogenic tumours, covering the whole of Nigeria, were involved. Sixteen examiners, certified oral and maxillofacial surgeons, were involved in the data collection. When questions arose, they were clarified by telephone. The accuracy of diagnosis was ascertained by histological examination carried out by certified pathologists attached to the reporting departments.
The study period ran from 1 January 2009 to the end of December 2012. The definitive diagnosis, including the histological variant of the tumour, was to be reported where applicable, while there was room for comments if these were felt necessary (see Appendix ).
All relevant data were entered into a computer and care was taken to eliminate double counting as a result of patients moving from one centre to another. The data were analyzed using SPSS version 17.0 software (SPSS Inc., Chicago, IL, USA). Simple descriptive statistics and the χ 2 test were employed when appropriate. Statistical significance was inferred at a P -value of ≤0.05.
Since the population of Nigeria grew from approximately 151 to 164 million during the study period, the average for the 4-year period was calculated to be 156.37 million and this figure was employed in all calculations.
Information was obtained for a total of 622 patients from all responding centres. They were 336 males and 286 females (male to female ratio 1.17:1), ranging in age from 5 to 89 years, with a mean age of 32.35 (standard deviation 14.83) years and modal and median ages of 25 and 29 years, respectively. The peak incidence occurred in the third decade of life ( Fig. 1 ); 72.8% of the patients were aged 11–40 years ( Table 1 ). The total number of odontogenic tumours, the histopathological diagnosis, and the distribution by gender are presented in Tables 2 and 3 .
|Age group, years||Males, n (%)||Females, n (%)||Total, n (%)|
|0–10||7 (1.1)||10 (1.6)||17 (2.7)|
|11–20||62 (10.0)||63 (10.1)||125 (20.1)|
|21–30||96 (15.4)||101 (16.2)||197 (31.7)|
|31–40||78 (12.5)||53 (8.5)||131 (21.0)|
|41–50||44 (7.1)||28 (4.5)||72 (11.6)|
|51–60||30 (4.8)||20 (3.2)||50 (8.0)|
|≥61||19 (3.1)||11 (1.8)||30 (4.8)|
|Total||336 (54.0)||286 (46.0)||622 (100.0)|
|P = 0.16|
|Ameloblastoma||Odontogenic myxoma||Adenomatoid odontogenic tumour||Keratocystic odontogenic tumour|
|( n = 476)||( n = 28)||( n = 35)||( n = 19)|
|Patient age, years|
|Mean (SD)||32.89 (14.28)||34.89 (14.16)||21.46 (13.5)||34.47 (17.22)|
|Mode||20 and 23||21||13 and 15||22 and 25|
|Duration of the lesion prior to presentation, years|
|Mean (SD)||3.13 (3.3)||2.91 (2.29)||2.71 (3.33)||3.91 (6.52)|
|Gender and jaw location|
|Gender||Jaw location||Total, n (%)|
|Male, n (%)||Female, n (%)||Mandible, n (%)||Maxilla, n (%)|
|Ameloblastic fibroma||5 (7.8)||10 (15.6)||11 (17.2)||4 (6.3)||15 (23.4)|
|Odontoma||3 (4.7)||5 (7.8)||1 (1.6)||7 (10.9)||8 (12.5)|
|Ameloblastic carcinoma||3 (4.7)||9 (14.1)||9 (14.1)||3 (4.7)||12 (18.8)|
|Odontogenic fibroma||6 (9.4)||2 (3.1)||5 (7.8)||3 (4.7)||8 (12.5)|
|Calcifying epithelial odontogenic tumour||5 (7.8)||3 (4.7)||7 (10.9)||1 (1.6)||8 (12.5)|
|Calcifying cystic odontogenic tumour (Gorlin C)||2 (3.1)||3 (4.7)||3 (4.7)||2 (3.1)||5 (7.8)|
|Cementoblastoma||2 (3.1)||3 (4.7)||2 (3.1)||3 (4.7)||5 (7.8)|
|Squamous odontogenic tumour||2 (3.1)||1 (1.6)||3 (4.7)||0 (0.0)||3 (4.7)|
|Total||28 (43.7)||36 (56.3)||41 (64.1)||23 (35.9)||64 (100.0)|
The relative frequency of odontogenic tumours was 0.99 per million per year, while that of ameloblastoma was 0.76 per million per year. Ameloblastomas constituted 76.5% of the tumours and they were reported in every age group.
The histological types of ameloblastomas are depicted in Fig. 2 . The most common histological variants of ameloblastoma were follicular (41.4%) and plexiform (30.7%). The ‘mixed’ variants (constituting 9.2%) included five variants ( Fig. 3 ). There appeared to be a striking predilection for the mandible for most of the tumours ( Tables 2–4 ), but the difference observed did not attain statistical significance when examined by individual jaw location (anterior, middle, or posterior; Table 4 , p = 0.28).