Epidemiological changes in oesophageal cancer at National Hospital, Bloemfontein: 1995, 2000 and 2005

ABSTRACT Background Oesophageal cancer is a common malignancy with a high mortality rate. The two main histological types are squamous cell and adenocarcinoma. An increase in oesophageal adenocarcinoma has been noted, especially in developed countries. Objectives The aim of this retrospective study was to investigate the profile of oesophageal cancer by reviewing medical records of patients diagnosed with oesophageal cancer in 1995, 2000 and 2005. Method The study sample consisted of 474 files of patients diagnosed, for the first time, with oesophageal cancer in 1995, 2000 and 2005, at the National Hospital in Bloemfontein and the outreach clinics in surrounding areas. Information reviewed from patient files included: age, race and gender of the patient, as well as topography, size, histological grade and type of the tumour. Results The number of newly diagnosed cases of oesophageal carcinoma decreased over the 10-year period. The mean age of patients was > 57 years. The majority of cases were Black patients: 90.5% in 1995, 93.2% in 2000 and 87.7% in 2005. More male patients were seen (71.5% in 1995, 70.1% in 2000 and 64.2% in 2005), although the number of female patients diagnosed with this malignancy increased by 7.3% from 1995 to 2005. The mid- and lower third of the oesophagus were affected most commonly, most lesions were 6 cm – 10 cm in length and classified as Grade II, moderately differentiated tumours. Squamous cell carcinoma was diagnosed in 76.9% of patients in 1995, 90.5% in 2000 and 94.3% in 2005. Conclusion The number of newly diagnosed cases of oesophageal carcinoma decreased over the 10-year period, but demographic and disease characteristics remained constant.


InTrODuCTIOn
Two histological types of oesophageal cancer that occur most frequently are squamous cell carcinoma and adenocarcinoma. Squamous cell carcinoma originates from the epithelium lining the oesophagus, while the malignant epithelial cells in adenocarcinoma are arranged in a glandular formation. In the past, squamous cell carcinoma occurred more frequently, but, over the past couple of decades, an increase in adenocarcinoma of the oesophagus has been noted. 1 However, no reports of previous studies explaining this observation could be found in the literature.
Huang 2 classified the natural history of oesophageal cancer into four different phases. The initial phase of oesophageal carcinoma stretches over an extended period, probably 20 years or more. The first changes that occur are small to moderate hyperplasia of the basal epithelial cells of the oesophageal mucosa. This is followed by the developmental phase in which cancerous cells develop. The developmental phase is clinically latent and may be present for a long period of time. These cells penetrate the basal layer of the epithelium and develop into infiltrating cancer that occurs as a granular plaque. Thereafter, the disease progresses to the overt phase, which is the main clinical phase that includes stages two and three of oesophagus carcinoma. In some cases, the primary cancer enlarges from the originally small, localised, early lesion. This lesion penetrates the deep layers of the oesophageal wall during the process of development. The primary cancer can also develop from a superficial lesion and increase rapidly in size, with penetration of the deeper layers of the oesophagus. The final phase of oesophageal carcinoma is usually short, with extensive extra-oesophageal invasion to the vital organs, for example, metastasis to the lungs. 2 Three screening modalities are available for early identification of oesophageal cancer: routine double contrast radiology, fibre-optic endoscopy, and oesophageal cytology. Oesophageal endoscopy employs the brush or balloon method, which has been found to show 95% accuracy in the identification of oesophageal cancer. The combination of balloon cytology and fibre-optic endoscopy provides a solid foundation for the early diagnosis of dysplasia and premalignant lesions. 2 The diagnosis of oesophageal cancer is often made late, when screening is no longer beneficial. In countries with a high incidence of oesophagus carcinoma, such as China, population screening programmes may be employed. In Western countries, however, where the incidence is usually lower, these screening programmes are not always applied. 2 Different factors are associated with oesophageal cancer and various studies across the world have been conducted to investigate these associations. A patient's ethnicity plays an important role, which is illustrated by the drastic difference between the incidence of oesophageal cancer in Black and White ethnic groups. 3 Oesophageal cancer in Black South Africans was very rare until the last couple of decades. 4 In South African Black men, for example, the incidence of oesophageal cancer is 40.9 per 100 000 males, compared to 4.4 per 100 000 males in Mozambique and 1. 5  Several studies were performed to determine whether food types have an effect on the incidence of oesophageal carcinoma. It was found that people consuming insufficient amounts of fruit or cereals, 7 vegetables 2 or excessive amounts of meat, 8 were at greater risk of oesophageal cancer. Chronic malnutrition was also found to increase the risk of oesophageal carcinoma. 3 A case-control study conducted in Sweden between 1995 and 1997, found that first degree family members did not have an increased risk of cancer in both adenocarcinoma and squamous cell oesophageal carcinoma cases. 9 On the contrary, a British study on the incidence of oesophageal carcinoma found that heredity did, in fact, play a role, albeit to a minor extent. 7 The aim of this retrospective study was to investigate the profile of oesophageal cancer by reviewing medical records of patients diagnosed with oesophageal cancer in 1995, 2000 and 2005.

