The role of alcohol in oral precancer: observations from a North-East England population.

Alcohol is known to be a risk factor for oral precancerous lesions, but evidence has been weakened by subjective estimates of alcohol intake from patients, and confounded by their use of tobacco. Red cell macrocytosis, assessed by calculation of mean corpuscular volume (MCV), may be a useful objective indicator of chronic alcohol intake. The aim of this study was to compare subjective and objective measures of alcohol intake in patients with oral precancerous lesions and assess the use of reported alcohol intake and MCV on assessing the degree of dysplasia at presentation and their role as markers of the behaviour of such lesions by assessing clinical outcome after treatment. Fifty-four new patients were recruited. All were smokers and had histologically confirmed single dysplastic oral precancerous lesions, but had had no previous treatment. Subjective data about their alcohol consumption were recorded, and blood samples taken for the assessment of MCV. All patients had laser excision of their lesions, which were assessed histopathologically for signs of dysplasia. Patients were followed up for 2 years. The significance of differences was assessed using Fisher's exact test. Alcohol intake of more than 28 units/week and MCV of over 100 were associated with increased dysplasia at presentation (p=0.01 and p=0.03, respectively). Thirty-six patients were disease-free at 2 years, but 18 developed further disease, often at new sites. Alcohol intake of more than 28 units/week was significantly associated with an increased risk of further disease (p=0.03), particularly recurrence at the same site (p=0.02).