[Hospitalization criteria for erysipelas: prospective study in 145 cases].

INTRODUCTION: Two recent studies conducted in France among general practitioners have shown that they only hospitalized between 20 to 50 percent of patients with erysipelas seen in private practice. We therefore conducted a hospital-based, prospective study designed to determine the hospitalization criteria for erysipelas, since a number of patients are also hospitalized directly through the emergency department.

PATIENTS AND METHODS: This prospective, hospital-based study, included patients hospitalized for clinical diagnosis of erysipelas in 9 centres in north-eastern France. Clinical data were recorded using a standard questionnaire filled-in during the first 72 hours of hospitalization. They included: demographic (name and first name initials, age, sex) and clinical (location of erysipelas) characteristics, as well as the origin of the patient (home, emergency department, other department), the reasons for hospitalization (severity of local or systemic signs, suspicion of deep vein thrombosis, location on the face, age over 60 years, associated diseases, absence of improvement after ambulatory treatment, socioeconomic reasons or on principle hospitalization). Patients hospitalized by (or without) a general practitioner were compared using Chi-2 and Student t tests.

RESULTS: One hundred forty-five adults (80 women and 65 men; mean age 61 +/- 20 years) were included in the study: 89 patients (61 p. 100) were hospitalized by a general practitioner whereas 56 (39 percent) went directly to the emergency department. In 128/145 cases (88 p. 100), erysipelas was localized on the lower limbs. The mean number of reasons for hospitalization per patient was of 2.2 +/- 1.2. The most frequent reason for hospitalization was an associated disease (diabetes, obesity, alcoholism, immunodeficiency), present in 77 cases (53 p. 100). Patients hospitalized through a general practitioner were older than those hospitalized directly through the emergency department (68 vs 51 years; p<0.001). Patients hospitalized by a general practitioner more frequently had an erysipelas located on the lower limbs (94 p. 100 vs 79 p. 100; p<0.001) and the mean number of reasons for hospitalization was greater (2.4 vs 1.7; p<0.001), especially suspicion of deep vein thrombosis and elevated mean age. The treatment started during hospitalization was intravenous penicillin G in 67 cases (46 p. 100), oral pristinamycin in 28 cases (19 p. 100) and intravenous or oral amoxicillin in 9 cases (6 p. 100).

COMMENTS: This study demonstrates the existence of two distinct courses of hospitalization for erysipelas. Patients hospitalized by a general practitioner were older and their erysipelas more frequently located on the lower limbs and deep vein thrombosis was suspected. Our study also shows the emergence of a population of patients younger and without medical supervision, for whom the general practitioner is replaced by the emergency department of the local hospital. Nevertheless, the most frequent reason for hospitalization in both groups is the existence of an associated disease, possibly responsible for further complications.