[The long Telegraph nail for humeral fractures]. / Le clou Telegraph long dans les fractures de l'humérus.

PURPOSE OF THE STUDY: We present an anterograde nailing system for humeral shaft fractures: the long Telegraph(R) nail. This nail enables static or dynamic anteroposterior distal locking and is easy to insert with the special aiming device. MATERIAL AND METHOD: This study concerned 148 patients who underwent surgery between September 1998 and February 2005, 86 women, men age 68 years and 62 men, mean age 53 years. After excluding patients lost to follow-up and pathological fractures, we analyzed 104 files to determine the rate of nonunion as a function of distal locking or not and dynamic versus static locking. We also reviewed 56 patients measuring the Constant score at mean 2.7 years follow-up. RESULTS: There were no cases of iatrogenic radial palsy and no cases of infection. There were four cases of nail fracture. The overall rate of nonunion was 7.7%. Nonunion was never observed with dynamic distal locking (p=0.03). Regarding the effect of the nail on shoulder function, the mean weighted Constant score was 105.4% with preservation of mean joint motion and a pain free shoulder in 36/56 patients. DISCUSSION: The long Telegraph(R) nail enables anterograde nailing of the humeral shaft with a simplified system for dynamic or static distal locking. In this series, dynamic distal locking appeared to provide the best rate of nonunion. If installed correctly, the system does not appear to affect shoulder function.

[Survey of the effects of smoking in an occupational cohort study. Value and limits of a new method]. / Prise en compte des effets du tabagisme dans les études de cohortes professionnelles. Intérêt et limites d'une nouvelle méthode.

BACKGROUND: This paper proposes a method to take smoking into account in occupational cohort studies when a lung cancer excess is observed and no information on smoking is available. METHODS: The model consists in assessing the proportion of smokers in the study cohort on the basis of (i) the relative risk observed for lung cancer (P1), and (ii) relative risks due to other smoking related diseases (P2). If P1 is higher than P2, it is concluded that the observed lung cancer excess is unlikely to be due to smoking. RESULTS: The model was applied to nine epidemiological studies, of which five showed a lung cancer excess (ranged between 1.12 et 1.32). The P2 proportions of four of these five studies ranged between 0.29 and 0.46, and were lower than P1 that ranged between 0.57 and 0.69. These differences reached statistical significance. These results suggest that occupational lung cancer risks might have existed. The model was validated by applying to cohorts without lung cancer excess and to cohorts of smokers and non smokers. The robustness was assessed using different values of the basic parameters: proportion of smokers in the general population and risks due to smoking for smoking related diseases. CONCLUSION: This model is an assessment of the relative importance of smoking and occupational factors in occupational cohort studies, in the absence of information on smoking.