The medicine dissertation is often the last step of a physician's medical studies. Long considered a rite of passage, it is now regarded as a scientific work deserving attention from the national and international scientific community. The objectives of this study were to document the means by which dissertations at the University of Lomé Medical School are diffused and to determine their impact on the scientific production of medical school faculty. This cross-sectional study included all dissertations in medicine at the University of Lomé from 1993 through 2002. We interviewed dissertation supervisors and co-supervisors (questionnaire-structured interviews) and consulted the Medline and CNRS/PASCAL databases, scientific communication registers and local journals to collect information. During this period, 240 dissertations were defended at the University of Lomé medical school. More than half came from the departments of medicine (n=85, 35.4%), surgery (n=57, 23.7%) and paediatrics (n=51, 22.5%). The majority dealt with epidemiological (47 %) and clinical (35.8 %) themes, and only a small minority concerned topics in microbiology and biochemistry (7.5 %). Ninety-nine dissertations (41 %) were published: 54 (22.5%) in indexed journals and 45 (18.5 %) in non-indexed journals; 130 (54%) were delivered as papers or posters at scientific meetings. The distribution of published dissertations according to department showed that the departments of paediatrics (51.3 %), medicine (21.2 %) and surgery (21 %) had the best rates of publication in indexed journals, compared with obstetrics and gynaecology (4 %) and basic science (13.8 %) departments. During the study period, faculty at the University of Lomé medical school published 264 articles in indexed journals: dissertations accounted for only 20.4 %. Dissertations made up a higher proportion of publications in the paediatrics department (32.2 %) than in surgery (20.7 %), medicine (19.8 %), basis sciences (14 %) or obstetrics and gynaecology (3.6 %). Our study shows that the proportion of published dissertations remains small and is even smaller if we consider only those published in indexed journals (22.5 %). Dissertations also account for an insignificant proportion of the publications by medical school faculty. This raises the question of the scientific interest that the faculty accord to these dissertations.