MeTHOD
A retrospective analytical study was performed to review medical records for the years 1995, 2000 and 2005. Four hundred and seventy-four files of patients diagnosed with oesophageal cancer for the first time in these years, at the Department of Oncotherapy at the National Hospital in Bloemfontein and outreach clinics in Bethlehem, Kimberley, Kroonstad and Welkom, were included in the study. Files were made available by the Biostatistics Division of the National Hospital. Information regarding the demography of patients, as well as the topography and morphology of the cancer, were obtained from the patients' files and recorded in a data sheet.
Analysis of data was performed by the Department of Biostatistics, University of the Free State, using SAS software. 10 Results of categorical variables were summarised as frequencies and percentages, and numerical variables as means, standard deviations or percentiles. The three different years (1995, 2000 and 2005) were compared by means of analysis of variance, Fischer's exact or chi-square tests, as appropriate.
The research protocol was approved by the Ethics Committee of the Faculty of Health Sciences, University of the Free State (ETOVS Stud. Nr. 26/07). Permission was also obtained from the Head of Clinical Services of the Universitas Academic Complex, which includes the National Hospital. Data obtained remained confidential, patients' anonimity being preserved through indicating only their oncotherapy reference number on the data form.
A pilot study was performed on 20 patient files from 1998 in order to assess certain aspects of the study, such as the availability of resources (funds and personnel), the amount of time required to complete the data sheet, and the executability of each step of the study. The pilot study also provided an indication of possible results that were to be anticipated.

resulTs
The Results of patient distribution according to race are shown in Figure 1.
The results showed that the parts of the oesophagus most commonly affected by cancer were the middle and lower third, while involvement of the upper third of the oesophagus remained below 10% for the three years investigated.     10.2% and 1.9% of patients did not experience any dysphagia in these three years respectively. These differences between 1995 and the other two years were statistically significant (p = 0.0022).
Based on the Eastern Cooperative Oncology Group (ECOG) performance status of patients with malignant diseases, less than 20% of the 330 patients for whom this information was available were evaluated as Grade 0, that is, they were fully active and able to carry out all pre-disease activities without restriction. 11 Results for ECOG performance status grading of the patients investigated in 1995, 2000 and 2005 are summarised in Table 1.
Information regarding the presence of co-morbid disease was available for 328 (69.2%) patients. The most commonly reported co-morbidity was respiratory disease, which occurred in 28.2% of the patients for whom information was available in 1995, 45.6% in 2000 and 28.7% in 2005 ( Figure 6).

DIsCussIOn
The decrease of 52% in the number of newly diagnosed cases of oesophageal carcinoma, from 221 cases in 1995 to 106 cases in 2005, could be attributed to several possible reasons. A possible explanation could be that, with the emergence of the HIV/ AIDS epidemic in southern Africa, many potential patients died of HIV/AIDS-related diseases before they reached the age at which oesophageal cancer is most commonly diagnosed, which is approximately 60 years of age. 12,13 Most HIV/AIDS-related   With regard to the histological type of oesophageal cancer, squamous cell carcinoma increased steadily. The percentage of cases presenting with adenocarcinoma remained constant Dysphagia is a common symptom in oesophageal cancer and was observed in more than 85% of the study population. Weight loss, which is directly associated with dysphagia, occurred in more than 90% of patients. However, it should be kept in mind that the treatment of malignancies and the disease process could also contribute to physical deterioration.
Due to the retrospective nature of the study, information regarding certain aspects of the disease was lacking in many patient files.

COnClusIOn
The findings of this study showed that, for the years 1995, 2000 and 2005, the majority of patients with oesophageal cancer were Black middle-aged to elderly males who were diagnosed with squamous cell carcinoma. Although an increase in the number of cases presenting with squamous cell carcinoma of the oesophagus was noted, other variables, such as race, age and gender, remained constant for these three years.

ACKnOwleDgeMenTs
We

Percentage of patients
Comorbid disease present Comorbid disease absent